Wikipedia:Arbitration Committee/Requests for comment/Article creation at scale

Add links
From Wikipedia, the free encyclopedia
Shortcut

Status as of 15:11 (UTC), Sunday, 28 April 2024 (Purge)

Introduction

This is the first of two RfCs about article creation and deletion at scale. Per the rules below, please feel free to add to questions/proposed changes for the first seven days; other suggestions, comments, questions or replies should be made within your own section.

This RfC has been announced at the articles for deletion talk page, the Arbitration Noticeboard, the administrators' noticeboard, the Bot policy talk page, Village pump (policy), Wikipedia talk:Notability, Centralized discussion, Notability (academics), Did you know, List of Wikipedians by article count and the wikiprojects Women in Red, Historic sites, Architecture, UK geography, Wikipedia talk:New pages patrol/Reviewers and Cheshire.

Background

Page-related actions done at scale can overwhelm the community's ability to adequately monitor and participate effectively. The issue is exacerbated in the case of article creation at scale because it escapes the normal notification system.

In the past, Wikipedia did not discourage article creation at scale under the assumption this was the best way to achieve broad coverage of vast subjects such as sports, plant and animal life, geography. There exists a policy that automated or semi-automated creation requires a bot request for approval. More recently, concerns have been raised in multiple venues that the continuing creation of such articles (or article creation at scale performed manually) has overwhelmed editors’ ability to track and assess these articles, and that the churn has become a waste of time and a cause of disruption. In a 2022 August decision, the Arbitration Committee (ArbCom) requested an RfC addressing "how to handle mass nominations at Articles for Deletion" (termed "AfD at scale").

A strong argument was made that the article creation at scale (sometimes known as mass, rapid, or large-scale creation) is one of the causes of dysfunction at AfD with regard to article deletions at scale, and that addressing this issue is a necessary precursor to the ArbCom-ordered RfC addressing AfD at scale.

For a list of proposed solutions other than those initially presented here, please see Archive 2 of WT:ACAS.

Statistics for mass creation

  1. Editors who have created more than seven articles in the past week, including lists and disambiguation pages
  2. Editors who have created more than seven articles in the past week, excluding lists and disambiguation pages
  3. Editors who have created more than ten articles in June
  4. Editors who have created more than ten articles in July
  5. Editors who have created more than ten articles in August
  6. Editors who have created more than 100 articles in the past year
  7. Editors who have created more than 100 articles in the past year, by month
  8. Editors who created more than than 10 articles in 2021, by month
  9. Editors who created more than than 10 articles in 2020, by month
  10. Editors who created more than than 10 articles in 2019, by month
  11. Editors by number of articles created in the past five years

Notes:

  1. None of these contain redirects that were converted into articles by the listed editor, but they do contain redirects that were converted into articles by other editors. I'm looking into fixing the latter; the former can be fixed for smaller datasets, but is too intensive for larger ones.
  2. External links counts can be suggestive about the quality of the article, it can also be meaningless - a low number may be because a large number of offline sources were used, while a high number may be because a template that provides links to a large number of database sources was added.
  1. Articles by editor by day over one year (1138 editor-days exceeded 10 articles; 163 exceeded 25)
  2. Articles by editor by week over one year (922 editor-weeks exceeded 20 articles, 150 exceeded 50)
  3. Articles by editor by month over one year (640 editor-months exceeded 40 articles, 123 exceeded 100)
  4. Articles by editor by year since 2020 (1156 editor-years exceeded 80 articles; 407 exceeded 200)

Note that these do attempt to exclude false positives from editors converting redirects created by the original editor, but some still exist, and this attempt does result in some false negatives. This is also the reason why a hard technical limit will be difficult; we will need some way to identify editors converting redirects into articles, and count those articles towards their count rather than towards the count of the original article creator. (Compiled by BilledMammal)

Purpose of this discussion

This RfC is to find and develop solutions to issues surrounding article creation at scale, partially in preparation for the RfC on article deletions at scale.

Rules

  1. All editors are required to maintain a proper level of decorum. Rudeness, hostility, casting aspersions, and battleground mentality will not be tolerated. Inappropriate conduct will result in a partial block (p-block) from this discussion.
  2. The sole purpose of this RfC is to determine consensus about policy going forward surrounding creation of articles at scale and to form consensus on those solutions. It is not a venue for personal opinion on past creation or creators of such articles or about previous tolerance of such creations, nor about past mass deletions, ditto. Editors posting off-topic may be p-blocked from this discussion.
  3. All comments must be about issues and proposed policy changes surrounding article creation at scale. Comments about any contributor are prohibited and will result in a p-block from this discussion. Any violations will be reverted, removed, or redacted.
  4. Please do not make changes in RfC questions that have already been posted. Anyone is permitted to post additional questions/proposals, below the existing ones. Moderators may at their discretion merge, edit, or condense questions at any point in the process. Any user may suggest such changes.
  5. Please make all additional proposals within seven days of the start of this discussion. Subsequent proposals may be brought up in an editor's own section for consideration and inclusion at the discretion of the moderators.
  6. Discussion is unthreaded. Please create your own comments section within the discussion section for each question, placing your username in the section header. Within your own section you may present your !votes, post questions to other editors, or respond to other editors; unthreaded discussions with other editors can be created on the talk page. Threaded discussion on the RfC will be moved to the talk page by moderators/clerk.
  7. Within a comment section each editor is limited to 300 words, including questions to and replies to other editors. (word count tool) Short quotes from other editors to provide clarity are excluded from the word count, but quoted material may be trimmed by moderators at their discretion. Moderators may at their discretion grant extensions following a request on the talk page that includes a brief explanation of why it is needed; please ping for such requests. Overlength statements will be collapsed until shortened.
  8. If you believe someone has violated these rules, please speak to a moderator on their user talk page. If you believe the moderators are behaving inappropriately, please speak to an arbcom member on their user talk page or by email.
  9. This discussion will be open for 30 days and will be closed by a panel of three editors with experience closing discussions and who will be appointed by the Arbitration Committee prior to the start of the RfC. The closing panel will summarize and evaluate what consensus, if any, exists within the community.
  10. Per their order and this amendment, any appeals of a moderator decision may only be made to the Arbitration Committee at Wikipedia:Arbitration/Requests/Clarification and Amendment. The community retains the ability to amend the outcomes of the RfC through a subsequent community-wide request for comment

Moderators of this discussion

The Arbitration Committee has appointed two moderators for this RfC:

Additional clerking help: MJL (talk · contribs)

Closers

The Arbitration Committee has appointed a panel of three closers for this RfC:

Proposals

The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion. A summary of the conclusions reached follows.
See the closing statement for details.

Question 1: Should we develop a noticeboard where mass creations and sources used for them can be discussed?

Proposed: A noticeboard will be created to allow for obtaining consensus for, making reports of, and having other discussions of mass creations and the sources used for such creations. (Details to be developed there.)

Support (Create noticeboard)

  1. Thryduulf (talk) 19:04, 3 October 2022 (UTC)[reply]
  2. Per what I wrote in the pre-RfC stage, especially the process described here. — Rhododendrites talk \\ 19:38, 3 October 2022 (UTC)[reply]
  3. --Enos733 (talk) 20:13, 3 October 2022 (UTC)[reply]
  4. This is very needed, article creation at scale has highly disruptive potential.Lurking shadow (talk) 20:56, 3 October 2022 (UTC)[reply]
  5. I suspect it's the only proposal that will achieve consensus here, effectively punting all this nonsense to a new location. Nonetheless, there are situations where it will be necessary, and it ought to be a net positive. Vanamonde (Talk) 21:01, 3 October 2022 (UTC)[reply]
    Re-affirming support. Yeah, we have an overabundance of noticeboards, but there isn't a place where someone can ask "hey is mass-creating from this list a good idea" and have a meaningful discussion about it. The closest we have is perhaps WT:N, which isn't very active. Vanamonde (Talk) Vanamonde (Talk) 15:11, 6 October 2022 (UTC)[reply]
  6. Worth a try, though I'm uncertain of how much good it will in practice do. Seraphimblade Talk to me 21:11, 3 October 2022 (UTC)[reply]
  7. Open to giving this a try. HouseBlastertalk 21:18, 3 October 2022 (UTC)[reply]
  8. A good start NW1223<Howl at me•My hunts> 22:52, 3 October 2022 (UTC)[reply]
  9. Per Seraphimblade. Not sure how much it will help, but it can't hurt. ♠PMC♠ (talk) 01:58, 4 October 2022 (UTC)[reply]
  10. More community eyes will lead to better results. Pinguinn 🐧 04:55, 4 October 2022 (UTC)[reply]
    Decent idea. Not sure why we need a mega-RfC to create a new, optional noticeboard, though? – Joe (talk) 10:20, 4 October 2022 (UTC) Moved to oppose. – Joe (talk) 10:11, 5 October 2022 (UTC)[reply]
  11. Echoing Seraphim. Don't have high hopes but its worth a shot. --WhoIs 127.0.0.1 ping/loopback 12:27, 4 October 2022 (UTC)[reply]
  12. A centralized place where consensus can be established and referred to for each mass creation sounds like the proper way to ensure that article quality is met. Each case will inevitably be unique no matter how well concerns about scale, rate or notability are codified. —  HELLKNOWZ ∣ TALK 15:51, 4 October 2022 (UTC)[reply]
  13. * Pppery * it has begun... 18:00, 4 October 2022 (UTC)[reply]
  14. Sounds like a good idea. If mass creations are then questioned the noticeboard discussion can be referenced. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:25, 4 October 2022 (UTC)[reply]
  15. Anything to help tamp down the eternal firehose of article creation. Even if mass article creations are a small percentage of it, that would still help ease pressure on AfC/NPP. —Jéské Couriano v^_^v a little blue Bori 20:52, 4 October 2022 (UTC)[reply]
  16. Beyond My Ken (talk) 03:51, 5 October 2022 (UTC)[reply]
  17. JoelleJay (talk) 05:41, 5 October 2022 (UTC)[reply]
  18. It's worth a shot. –dlthewave ☎ 21:40, 5 October 2022 (UTC)[reply]
  19. It would be low-traffic of course, but I do think that it would be really helpful for, in a centralised location, mass article creation to be beforehand discussed. To have multiple people look over the source(s) used and the notability of the to-be-created articles would resolve most of the problems around mass creation. It could turn into some consensus-less hell what with the dichotomy on the merits of mass creation, but I reckon it's worth a try. J947 † edits 22:24, 5 October 2022 (UTC)[reply]
  20. Not sure how effective it would be but worth a try. Rlendog (talk) 14:59, 6 October 2022 (UTC)[reply]
  21. Support --Jax 0677 (talk) 23:22, 9 October 2022 (UTC)[reply]
  22. While I agree that current noticeboards may be able to handle this, there should at the least be a standardized method for notifying the community about mass creation requests. Reywas92Talk 21:41, 10 October 2022 (UTC)[reply]
  23. While the noticeboard itself may be another venue for "taking the wiki", those editors who actively avoid the place when creating articles at scale can more easily be considered not having the projects best interests at heart. Also, I disdain the suggestion that the creation of noticeboards at scale has any bearing on whether any one noticeboard is of practical use. LessHeard vanU (talk) 15:56, 11 October 2022 (UTC)[reply]
  24. Could be useful but may not always be needed though. Crouch, Swale (talk) 20:22, 11 October 2022 (UTC)[reply]
  25. Support Almost certainly we will need a centralised location for discussion. It will be massively disruptive at the beginning and a board is one way to manage it. scope_creepTalk 16:36, 14 October 2022 (UTC)[reply]

Oppose (Create noticeboard)

  1. Wikipedia already has 39 noticeboards and we should avoid adding more. A noticeboard related to something niche probably won't attract a broad audience and will be of limited usefulness. Hut 8.5 12:03, 4 October 2022 (UTC)[reply]
  2. Let's consider how the last time we tried adding a new noticeboard went. Article creation at scale is a problem, but do we really have distinct events often enough to justify a separate noticeboard? Why can't this be handled at AN or ANI? Trainsandotherthings (talk) 14:02, 4 October 2022 (UTC)[reply]
  3. We already have more noticeboards than we can keep track of. Any discussion about whether or not to embark upon a particular spree of article creation can be conducted at an existing page (e.g., Women in Red) and advertised centrally. Moreover, "article creation at scale" is sufficiently ill-defined and fuzzy around the edges that we'd only be inviting meta-arguments and wiki-lawyering about what belongs on the new noticeboard. XOR'easter (talk) 02:11, 5 October 2022 (UTC)[reply]
  4. The sort of topics which tend to be done at scale have quite different themes -- minor planets; species; athletes; settlements; &c. These all have existing projects which cover these specialist areas and so the particulars of the topic are best discussed there. If there are technical issues then these are addressed by existing discussion forums such as WP:BOTN. Per WP:NOTFORUM and WP:BIKESHED, we really don't need yet another talking shop. Andrew🐉(talk) 09:57, 5 October 2022 (UTC)[reply]
  5. Moved from support per XOR-easter and Andrew D. I do think it's a good idea to discuss mass creation projects with others before embarking on them, but they're absolutely right – it's better to have that discussion with editors who have subject-matter expertise. On reflection I can see how a mass creation noticeboard could easily be dominated by an unrepresentative subset of the community that is opposed to mass article creation in principle, much as Wikipedia:Village pump (WMF) ended up being dominated by people who have an ingrained dislike of the WMF. – Joe (talk) 10:11, 5 October 2022 (UTC)[reply]
  6. It's a good idea to discuss but we already have too many noticeboards. Projects or already-existing noticeboards should be preferred. RSN is a particularly good venue for discussing sources (CF the discussions we have already had there about GNIS, GeoNET Names Server, which are some of the worst sources for mass-creation). FOARP (talk) 12:24, 5 October 2022 (UTC)[reply]
  7. As to my knowledge there are only a few editors who masscreate not so much informative stubs. The issue was more that there has not been presented a viable solution to deal with it. To temporarily prohibit those few to release articles into mainspace and instead allow them to create drafts and submit them for review would be a better solution.Paradise Chronicle (talk) 12:41, 5 October 2022 (UTC)[reply]
  8. I'm not convinced the number of mass-creators is high enough to warrant a separate noticeboard and its incumbent bureaucracy. —David Eppstein (talk) 16:25, 5 October 2022 (UTC)[reply]
  9. I am leaning in opposition to this proposal because, while this question is in scope of this RfC, I think we're putting the cart in front of the horse. A noticeboard might make sense in the case that we have a special deletion procedure that relates to mass-created articles, or if we expect mass-creations to happen so frequently that we need a user conduct noticeboard, but I don't see evidence that ANI is unable to handle user conduct issues that pertain to mass creation. Absent a change in deletion policy or guidelines, the current proposal would essentially create a fork of ANI to deal with conduct issues that pertain to mass creation, and I'm not sure that a board that's going to only be visited by those who have an active interest in mass creation (or opposing mass creation) would be a good idea in terms of its ability to deal with conduct issues, especially since WP:CBAN generally restricts the discussion venues for community sanctions to AN and ANI, which I think is wise given that we want the general community to view proposed sanctions and discuss them before imposing sanctions in the name of the whole community. However, if there are changes in deletion policy, then a discussion to create a noticeboard might be apt at that time. — Red-tailed hawk (nest) 17:25, 5 October 2022 (UTC)[reply]
  10. We should be reducing the number of noticeboards, not increasing them. Article creation at scale can be either good or bad, depending on how it is done, but current procedures andnoticeboards should be able to handle this. Phil Bridger (talk) 18:43, 5 October 2022 (UTC)[reply]
  11. I'm worried that this would create a little-frequented corner where it would be easy for a small number of regulars to dominate, as occurs at some of the more-specialist deletion arenas. Given the breadth of opinion on this topic that seems unwise. I also agree that topic or source-specific expertise is more useful than an all-purpose board. Espresso Addict (talk) 19:30, 5 October 2022 (UTC)[reply]
  12. Please don't; the creation of WP:XRV comes to mind. Which was an idea I enthusiastically supported, and now not even I'm watching it regularly. There's one noticeboard I regularly visit (WP:AN) and one I occasionally stumble upon (WP:ANI), and that's about it. I do post to WP:BLPN when I feel it's needed, but I don't visit BLPN to help with other cases there. That's unfortunate and suboptimal of course, but probably not unusual. The existence of a separate noticeboard for an issue is more likely to be noticed by those who currently need it, rather than those who are currently needed by that noticeboard. ~ ToBeFree (talk) 21:31, 5 October 2022 (UTC)[reply]
  13. Per Hut 8.5, XOR'easter, Joe Roe and David Eppstein. We already have too many notice boards. We do not have the manpower for an additional noticeboard with such a narrow focus. James500 (talk) 00:25, 6 October 2022 (UTC)[reply]
  14. As Hut 8.5 noted, we already have 39 noticeboards with varying levels of traffic, and I think there's a high risk that the only regular contributors to a new noticeboard will be the same editors who have already been arguing back and forth about the mass-creation/SNG debates for the past several months. TheCatalyst31 Reaction•Creation 02:51, 6 October 2022 (UTC)[reply]
  15. As others have noted, we have too many noticeboards already; the last thing we need is another one. —pythoncoder (talk | contribs) 04:03, 6 October 2022 (UTC)[reply]
  16. Others have put this better than I could. We already have many noticeboards and the numbers of "mass creators" isn't that high to warrant a separate noticeboard. A noticeboard with less reports would be on fewer people's watchlist, and would be prone to gaming by a small subsection of the community. —CX Zoom[he/him] (let's talk • {C•X}) 08:27, 6 October 2022 (UTC)[reply]
  17. Per others. "Wenn Du nicht mehr weiter weißt, bilde einen Arbeitskreis" ("If you're ever stuck, just form a committee") is among the more infamous problem resolution strategies the German-speaking world has produced, and this very distinctly feels like the WP version of it. Dr. Duh 🩺 (talk) 08:50, 6 October 2022 (UTC)[reply]
  18. Per Espresso Addict and others above. In addition, I'm concerned about the lack of specifics here: AARV floundered partly because there was so much bickering about how to flesh out the details, and this proposal has even fewer details than that one did. Noticeboards that aren't well thought out ahead of time turn into drama sinks, and I'm afraid that's what would happen here. Extraordinary Writ (talk) 06:26, 7 October 2022 (UTC)[reply]
  19. Existing noticeboards should suffice. wjematherplease leave a message... 12:32, 7 October 2022 (UTC)[reply]
  20. There are existing places these issues can be raised - ANI for example. Those are preferable to having another obscure location that no one can find. I can see the point of this, but don't think it's, right now, necessary. Blue Square Thing (talk) 08:18, 8 October 2022 (UTC)[reply]
  21. Weak oppose - as above, but in particular because of the likelihood that it would be too easily dominated by a small group of editors with a particular agenda. Ingratis (talk) 09:34, 8 October 2022 (UTC)[reply]
  22. Unnecessary as it would rarely be of use and there are other forums available. BoJó | talk UTC 09:40, 8 October 2022 (UTC)[reply]
  23. I would oppose this proposal, unless such a noticeboard also encompasses mass actions in general. NotReallySoroka (talk) 05:22, 11 October 2022 (UTC)[reply]
  24. Noticeboards are magnets for drama and unnecessary fights. We already have bot noticeboards for the kinds of software-assisted creations that there is clear consensus to ask permission first. Otherwise, it's not necessary and will just deter good editors who end up having their every move stalked by people who have a problem with articles about a particular subject. Steven Walling • talk 21:00, 11 October 2022 (UTC)[reply]
  25. I would much rather we have a set of rules around the mass creation of articles. A noticeboard would instead allow a small group of editors to become the gatekeepers of new content. Supertrinko (talk) 02:00, 13 October 2022 (UTC)[reply]
  26. Oppose per Hut. When broader community input is needed WP:VPR is the correct location. BilledMammal (talk) 03:52, 13 October 2022 (UTC)[reply]
  27. Per Trainsandotherthings: Article creation at scale is a problem, but do we really have distinct events often enough to justify a separate noticeboard? feminist (talk) 17:00, 13 October 2022 (UTC)[reply]
  28. WP:VPR is more conducive to peaceful discussion. I strongly oppose taking it to WP:ANI, (supposedly) a noticeboard for urgent incidents and chronic, intractable behavioral problems (emphasis in original). A subpage of some sort could be created indexing mass-creation-related discussions for easy reference, which I feel is the only benefit of a separate noticeboard. Ovinus (talk) 01:00, 15 October 2022 (UTC)[reply]
  29. We already have too many noticebords, no need to add yet another one. Per Joe Roe, this noticeboard is unlikely to be watched by a broad spectrum of editors. It is better take input from a general forum like WP:VPR. ಮಲ್ನಾಡಾಚ್ ಕೊಂಕ್ಣೊ (talk) 06:35, 16 October 2022 (UTC)[reply]
  30. Absolutely not, we have VPR for that.--Ortizesp (talk) 14:40, 20 October 2022 (UTC)[reply]
  31. As other editors have said, we can use VPR for the occasional mass article creation, and its other purposes. Robert McClenon (talk) 05:23, 27 October 2022 (UTC)[reply]
  32. Oppose all aspects of policing article creation like until after WP:BOLD is abolished. There's no reason to presume someone is acting in bad faith merely because they do a lot of work. --Jayron32 18:22, 28 October 2022 (UTC)[reply]
  33. As stated in the introduction, current policy mandates that mass creation of articles is treated as a bot task. We already have a noticeboard for considering bot tasks. WaggersTALK 11:45, 1 November 2022 (UTC)[reply]
  34. It is unnecessary. Mass creation already requires bot approval. BAG will not give approval unless community consensus is first demonstrated. We don't need to regulate where that discussion takes place. The appropriate place will vary depending on circumstance. SpinningSpark 18:22, 1 November 2022 (UTC)[reply]

Comments (Create noticeboard)

Comments from Thryduulf (Q1)
  • I think this will provide clarity to everyone, help keep discussions focused and constructive. This VP:PRO discussion is something we should aim to improve on. Thryduulf (talk) 19:04, 3 October 2022 (UTC)[reply]
  • @Hellknowz: A small number of examples work well for bots and would likely work well here also. Thryduulf (talk) 18:49, 4 October 2022 (UTC)[reply]
Comments from Rhododendrites (Q1)

If you meet the definition of "article creation at scale" (see my comments at Q3), then you must post a notice to this noticeboard with the following information:

  1. The approximate number of articles you will create.
  2. The approximate time frame for creation.
  3. A description of the overall topic/theme.
  4. Which notability criteria you will be using.
  5. What kind of sourcing you will use to demonstrate that each article meets the criteria (subject to the results of Q2).

Upon creation of the noticeboard, a subsequent RfC (or other discussion) will determine how long these discussions stay open, who approves them, if there's an appeals process, etc. — Rhododendrites talk \\ 19:48, 3 October 2022 (UTC)[reply]

Comments from Valereee (Q1)

Trainsandotherthings, FWIW a commenter at the ArbCom case did this analysis and reported that "There does not appear to be any page of ANI archives that don't have at least one thread about AFD." 15:45, 4 October 2022 (UTC)[reply]

Comments from Hellknowz

@Rhododendrites: I feel like it would be much more useful to others if editors were to show/make an example draft/article. If they are serious about the work of creating a whole lot of articles, then surely they can just make one. Presumably, mass creation results in articles that are all basically "the same". So seeing an example would be so much more useful (and much easier to participate in a discussion about it) than a checklist of criteria. You could just look at it from the AfC / AfD perspective. —  HELLKNOWZ ∣ TALK 16:22, 4 October 2022 (UTC)[reply]

Comments from Espresso Addict (Q1)

I'm worried that this would create a little-frequented corner where it would be easy for a small number of regulars to dominate, as occurs at some of the more-specialist deletion arenas. Given the breadth of opinion on this topic that seems unwise. Espresso Addict (talk) 01:45, 5 October 2022 (UTC)[reply]

Question from Scolaire

How will a user who decides to mass-create articles know that such a noticeboard exists? Will a notice be sent to the talk page of every user? Will every Create page have a banner in red saying "If you intend to mass-create you must go to the noticeboard first"? Scolaire (talk) 15:00, 5 October 2022 (UTC)[reply]

Comments from Hut 8.5 (Q1)

Successful noticeboards usually fall into one of two groups:

  1. The noticeboard has a broad scope, or at least a scope which lots of people are interested in, and it attracts lots of people. These noticeboards can be used to establish community consensus for something and allow the community to scrutinise things. For example WP:ANI, WP:VPR, WP:RS/N.
  2. The noticeboard has a narrow scope and is of interest to a small audience of specialists. These noticeboards can be used to make sure these specialists are aware of developments in that area, or to get those specialists to fix a problem. For example WP:BOTN, WP:CP, WT:WPSPAM.

This noticeboard is neither. If the idea is for it to scrutinise mass article creation then it needs to attract lots of people, but it has a narrow scope and will probably only attract a handful of people who are very interested in the topic. It also probably won't get much use. It was mentioned above that WP:ANI has lots of threads about AfD, but most AfDs have nothing to do with mass article creation. Hut 8.5 11:58, 6 October 2022 (UTC)[reply]

Comments from Lurking Shadow

If we don't have a noticeboard for mass-creation, where should these talks take place?

WP:AFD is deletion. I have seen bulk deletion requests, they don't work.

WP:ANI is for conduct issues, not content issues. Mass-creation can be both. ANI would only address the conduct issues. It is ill-suited to solve content issues.

WP:BOTN is specialized on technical problems with bots, and not conduct issues. Not a broad audience.

WP:BRFA would be the right place for approvals, but not if there were unapproved mass-creations. Not focused on conduct either, and not a broad audience.Lurking shadow (talk) 16:22, 6 October 2022 (UTC)[reply]


Comments from ONUnicorn (Q1)
My thoughts largely echo those of Scolaire. How will users (especially new users) learn of the noticeboard? Regarding Lurking Shadow's question of "If we don't have a noticeboard for mass-creation, where should these talks take place?" Perhaps at WP:VPPRO. Users could say something along the lines of, "Hey, over the next few months I intend to create short articles on every corporation established in Delaware between the years of 1790-1850. I will use this database, that book, and those periodicals as sources." And a discussion could be had there. ~ ONUnicorn(Talk|Contribs)problem solving 19:58, 11 October 2022 (UTC)[reply]

Comments from Editor Scope_creep (Q1)

VPR is not the place to discuss. We don't even known the types of problems that will raised, never mind finding ways to fix, update code, discussion to reach consensus, approval and so on. There are so many unknowns. A noticeboard is an ideal way to centralise these things. With AI going to be used to generate all these new articles, where do you record that information, for example, during an approval process. On the talk page. Not a chance. On some other board. Its not specific to that board. Not having a board also means a complete lack of focus, with a lack of participatory action by interested editors. We have seen this in the past with NPP dissapearing in 2021. There is no reason that I can think of that would be net positive, if it not used. Lastly, the number of boards is irrelavant. It is the centralised recording of information, where everybody knows where it is and is easily accessible, that is important here. scope_creepTalk 16:49, 14 October 2022 (UTC)[reply]

Comments from Uanfala

Most of the opposition above stems from concerns around how popular the noticeboard will become (given the scarcity of mass creations and the existence of so many other noticeboards) and whether its participants will be representative of the wider community. However, these issues are only relevant if you see the place as a formal venue with the power to grant approval or withhold permission for mass creations. But these aren't applicable if the venue is an informal one: a place where people can keep track of the bigger ongoing creation projects, exchange ideas about how to do that more efficiently, share advice on when bots may be useful or if community consensus may need to be sought first, get wikiprojects involved if there are content issues, etc. I imagine this could be of use primarily to mass creators themselves and to NPP reviewers, and I don't see why setting up such an informal discussion venue would need community sanction from this RfC. – Uanfala (talk) 10:32, 29 October 2022 (UTC)[reply]

Comments from Editor X

Please open your own section with username in the heading

Question 2: Should we require (a) source(s) that plausibly contribute(s) to WP:GNG?

Proposed: Modify the General notability guideline (GNG)/Subject-specific notability guidelines (SNG) at WP:Notability (as appropriate) to add: (Please rank your choices by listing, in order of preference from most preferred to least preferred; ranking all options you don't consider completely unsuitable will assist closers in determining consensus.)

A: All articles created under SNGs (other than those which confer notability) must be cited to at least one source which would plausibly contribute to GNG: that is, which constitutes significant coverage in an independent reliable secondary source.

A-2: At least two sources.

B: All articles (except those not required to meet GNG) must be cited to at least one source which would plausibly contribute to GNG: that is, which constitutes significant coverage in an independent reliable secondary source.

B-2: At least two sources.

C: All WP:MASSCREATEd articles (except those not required to meet GNG) must be cited to at least one source which would plausibly contribute to GNG: that is, which constitutes significant coverage in an independent reliable secondary source.

C-2: At least 2 sources.

D. No change.

Statements (Require GNG-quality source(s))

Please rank your choices by listing, in order of preference from most preferred to least preferred; ranking all options you don't consider completely unsuitable will assist closers in determining consensus. Sign as usual with 4 tildes.

  1. D only (I don't think change will improve things, especially given the very significant variety in SNGs) --Enos733 (talk) 20:16, 3 October 2022 (UTC)[reply]
  2. C2 > C > A2 > A > D > B2 > B. Much of the conflict that led to this RfC is driven fundamentally by a mismatch between criteria used for creation and for deletion. This is primarily the result of SNGs that do not independently confer notability being used to justify mass-creation using databases and lists. Neither such SNGs nor such sources are, at present, admissible as evidence for keeping at AfD, where such articles inevitably end up. Requiring the articles to include sources supporting GNG addresses this mismatch. I would prefer two sources to one, but requiring it of every single article is a bit of an over-reach. Vanamonde (Talk) 20:53, 3 October 2022 (UTC)[reply]
  3. B2 > B > A2 > A > C2 > C > D. Vanamonde's point about the disconnect between creation and deletion is a good one, and requiring editors to create articles that meet some (very low) quality standards will improve the quality of the encyclopedia while also naturally preventing problematic mass-creations of articles without affecting highly productive editors that also produce high quality articles. BilledMammal (talk) 00:08, 4 October 2022 (UTC)[reply]
  4. D (only). This proposal conflates two fundamentally different concepts: notability and verifiability. We cite sources in articles so that our readers can verify the information in them, not so that editors, and the best sources for verification are not always the sources that show notability. If there is doubt about the notability of a topic, it should be addressed with cleanup tags, talk page discussions, and/or an AfD nomination, not by shoehorning sources into articles where they're not needed and have no value to our readers. Notability is a property of topics, which does not depend on the state of sourcing of the article. I also have significant misgivings that options A and B are valid outcomes of this RfC (see below). – Joe (talk) 10:40, 4 October 2022 (UTC)[reply]
  5. D only. A and B are an attempt to make large changes to notability guidelines by the back door and are frankly not valid outcomes of this RfC. An RfC about large-scale article creation should restrict itself to proposals about large-scale article creation as people who have no interest in large-scale article creation probably won't participate, so support for these proposals here does not indicate there is community consensus for them. A and B will make big changes to numerous SNGs, including some which have always been understood to be independent of the GNG (such as WP:PROF and WP:GEOLAND). The GNG also does not require multiple sources. I could get behind Vanamonde's suggestion below that mass-created articles only should have evidence that the subject meets the GNG or a criterion which is independent of the GNG. Hut 8.5 12:12, 4 October 2022 (UTC) Amend per Vanamonde93's comment below: not opposed to C if it means something similar to the 4A proposal. Hut 8.5 17:57, 4 October 2022 (UTC)[reply]
  6. B2>C2>B>C>A(2). Two sources should be a minimum to show sufficient coverage. Although at the end of the day I'd support any of these changes. --WhoIs 127.0.0.1 ping/loopback 12:31, 4 October 2022 (UTC)[reply]
  7. C-# > D. For C specifically, the number of sources should be sufficient and varied to write an article. This is how all GNG works and I don't see why mass creation should lower this bar. From my experience, AfD sees 3 in-depth sources as barely enough. A and B options sound like they are outside the scope of the RfC and change how all articles are approached, so I cannot see how those are valid options here. —  HELLKNOWZ ∣ TALK 16:01, 4 October 2022 (UTC)[reply]
  8. D > C > C2. I don't think change will improve matters, but C is acceptable. A and B are not. Thryduulf (talk) 16:08, 4 October 2022 (UTC)[reply]
  9. C,B,A,D (don't really care whether the requirement is 1 or 2 sources). This seems to have worked for WP:SPORTSCRIT, so it would be beneficial to expand it to here. I concur that A and B are kind of outside scope, so prefer to C to them. * Pppery * it has begun... 17:58, 4 October 2022 (UTC)[reply]
  10. C,B,A,D I don't see that having one reference that supports GNG is onerous, and should take the some of stress of AfD. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:29, 4 October 2022 (UTC)[reply]
  11. D only - the other options are tempting, but we the community should have no appetite for a repeat of the Pending Changes "trial". That that page even exists should be telling. —Jéské Couriano v^_^v a little blue Bori 20:56, 4 October 2022 (UTC)[reply]
  12. B2 > B > A2 > A > C2 > C > D. If an editor is creating an article that needs to meet GNG, then presumably they've already found sources (otherwise, how would they know that it meets GNG?) so simply including these sources in the article is not a big ask. This would take a lot of pressure off of the creation and deletion processes, both of which currently require other editors to search for sources to prove that it's not notable before challenging. –dlthewave ☎ 23:01, 4 October 2022 (UTC)[reply]
  13. D only. I'm extremely uncomfortable with this question as noted below and particularly per Hut 8.5 above. Espresso Addict (talk) 00:45, 5 October 2022 (UTC)[reply]
  14. D only, with a trout to whoever decided to insist upon ranking "all seven options". (Bikeshedding should not be obligatory.) This entire discussion is predicated upon the idea that the GNG is objective, unambiguous, and easily applicable, when really all it does is transfer the ambiguity to questions about what counts as "significant" coverage. Let's not make things worse than they already are. XOR'easter (talk) 02:17, 5 October 2022 (UTC)[reply]
  15. D only. WP:N notes quite explicitly that It meets either the general notability guideline (GNG)... or the criteria outlined in a subject-specific notability guideline (SNG provided the article is not barred by WP:NOT, so SNGs generally do confer notability except when they explicitly state that they don't (or, like WP:NCORP, explicitly state stricter-than-GNG source requirements per community consensus); WP:DEL-REASON#8 explicitly refers to consensus to delete articles whose subjects fail to meet the relevant notability guideline, rather than simply articles whose subjects fail some notability guideline. We have notability guidelines that do not so much as require significant coverage of the article subject to exist, such as WP:NGEO's guidance legally recognized populated places and WP:COMPOSER#1. Requiring GNG-quality sources is not warranted based on our current notability guidelines; if there exist guidelines that don't require SIGCOV, then requiring evidence of SIGCOV is not a good measure of whether mass creation is appropriate or not. Rather than looking for "GNG-quality sources", it would probably be better to frame this as something along the lines of "contains sourcing in the article that shows that the article subject satisfies at least one notability criterion", but this bumps into the wisdom of WP:NEXIST. An alternative might be to create a speedy deletion tag for all mass-created articles that do not contain a sourced claim of significance, which seems like a natural extension of WP:A7 to handle this sort of situation. But I don't think that GNG is the right framing here when the vast majority of controversial mass creations are under SNGs and this proposal's options basically excludes anything that could possibly claim SNG notability. — Red-tailed hawk (nest) 03:56, 5 October 2022 (UTC)[reply]
  16. B2>B>A2>A>C2>C>D. This would only apply to subjects that already must have two+ GNG sources in existence; requiring those sources to actually be cited in the article from the start will, by definition, not affect whether the subject merits an article. So this would have zero change to notability guidelines aside from any special informal temporal leeway certain topics may currently receive when having to demonstrate notability. JoelleJay (talk) 05:53, 5 October 2022 (UTC)[reply]
  17. C 1 or 2. Vanamonde makes a good point between the mismatch in article creation expectations and article deletion expectations. If not a single reliable source can be found for mass created pages, then there are probably better ways to present the content. A and B are nice enough, but I find the arguments they are somewhat out of the scope of this RfC reasonable. CMD (talk) 08:19, 5 October 2022 (UTC)[reply]
  18. D only This has nothing to do with the issue of scale. WP:GNG is not a policy, explicitly provides for exceptions and so is not mandatory. Andrew🐉(talk) 10:04, 5 October 2022 (UTC)[reply]
  19. All articles should meet GNG>All articles should have at least one instance of SIGCOV>B2>B1>C2>C1>A2>A1. Strongest possible oppose of D. I do not see mass-creation problems in any area where GNG applies, it is entirely a creation of SNGs, particularly GEOLAND and NSPORTS (at least whe NSPORTs was being interpreted as if it were a stand-alone SNG). It is caused by just going through a database which does not provide enough coverage to actually write a meaningful article on and creating articles. I do not see what possible basis there is for GEOLAND being somehow kept out of this issue when it is far and away one of the worst areas for mass-creation problems. Requiring at least one instance of SIGCOV means that in every instance at least some kind of meaningful article can be written. FOARP (talk) 12:31, 5 October 2022 (UTC)[reply]
  20. B2 > B > A2 > A > C2 > C. At least two independent sources are necessary to write a neutral article; an article which doesn't have them is going to fail multiple policies in any case. And assuming they're not just making stuff up whole-cloth, the person who originally creates an article ought to have those sources on-hand in any case; it's logical to ask them to provide them at the start. While we allow unsourced text to exist until it is challenged, an entirely unsourced article is another matter entirely. And trying to apply this only to some formally-defined "article creation at scale" is doomed to fail - the simple fact is that if someone is not mass-creating articles, providing the two sources they used to write the article should be trivial for them. The argument that this would somehow change our existing notability guidelines (which already set this requirement) is nonsensical. --Aquillion (talk) 14:59, 5 October 2022 (UTC)[reply]
  21. C2, C. We should stick to making rules for mass-created articles. Two sources is preferable because the bar should be higher for an experienced user mass-creating new articles than for the ordinary Joe. Scolaire (talk) 15:05, 5 October 2022 (UTC)[reply]
  22. D > C > B > A. Per Scolaire and WP:CREEP, we should stick to making rules for mass-created articles. I don't think the wording of any of A, B, or C is clear enough to both have the desired effect of preventing mass creation from lines in databases, and to avoid damaging the creation of properly-sourced stubs. And in practice, something like this is already the case, with unsourced stubs generally quickly getting draftified by the new page patrollers, so we don't need extra rules saying that something must happen without providing any mechanism for making that thing happen. —David Eppstein (talk) 16:30, 5 October 2022 (UTC)[reply]
  23. C The best choice on this entire page. "A" and "B" are basically the elimination of SNG.....probably a good thing in the long run but not something that you do in a mass creat RFC. North8000 (talk) 19:06, 5 October 2022 (UTC)[reply]
  24. B>B2>A>A2>C>C2>D, although A and B may be out of scope, as noted multiple times above and below. The burden of deleting a page (WP:AFDBEFORE) is considerably higher than the burden of creating one (WP:BLPPROD?). I can create 1000 pages about non-notable subjects citing nothing but interviews, and deleting each of them requires others to check various search engines for material I should have provided in the first place. And if one of the 1000s is actually notable, I have even achieved a goal "positive to the encyclopedia" on behalf of others' free time. AFC/draftification and WP:A7 exist, but they don't entirely solve the problem. Requiring at least one independent reliable source as proposed, for whichever cases this RfC can actually decide about, perhaps even two, sounds fine to me. ~ ToBeFree (talk) 21:58, 5 October 2022 (UTC)[reply]
  25. D only per Joe Roe, Hut 8.5, XOR'easter, Red-tailed hawk and David Eppstein. Proposals A and B are completely outside of the scope of this RfC. C assumes that GNG is objective and unambiguous. C is not compatible with WP:NEXIST. James500 (talk) 00:26, 6 October 2022 (UTC)[reply]
  26. B2 > A2 > C2 > B > A > C > D. All articles, like all mainspace content, should be supported by multiple sources. I'll take as close to that ideal as I can get. Levivich (talk) 01:18, 6 October 2022 (UTC)[reply]
  27. D only A and B aren't related to mass creation in the first place, what counts as "would plausibly contribute to GNG" is too open to interpretation to be useful for a deletion or draftification criterion, and with the general trend away from SNGs that language seems a little odd. (I especially oppose any requirement for two sources; there are DYK-quality articles with only one non-database source.) TheCatalyst31 Reaction•Creation 03:03, 6 October 2022 (UTC)[reply]
  28. D ≫ C > C2: I don't think that change will help here. Notable topics may suffer from scarcity of sources, while non-notable ones may have enough in the form of generic listings on web.(stricken as there is a carve-out for SNGs) Not every article has to be a GA. If the proposals covered one or two subject areas, I would have considered to support. But a blanket rule is too far reaching that will hurt content creation more than improve it. Even more far reaching are the A/B proposals, which I think shouldn't be enacted at any cost. —CX Zoom[he/him] (let's talk • {C•X}) 08:53, 6 October 2022 (UTC)[reply]
  29. D > C (subject to a contingency) >> (A or B) - C only works if we have an adequate definition of "mass creation". Some of the definitions that have been proposed, like 50 over a month or 500 over a year amount to less than 2 a day, which is really not mass creation. Maybe something like 20 a day or 100 a week would qualify, and thus be appropriate to require some additional indication of notability upon creation. A and B might make sense at some point but are really outside the scope of this RfC. C2, B2 and A2 all go far beyond what should be required at the time of creation - there is time to add additional sources if there is a reliable source in the article at creation, and in addition B2 and A2 are way beyond the scope of an RfC on mass creation. Rlendog (talk) 15:06, 6 October 2022 (UTC)[reply]
  30. C, the vast gulf in effort required to mass-create versus mass delete articles means that some form of quality control has to be introduced, and one GNG-approaching source is an acceptable compromise here, the massive amount of wasted editor time over mass-created sports biographies showcases this excellently. Devonian Wombat (talk) 02:16, 7 October 2022 (UTC)[reply]
  31. C# with option C-2 > C. THis is because our problem addressed here is for masscreated articles, and this requirement should mean that the articles should pass GNG. And thus not be a problem for being useless or deletable. Graeme Bartlett (talk) 03:41, 7 October 2022 (UTC)[reply]
  32. B > C > A > B2 > C2 > A2, not D. Requiring at least one instance of significant coverage should be a bare minimum for any article, but is an absolute must in order to prevent disruptive mass-creation. We are well beyond the point of doing nothing, so no change is not an option. wjematherplease leave a message... 12:30, 7 October 2022 (UTC)[reply]
  33. Other: Depends if the article is a BLP. I feel that we need to raise standards incrementally, and the first standard we raise should be the one for biographies of living people. I feel that from now on, nobody should start a new biography of a living person in the mainspace unless it cites at least two GNG-quality sources, but if that's difficult we could make it one GNG-quality source and one other non-database source for the time being.—S Marshall T/C 19:02, 7 October 2022 (UTC)[reply]
  34. D only. This really won't help. WhatamIdoing (talk) 04:20, 8 October 2022 (UTC)[reply]
  35. C>C2>D. The others are out of scope of this RFC. We are tolerating these quality errors from newbies because we don't want to bite them away, but mass-created articles with severe problems like that coming from likely more experienced editors are just not tolerable.Lurking shadow (talk) 06:53, 8 October 2022 (UTC)[reply]
  36. D mainly; C is just about OK only. Of course articles should have reasonable sources. But trying to put in place "rules" which are really quite precise will cause more harm than good and lead to gaming Blue Square Thing (talk) 08:22, 8 October 2022 (UTC)[reply]
  37. C2 > B > A; NOT D. The problem is mass creation so options A and B are effectively academic, though it is worth making the point that any new article must have SIGCOV from at least one reliable non-database source. BoJó | talk UTC 09:37, 8 October 2022 (UTC)[reply]
  38. D Only The other options are inadequate, will leave a great opening for continued creation of poorly sourced stubs in areas covered by SNGs or not subject to GNG. - Donald Albury 14:55, 8 October 2022 (UTC)[reply]
  39. D only. Per several others. BeanieFan11 (talk) 17:52, 10 October 2022 (UTC)[reply]
  40. A2 > A> B2 > B > C2 > C, not D All articles, mass-created or not, should be expected to have a modicum of significant coverage that is not merely databases. The failure to include significant coverage in new articles is what leads to issues at AFD. Failure to include significant coverage in mass-created articles increases the probability of incorrect, low-quality, and non-notable content and moreover results in intractability of these issues as the scale of the articles make it daunting to attempt to address the problem at AFD as people call for procedural keeps. Reywas92Talk 21:47, 10 October 2022 (UTC)[reply]
  41. B > A > D > C - Requiring all articles to cite at least one source would ease many problems around here. Honestly, my ideal wording would be B, if it would leave out the parenthetical. ~ ONUnicorn(Talk|Contribs)problem solving 20:11, 11 October 2022 (UTC)[reply]
  42. D only. As said several times A and B are out of scope anyway, and C presupposes that GNG is clear and unambiguous, whereas much of AfD is actually occupied by squabbles between different subjective interpretations of what is or isn't GNG-worthy. Ingratis (talk) 06:55, 12 October 2022 (UTC)[reply]
  43. B2 > C. B2 would overall best for the encyclopedia, but I think C more directly addresses the mass creation issues which have caused such trouble, so I'll settle with that for the time being. -Indy beetle (talk) 08:32, 13 October 2022 (UTC)[reply]
  44. B2 > C. WP:SPORTCRIT already contains language comparable to B1, and it is my experience that it has slowed the creation of sports biographies that are sourced only to statistical databases (particularly for editors that engaged in "mass" creation via those databases). I think B2 will reduce the flow even further. The quantity of these types of biographies (and the toll it places on our deletion processes) is such that D cannot be the preferred outcome. Jogurney (talk) 19:53, 13 October 2022 (UTC)[reply]
  45. A2>A>B2>B>C2>C>D Badly sourced articles being deletable would solve many problems. The most important articles already exist in an acceptable state(e.g. Earth). The less important articles need sources; if they have reliable sources they can far easier be checked for misinformation. The job of the cleaning squad would become drastically easier. We only have limited human resources already being stretched, let's solve that.Lurking shadow (talk) 17:20, 14 October 2022 (UTC)[reply]
  46. C2 > C > A2 > A > D > B2 > B. scope_creepTalk 18:42, 14 October 2022 (UTC)[reply]
  47. D only. No need to change the existing rules. A and B are unrelated to this RFC. ಮಲ್ನಾಡಾಚ್ ಕೊಂಕ್ಣೊ (talk) 06:50, 16 October 2022 (UTC)[reply]
  48. D only no change needed, this RFC is nonsensical. A and B out of scope, and we are becoming overreliant on GNG to begin with.--Ortizesp (talk) 14:41, 20 October 2022 (UTC)[reply]
  49. B > C > A > B2 > C2 > A2 Requiring a single SIGCOV source is a perfectly reasonable quality standard, and doesn't put any strain whatsoever on competent page creators. There's no reason why article creators should not put at least that much effort into their work. Requiring two is perhaps too much, but still better than nothing. Avilich (talk) 18:31, 20 October 2022 (UTC)[reply]
  50. C2 > C > B2 > B > A2 > A > D As stated above "Badly sourced articles being deletable would solve many problems." WomenArtistUpdates (talk) 00:51, 21 October 2022 (UTC)[reply]
  51. B2 > B > A2 > A > C2 > C > D per Dlthewave, ToBeFree and others. The needs of Wikipedia have changed since 2005 but many of our policies and guidelines have not. Rather than further expansion, we mostly need further curation, and additional articles should meet a basic quality standard to avoid degradation of our reputation and accuracy. I cannot think of an example (excluding our almanac and gazetteer articles) where, in 2022, it would be appropriate to create an article that did not contain two sources that contribute to notability.
    I have never bought the argument that there are topics that are notable because of unreliable/useless sources that can't be included in the article. Even if this was true for a topic, a responsible volunteer would put the sources on the talk page. And we need to shift the burden of labour onto creators a lot more than we do, because of the huge burnout and understaffed ("under-volunteered?") nature of NPP, AFC, AFD and related areas. — Bilorv (talk) 13:09, 21 October 2022 (UTC)[reply]
  52. Weak support for C, but really I think the guidance should go something like this: "Mass created articles must include sufficient sourcing to show notability, and cannot be based only on simple statistical databases. While there are no firm requirements about the level of quality an article must reach when created, many in the community have a strong preference for mass created articles to be more than one- or two-sentence stubs." (This obvious extends to quality, but doesn't mandate anything about length). 2B and 2A are outside of this RfC. (copied from comments section below in case it matters). — Rhododendrites talk \\ 12:23, 24 October 2022 (UTC)[reply]
  53. D; while an argument can certainly be made for a modification of our guidelines, per Joe Roe and others, I feel like a wide-ranging alteration to the general notability guidelines should be made at an RfC about the general notability guidelines, not as one item of a seventeen-part RfC about mass article creation. It's already difficult enough for people to pay attention to the ever-blossoming world of gigantic policy RfCs, but I think it stretches the bounds of credibility to expect people to pay attention to every RfC on the whole site on the off-chance that an RfC about roller coasters has grave implications for rutabagas. jp×g 09:00, 25 October 2022 (UTC)[reply]
  54. C or C2. It is clear that mass creating articles causes problems when the creator has not provided sources. If a user is mass creating articles they should be able to demonstrate that the article is notable as the time needed to clean up if they actually are not is a problem. For individual articles this is not as much of an issue, but for mass creations I think the burden should be on the editor who created the article. Dreamy Jazz talk to me | my contributions 23:20, 27 October 2022 (UTC)[reply]
  55. D only. Notability attaches to the subject, not the article content. So long as WP:CSD A1/A3 standards are exceeded, we're fine. --Jayron32 18:26, 28 October 2022 (UTC)[reply]
  56. D only Wrong forum. At some point we will have such a requirement. But this discussion isn't the right place for it. Also I think the other options are a poor idea that would hurt getting and retaining new editors and so are bad for the encyclopedia. CSD criteria are enough. Hobit (talk) 17:42, 30 October 2022 (UTC)[reply]
  57. D only. I don't see why we need special notability rules for mass creation. SpinningSpark 18:24, 1 November 2022 (UTC)[reply]
  58. Anything other than D we have passed the point of needing to import databases. --Guerillero Parlez Moi 11:00, 4 November 2022 (UTC)[reply]

Comments (Require GNG-quality source(s))

Comments from Thryduulf (Q2)
  • Combine the massively variety in style, format and purpose of SNGs with the very subjective nature of what constitutes a source that "passes" the GNG and changes would not improve matters. Better imo to discuss things individually at the board proposed in Q1. 19:26, 3 October 2022 (UTC)
Comments from Rhododendrites (Q2)

Whoa. 2B (and to a much lesser extent 2A) extends far beyond the scope of this RfC IMO, applying to all articles. This would be a radical change and should be separated if anyone wants to really propose it.

Weak support for C, but really I think the guidance should go something like this: "Mass created articles must include sufficient sourcing to show notability, and cannot be based only on simple statistical databases. While there are no firm requirements about the level of quality an article must reach when created, many in the community have a strong preference for mass created articles to be more than one- or two-sentence stubs." (This obvious extends to quality, but doesn't mandate anything about length). — Rhododendrites talk \\ 19:52, 3 October 2022 (UTC)[reply]

Comments from Vanamonde (Q2)
  • I believe a cleaner way to do this would be to simply prohibit mass-creation that is based on criteria that do not independently grant notability. However, this idea has not made it into the RfC. Some of the proposals above do so indirectly, at least per my view of what mass-creation is, and so have my support. Vanamonde (Talk) 20:55, 3 October 2022 (UTC)[reply]
    @Hut 8.5: If I'm understanding you correctly you're not opposed to the C options, but you don't directly refer to them in your !vote; have I misunderstood? Vanamonde (Talk) 16:47, 4 October 2022 (UTC)[reply]
  • Valereee, didn't receive your ping, but that's what 4A below is trying to achieve. I don't see it reaching consensus though. I have to admit the community's opinions about notability and deletion are even more off-the-wall than I expected. Vanamonde (Talk) 15:53, 5 October 2022 (UTC)[reply]
  • @FOARP: I'm not sure why you struck your reply to me, because as it happens I agree with you; there are problematic creations under GEOLAND, including mass creations. However, that's an issue with GEOLAND that needs addressing; it doesn't mean we should permit mass-creation under criteria that are even looser. Vanamonde (Talk) 17:21, 7 October 2022 (UTC)[reply]
  • @ONUnicorn: I'm surprised you're ranking D over C given your argument, and wondered if it were a typo. In case you think I'm nit-picking, A & B don't have a ton of support, so in a ranked choice consensus determination, your position will default to D. Vanamonde (Talk) 23:48, 12 October 2022 (UTC)[reply]
Comments from Enos733 (Q2)

I agree with the comments of Thryduff and Rhododendrites above (and can support Rhododendrites's proposal). We do want high-quality articles, but we must balance that with the idea that this project is "freely editable" and we should be hesitant to enact procedures that enact barriers towards the sharing of knowledge (we have procedures for dealing with vandalism and deleting articles that do not fit with this project). --Enos733 (talk) 21:11, 3 October 2022 (UTC)[reply]

Comments from Joe Roe (Q2)

Options A through C represent a fundamental change to our core content policies (throwing out WP:NEXIST and modifying every single SNG) and options A and B would apply this to all articles. This is an absurd overreach of the stated scope of this RfC, which was already stretching ArbCom's request for comment on article deletion, and I don't think a local consensus on them here would be enforceable. The moderators should remove or modify it so we don't waste our time. – Joe (talk) 10:49, 4 October 2022 (UTC)[reply]

Comments from Valereee (Q2)

Vanamonde93, if you can figure out a way to propose something around based on criteria that do not independently grant notability that you think can work for a general !vote, please do. In distilling the workshop suggestions, I couldn't. Anything I didn't include was for reasons of not being able to figure out how to distill something that addressed major concerns, but if you have an idea definitely add it!

Vanamonde93, well, duh, I forgot to sign w/ping lol...yeah, there's certainly a lot of pushback on any proposal to make any changes to try to solve the creation-at-scale problems that many feel are a root of the AfD-at-scale problems. From my point of view as just-a-moderator, a creations RfC that cannot find consensus for any proposed solutions is still a successful effort, as it removes concerns about lack of due diligence from the deletions RfC. But it probably won't be pretty at that RfC. Valereee (talk) 16:01, 5 October 2022 (UTC)[reply]
Scolaire, you can add another question at will for several more days. Since it's related to Q2, I'd suggest adding it as a Q2A. Try to word it in a way that doesn't cause a problem with mutual exclusivity. Valereee (talk) 16:14, 5 October 2022 (UTC)[reply]
Comments from Espresso Addict (Q2)

Echoing above comments; some of these options appear a considerable overreach of the "Article creation at scale" RfC remit. They should be removed, before this poisons the whole RfC. Espresso Addict (talk) 00:15, 5 October 2022 (UTC)[reply]

@Dlthewave: I assume because some sources do exist but (1) are not online (eg print books not on Google Books); (2) are in old, offline/paywalled newspapers/magazines/journals; and/or (3) are not in English or any language the editor reads. Eg. the article I'm currently working on, a listed building, is primarily sourced to the listing description, which references two books neither of which is previewed on Google Books, one oop, the other rather expensive; I have found another potential piece of significant coverage but it is in a 1932 journal/magazine which has no online content listing, is all paywalled and may not be online back to 1932. As it is entirely impossible for me to visit an academic library, my personal solution has been to buy the two books at significant personal expense, and I may plan to ask the resources exchange for the journal/magazine. However, I do not think it is unreasonable for someone to start the article based purely on the listing. Espresso Addict (talk) 23:06, 6 October 2022 (UTC)[reply]

@FOARP: WP:GEOLAND is a very diverse guideline that covers a wide range of articles, including heritage-listed structures and substantial geographic entities such as mountains. Personally I find its settlements advice completely unhelpful (afaik, Scotland has no legally recognised places lower than cities?) which has contributed to my stopping working in this area altogether. Perhaps it would be better to try to refine the guidance further rather than require everything falling under GEOLAND to address GNG, which would potentially bring into question a huge number of relatively harmless articles, say on heritage-listed structures, where there's a single easy source but sparse other sources that are online & non-paywalled. Espresso Addict (talk) 01:56, 8 October 2022 (UTC)[reply]

@S Marshall: Agree BLPs are more of a problem than other topics. Espresso Addict (talk) 11:05, 8 October 2022 (UTC)[reply]

@WhatamIdoing: I like the facts approach, but fear it's gameable; consider a list of publications for a scientist. (Is each paper a fact? Or worse, is each paper n facts (co-author 1, co-author 2....)?)

@S Marshall: "a database of species or diseases or asteroids maintained by scientists is a good source, but a database of Olympians maintained by amateurs isn't" the key might be experts vs amateurs. Espresso Addict (talk) 01:00, 9 October 2022 (UTC)[reply]

@JoelleJay: There are plenty of article types that don't need secondary analysis to be useful. Espresso Addict (talk) 01:40, 9 October 2022 (UTC)[reply]

@Dlthewave: Agree good practice would be to list sources that can't be verified somehow; I use often use Further reading for this. Espresso Addict (talk) 03:33, 12 October 2022 (UTC)[reply]

Comments from JoelleJay (Q2)
Responses to Red-tailed hawk

@Red-tailed hawk (and others), I'm not sure I understand your reasoning? The proposed changes would only affect GNG-based articles, which includes around half of the SNGs, e.g. NSPORT -- so there would be no conflict with SNGs that don't ultimately require GNG. Currently the burden of verifying a subject is notable is almost entirely relegated to AfD and NPP, rather than the article creator (where the responsibility should lie), and the proposals aim to address this. 06:17, 5 October 2022 (UTC)

Red-tailed hawk, I discussed GNG-equivalent SNGs here, although I realize now it's not explicitly mentioned in some that GNG is required, they just restate it. Nevertheless, the proposals would impact the SNGs that do rely on GNG and would've prevented Lugnuts-type mass creation of database-derived athlete microstubs. And regarding NCORP, I've given my thoughts on its differences from the GNG here. JoelleJay (talk)
Red-tailed hawk, the proposals wouldn't affect non-GNG SNGs but would have stopped Lugnuts' sports bios: NSPORT has always required GNG for establishing notability (see: first line, third line, FAQs 2 & 5, and the first section). The part that allowed the creation of non-GNG-demonstrating articles was the second sentence, which basically presumed a subject meeting an SSG also met GNG, and therefore only SSG verification was needed to pass AfC/NPP, and a strong rebuttal to this presumption was required at AfD. But the expectation always was that each subject must demonstrate GNG with sourcing "eventually".
Requiring citations to SIGCOV ab initio therefore would not change which subjects were notable, it would just prevent editors from making articles on subjects if they don't know they're notable...which they shouldn't be doing anyway, since a comprehensive NPOV article can't even be written without coverage. 00:44, 6 October 2022 (UTC) EDIT: 21:20, 7 October 2022 (UTC)

I think that it would be more relevant to say that the database record should contain 10+ facts that would belong in the article, if that article reached FA. This would NOT be acceptable for GNG because it still doesn't contain secondary analysis. It doesn't matter how many isolated facts a database contains on a subject, if no person is actually discussing them specifically they are not DUE in an article, let alone a sufficient basis for an article. 01:12, 9 October 2022 (UTC)

Policy: Do not base an entire article on primary sources, and be cautious about basing large passages on them. No articles should be based on primary sources. And if a whole source type is unacceptable as the basis of an article, it is not possible for it to contribute to notability. 04:23, 9 October 2022 (UTC)
@S Marshall Whether a database is scientific/professional or not is irrelevant, the only factor that should be considered is whether someone actually discussed the subject specifically in their own words. That rules out ALL the problematic databases, which only contain facts with no interpretation/commentary, while permitting the ones that actually do have secondary coverage. Being able to write something on a subject (which we can always do by stringing together isolated "facts") is only half of the point of N; the other half is to be able to maintain NPOV/no OR/comply with NOT, which can only occur when other people have personally analyzed those facts and discussed the ones that are important, and why. If we just went by the first purpose we could have articles on every single child with a MaxPreps profile. JoelleJay (talk) 22:10, 9 October 2022 (UTC)[reply]
@S Marshall, the problem with that is there is no indication of which info is encyclopedic, let alone that the topic is encyclopedic merely through being mentioned in a database. That page could never be used to support notability because no one could write an article based solely on it: it would require massive OR for an editor to interpret all of that info, distill what's important, and then contextualize it into a prose WP page. That's what makes it primary. JoelleJay (talk) 23:10, 9 October 2022 (UTC)[reply]
This is the problem with all of the databases that do not contain prose interpretation, which makes it very easy to just exclude off the bat any such database entry from contributing to notability. All one has to do when checking a database entry citation is see whether the subject is discussed specifically, by a human, in prose. JoelleJay (talk) 23:18, 9 October 2022 (UTC)[reply]
@S Marshall, databases that don't contain original prose written specifically about the subject are comparable to specialized web browsers or news aggregators: they may be reliable, but they are not synthesizing anything specifically for any entry, they are merely tools to autocurate and present facts without further analysis. Some of those facts might be from a secondary publication, but they are more likely to be cited to a primary research article or even uploaded directly by a researcher. For the facts that are referenced to secondary sources or have secondary coverage elsewhere, those publications should be the basis of the article rather than the database. For brevity, I'm also going to just quote some parts of NOT and OR (all emphases mine) and expand on those on a talk page: Information should not be included in this encyclopedia solely because it is true or useful... Verifiable and sourced statements should be treated with appropriate weight. And A primary source may be used on Wikipedia only to make straightforward, descriptive statements of facts that can be verified by any educated person with access to the primary source but without further, specialized knowledge. JoelleJay (talk) 19:28, 10 October 2022 (UTC)[reply]
  • @ONUnicorn, the argument for requiring a SIGCOV source for mass-created articles in particular is a) because of how many orders of magnitude more time, editors, and effort dealing with deletion of those articles is compared to creation; b) it necessarily slows down creation based on inclusion in a non-notability-granting database if the creator must actually know and be able to halfway demonstrate that the subject is notable rather than merely presuming this is the case; and c) because it gives the community the ability to assess someone's article sourcing in the midst of their mass creation spree and tell them, with P&G support, to cease generating articles based on X source, thus preempting however many pages are left in their queue. JoelleJay (talk) 00:27, 14 October 2022 (UTC)[reply]
Comments from FOARP (Q2)

Why exempt GEOLAND from this when it is far and away one of the worst areas for mass-creation? Just look at the mess that C46 left us with in Iran and California based on bare database sources. FOARP (talk) 12:57, 5 October 2022 (UTC)[reply]

Comments from Scolaire (Q2)

Perhaps we need a C3 and a C4 that don't specifically invoke GNG, and don't link to WP:MASSCREATE, e.g. "All mass-created articles (except those specifically excluded) must be cited to at least one (or two) source(s) which demonstrate(s) significant coverage in an independent reliable secondary source." Poorly thought out. I don't know how we would actually edit WP:N. Scolaire (talk) 16:23, 5 October 2022 (UTC)[reply]

Comments from Red-tailed hawk (Q2)

For what it's worth, options A and B are clearly outside of the scope of this RfC, as Rule#2 states that the sole purpose of this RfC is to determine consensus about policy going forward surrounding creation of articles at scale and to form consensus on those solutions. Simply put, options A and B are not narrowly tailored towards the mass creation of articles; while they could be something the community puts together as a sweeping change to WP:NEXIST, they are clearly out-of-scope for this RfC as they affect policy in a way that is much, much broader than the narrow focus of dealing with creations of articles at scale. That being said, Option C, as it only focuses on (a subset of) mass-created articles, is within the scope of this RfC, as is the option arguing for no change in policy. — Red-tailed hawk (nest) 16:38, 5 October 2022 (UTC)[reply]

@JoelleJay: Would you please provide a list of the SNG pages you say require GNG to be met? I'm looking through them and I'm not getting anywhere near 50-50. — Red-tailed hawk (nest) 22:30, 5 October 2022 (UTC)[reply]
@JoelleJay: With due respect the part where SNGs are just effectively restating GNG is somewhat nebulous, far too much so for the reach of this proposal. This proposal squarely asks about articles that are required to meet GNG, not meet something else that we think is similar to GNG but might have a hair of difference. It actually wouldn't have even stopped Lugnuts, since Lugnuts's mass creations were all in line with an SNG that, at the time, did not require GNG levels of coverage nor any evidence of a source that would cover it significantly. If we'd like to change notability guidelines, which I take it the statement RE NSPORT is getting at, there's a place for that, but this is not that place. — Red-tailed hawk (nest) 00:04, 6 October 2022 (UTC)[reply]
Comments from Dlthewave (Q2)

CX Zoom, I'm having trouble imagining a topic that's notable yet suffers from a lack of sourcing. Notability is based on significant coverage, so generally if sourcing doesn't exist then it's not notable. And if the sourcing does exist, why not simply include it in the article from the get-go? The exception would be SNGs that presume notability, but this proposal has a carve-out for those. –dlthewave ☎ 16:02, 6 October 2022 (UTC)[reply]

@Espresso Addict: and others, I wonder if a good middle ground would be to include sources in a bibliography if they can't be cited directly for whatever reason. This would at least demonstrate the notability of the topic. –dlthewave ☎ 03:21, 12 October 2022 (UTC)[reply]

Comments from WhatamIdoing (Q2)

None of these proposals will solve any of the actual problems. The problem isn't that the editors aren't providing sources; the problem is that some editors don't believe that the provided sources "plausibly contribute to WP:GNG", and there is no rule we can make that has the power to stop people from feeling that way.

Part of the problem is our unwillingness to settle at WP:N exactly what WP:SIGCOV means. I once had an editor tell me that it was SIGCOV if the whole fact of "<Person> got married today in <city>" could be read in a single source, and didn't have to be cobbled together from one source saying "<Person> got married today" and another saying "<Person> is in <city> today". I've also had editors say that SIGCOV is 100 consecutive words of prose in a single source, 300 words or prose, etc. My own interpretation relies on WP:WHYN: We have "significant coverage" when we have enough coverage to write a decent article. But several editors who like this "plausibly contribute to WP:GNG" phrase don't think that's relevant at all; they appear to think that SIGCOV is a matter of how much serious-souding prose you find, because when I show them that it's possible to write a decent bigger-than-stub article from a database, they persist in believing that databases are incapable of providing SIGCOV. And if you are inclined to believe that because some entries in some databases are insufficient, then all entries in all databases should be assumed insufficient, then I point out that https://omim.org/entry/609423 – a single database record – contains more consecutive sentences of prose than at least 90% of our articles.

The bottom line is that this proposal will not solve any actual problems. I believe these problems can be solved, but this won't contribute to it. A clear definition of SIGCOV would go much further than any of these efforts. WhatamIdoing (talk) 04:19, 8 October 2022 (UTC)[reply]

@S Marshall, I agree with you about the value of being cautious wrt BLPs. I thought you might want to consider the work of Wikipedia:WikiProject Medicine/Dermatology task force 10+ years ago. It basically was a single editor who set himself a target like "I will turn every red link on the List of skin conditions blue" – and he did. He created hundreds of single-sentence, single-source stubs. All of them were notable subjects. (In practice, all recognized-by-mainstream-medicine human diseases are notable under GNG, because even if the cited source only contains a passing mention, others will go into depth.) Most editors have enough sense to know that certain categories should be assumed notable (e.g., heads of state, prescription drugs, human diseases, the latest iPhone...), but for those subjects that are less obvious to some reviewers, how could we explain to someone that this list/this source is okay, but that list/those sources aren't?
Also, do the problematic aspects of mass creation of articles depend on whether the subject's notability is obvious? WhatamIdoing (talk) 22:56, 8 October 2022 (UTC)[reply]
@S Marshall, I think that it would be more relevant to say that the database record should contain 10+ facts that would belong in the article, if that article reached FA. That would rule out a simple "List of Olympians from Ruritania" (number of facts: name+year+sport+country+score/ranking = 5) but would permit database entries that are actually useful for writing a whole article. (We could swipe the wording for such a provision right out of WP:ELNO#EL1.) WhatamIdoing (talk) 00:10, 9 October 2022 (UTC)[reply]
@Espresso Addict, so long as editors keep using "gameable" as a synonym for "editors trying to follow the rules", then everything is gameable. Perhaps we should embrace that, and write the rules so take advantage of it. "We can't just require articles to have five top-quality sources, because then (*gasp*) people might actually do that! We can't have people going around and creating articles with multiple top-quality sources! What will become of Wikipedia!"
In terms of your example ["consider a list of publications for a scientist. (Is each paper a fact? Or worse, is each paper n facts (co-author 1, co-author 2....)"], would each one of those papers be included in a Featured Article? Any paper that wouldn't be mentioned is irrelevant.
If the paper would be mentioned, then I think most editors would interpret 'the fact' as being "He wrote a paper". I don't think they would stand for someone writing "1. He wrote a paper. 2. The co-author was Francis Crick. 3. The title was "Molecular Structure of Nucleic Acids: A Structure for Deoxyribose Nucleic Acid". 4. It was published in 1953. 5. It was published in Nature. 6. It was in volume 171. 7. It was two pages long." WhatamIdoing (talk) 01:24, 9 October 2022 (UTC)[reply]
@JoelleJay, yes, that's in the policy, but it was added by SlimVirgin, who argued for years that when journalists stood in front of a burning building and submitted a story based on what they saw with their own eyes, this was a "secondary source" on the grounds that the journalists were neither arsonists nor firefighters. More than a decade later, editors now (mostly) agree that Wikipedia:Secondary does not mean independent and that WP:PRIMARYNEWS exists, but SV represented the other end of that spectrum.
Consequently, I'm not sure that it fully represents either the actual or the desired practice as manifested by the work of tens of thousands of editors each month. Strict enforcement of that rule would, for example, result in deleting a large fraction of articles about music albums, which are sourced either entirely to primary sources (e.g., the track listing) or to sources that are partly primary and partly secondary (music reviews). We have, and seem to want, such articles, and when they appear at AFD, it is rare for anyone to complain about NOR violations because the article only names primary sources. It may be one of our rules that does not mean quite what it says. WhatamIdoing (talk) 21:12, 9 October 2022 (UTC)[reply]
S Marshall

Espresso Addict, I'm thinking about mass creation in terms of articles about sportspeople and athletes, because that's the actual cause of the drama here. Mass creation often arises when an editor sets themselves a target like "I will turn every red link on this list blue". I think it matters what the list is: asteroids or roads or species of beetle are one thing, and biographies of living people are another. I do not have a problem with Whatamidoing's database source but I do have a problem with the database sources used by Lugnuts among others, and I think it's partly about who compiles the database and partly about its checkability and rigour.

I feel that we can't agree on what mass creation is and therefore we need to generate solutions that don't require a definition of mass creation. Hence my wording above.—S Marshall T/C 10:15, 8 October 2022 (UTC)[reply]

@WhatamIdoing, I'm starting to wonder if it might be useful to distinguish scientific databases from non-scientific ones? Imagine a rule where a database of species or diseases or asteroids maintained by scientists is a good source, but a database of Olympians maintained by amateurs isn't. This idea might need some refinements because for example it makes our articles about roads and railway stations look vulnerable, and there's scope to wonder whether we feel that's a bad thing.—S Marshall T/C 23:39, 8 October 2022 (UTC)[reply]
@JoelleJay, you and I often agree but I join issue with you on this particular point. I'm very persuaded by WhatamIdoing's remark about scientific databases. Take for example this database entry, which is incredibly condensed and contains zero prose sentences. It's also verifiable and well-referenced. It describes the star HIP 56948, and would unpack to a complete article. WhatamIdoing gives similar examples of databases about diseases and I'm sure there are equivalents for species—and, I should expect, court cases as well, although these last can always be supported by prose sources containing legal analysis because judgments are published, at least in Western democracies. I feel it would be an error to rule these out.—S Marshall T/C 22:37, 9 October 2022 (UTC)[reply]
@JoelleJay, again I disagree. WhatamIdoing has already shown here how an editor who understands the science can unpack a scientific database entry into a full article. The only issue is that you do have to understand the science to do it, but that's settled policy already. Verifiability means verifiable by someone. It doesn't mean verifiable by anyone.
So for example if I cite a source in a foreign language, I've cited the source. Randy from Boise might not be able to follow my citation but that's his problem. The burden is not on me to drive to his house and give him language lessons. If he wants to check my translation, the burden is on him to find an independent Wikipedian who speaks the language and ask them. (And not use Google Translate because it's unreliable.)
In the case of scientific databases, you've got to, er, speak science to unpack them and not everyone does. That's OK. Randy from Boise's ignorance is, and rightly should be, his problem and not an article writer's problem.—S Marshall T/C 23:36, 9 October 2022 (UTC)[reply]
Comments from Donald Albury (Q2)

After thinking long and hard, I have decided that I do not support any of the alternatives under this question. As worded, the question variants exempt all articles that are subject to an SNG, or are not subject to the GNG. That leaves categories that have been particularly subject to mass creation of articles, such as populated places, outside the scope of this "solution". - Donald Albury 14:41, 8 October 2022 (UTC) Edited 14:42, 8 October 2022 (UTC)[reply]

Comments from Redfiona99

From my memory, the major problem was with people mass-creating from databases. Would there by anyway of requiring that, if an article was created from a database, the creator would also have to add at least one other, non-database, source? Red Fiona (talk) 17:14, 8 October 2022 (UTC)[reply]

Comments from XOR'easter (Q2)

Redfiona99, I think I see where you're coming from, but that just leads us back to the problem that not all databases are created equal. A poorly-phrased rule could lead to unwarranted bias against perfectly good sources just because of how their websites happen to be formatted. XOR'easter (talk) 17:19, 8 October 2022 (UTC)[reply]

Comments from ONUnicorn (Q2)

@Vanamonde93: - Ranking D over C was not a typo. There are several reasons why I prefer D (the status quo) to C.

  • Current policies and guidelines create an environment where the information in articles has to be verifiable - that is, sources have to exist from which readers can verify the information contained in the article; but, critically, except for some articles about living subjects, there is no current requirement that sources actually be cited. In recent years there has been an increasing practice (largely stemming from the efforts of NPP) of requiring sources, and shunting unsourced new articles to draft, but there is still no requirement in policy for sources to be present in the article.
  • Options B and A introduce a requirement for articles to cite at least one source. I do not like the parenthetical notation from B that exempts "those not required to meet GNG", and I like A's wording about SNGs even less. I could write an entire essay about the way the SNGs and GNG relate to one another and to the broader concept of notability - but I won't here. Suffice to say that I think the parenthetical in B will cause confusion, and the language of A will REALLY cause confusion. I would prefer a straight requirement for all articles created after X date (The date the change in policy goes into effect) to actually cite at least one source.
  • I understand the objections of people opposing A and B that A and B are beyond the scope of this RFC, which is, ostensibly about "article creation at scale", which people are interpreting to be mass creations. People are not necessarily objecting to a requirement that articles cite at least one source (though it is likely some people would object to such a requirement), they are objecting to that being a result of this RFC as they feel it is beyond the scope of this RFC.
  • C applies a requirement to cite at least one source to all mass created articles.
    • "Article Creation At Scale" is currently undefined. There are some proposals below (question 3) to define it, but those don't seem to be going anywhere.
    • Assuming we are currently using "Article Creation At Scale" and "WP:MASSCREATE" interchangeably, I do not think it is right to apply a higher standard for sourcing to articles created under some undefined or poorly defined concept than we do to every other article in this encyclopedia.
    • If we are not using them interchangeably, then the requirement would only apply to bot-created articles under the bot policy - which I don't think is what supporters are really going for.
    • I have not seen anyone give a good reason why (other than all articles being outside the scope of this RFC) we should require mass created (however we are defining it) articles to cite at least one source when we do not have that requirement for any other class of articles except BLPs.
  • Given all this, C seems egregiously unfair. If people feel B and A are beyond the scope of this RFC, then maintaining the status quo seems like a better option to me than C.
  • If people would like to see a requirement that all new articles cite at least 1 source that plausibly meets the minimum requirements of reliability and independance from the article subject to remain in main space (which is honestly what I would like to see moving forward), or something similar; but they don't think that is within the scope of this RFC, then perhaps we need a separate RFC for that. ~ ONUnicorn(Talk|Contribs)problem solving 15:33, 13 October 2022 (UTC)[reply]
Comments from Ovinus (Q2)

@XOR'easter: What are your thoughts on having source-specific discussions when an editor would like to create a bunch of articles sourced to that database? Ovinus (talk) 00:55, 15 October 2022 (UTC)[reply]

Comments from Pbsouthwood (Q2)

Is changing the basic requirements for all article creation within the scope of this RfC? · · · Peter Southwood (talk): 19:49, 1 November 2022 (UTC)[reply]

Comments from Editor X

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 2A: Should we require the inclusion of (a) source(s) that plausibly contribute(s) to WP:GNG?

Proposed: Modify the General notability guideline (GNG)/Subject-specific notability guidelines (SNG) at WP:Notability (as appropriate) to add: (Please rank your choices by listing, in order of preference from most preferred to least preferred.)

A: All articles created under SNGs (other than those which confer notability) must include at least one source which would plausibly contribute to GNG: that is, which constitutes significant coverage in an independent reliable secondary source. To satisfy this requirement, source(s) may be cited in the body of the article or included in a bibliography or "further reading" section.

A-2: At least two sources.

B: All articles (except those not required to meet GNG) must include at least one source which would plausibly contribute to GNG: that is, which constitutes significant coverage in an independent reliable secondary source. To satisfy this requirement, source(s) may be cited in the body of the article or included in a bibliography or "further reading" section.

B-2: At least two sources.

C: All WP:MASSCREATEd articles (except those not required to meet GNG) must include at least one source which would plausibly contribute to GNG: that is, which constitutes significant coverage in an independent reliable secondary source. To satisfy this requirement, source(s) may be cited in the body of the article or included in a bibliography or "further reading" section.

C-2: At least 2 sources.

D. No change.

Statements (Require inclusion of GNG-quality source(s))

Please rank your choices by listing, in order of preference from most preferred to least preferred. Sign as usual with 4 tildes.

  • D only. Notability attaches to the subject, not the article content. So long as WP:CSD A1/A3 standards are exceeded, we're fine.
  • D - there's already enough in the GNG to deal with this. Blue Square Thing (talk) 09:01, 30 October 2022 (UTC)[reply]
  • D only. It is irrational to create different notability rules for mass-created articles. SpinningSpark 18:28, 1 November 2022 (UTC)[reply]

Comments (Require inclusion of GNG-quality source(s))

Comments from Dlthewave (Q2A)

This proposal replaces "must be cited to at least one source" with "must include at least one source", which may be satisfied with either a citation or further reading/bibliography section. This would require demonstration of GNG coverage (unless exempted by an SNG), while allowing article creation if sources have been found but cannot be accessed or cited for whatever reason. –dlthewave ☎ 16:28, 12 October 2022 (UTC)[reply]

Comments from Editor X

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 3: Should we create a definition of "article creation at scale"? By rate, source, similarity, other?

Proposed: "Article creation at scale" is the creation of over 25 similar/similarly-structured articles per day or 50 per week or 100 per month or 200 per year using the same source.

This definition, once finalized, would be usable for establishing limits for the need to request consensus to create at scale, for requesting permission to create at scale, or for other discussions surrounding article creation at scale. (This proposal is intended to be refined and may not be finalizable here in this RfC but can be used for input for later proposals.)

Support (Create definition)

    • Thryduulf (talk) 19:06, 3 October 2022 (UTC) Reconsidered in light of others' comments. Thryduulf (talk) 07:40, 4 October 2022 (UTC)[reply]
    • Supporting on principle, but it's odd to ask if we should have a definition while simultaneously making decisions about the undefined thing. — Rhododendrites talk \\ 19:45, 3 October 2022 (UTC)[reply]
    • Support. How about the definition would be the top 10 article creators? Or 20?Paradise Chronicle (talk) 23:06, 29 October 2022 (UTC)[reply]

Oppose (Create definition)

  1. I supported, but was confused by the difference between the heading, which asks "By rate, source, similarity, other?" and the actual proposal, which answers that question: "by rate, similarity, and source". I moved to oppose mainly because "similar/similarly-structured" needs more clarity and because "using the same source" should be one way in which they can be similar/similarly-structured. See my comments below for an alternative. — Rhododendrites talk \\ 19:58, 3 October 2022 (UTC)[reply]
  2. Support creating a definition, oppose this definition. As I see it, mass-creation is simply creating articles without individually checking them for notability. Period. It's sometimes justifiable; it sometimes draws from lists or databases whose entries are inevitably notable; and sometimes it doesn't, but the products may still be good. However, putting numbers on it misses the crux of the matter, and also allows for endless dispute about timing and rates (see how bad the wikilawyering can get just with respect to 1RR restrictions). Vanamonde (Talk) 21:00, 3 October 2022 (UTC)[reply]
  3. I'm not sure I support a direct numeric value as anything but guidance; hard numbers can be gamed. The idea here is "If you intend to create a whole bunch of very similar articles, get community feedback rather than just plowing ahead with it." That's not unnecessary at 49 in a week but suddenly essential at 50. Seraphimblade Talk to me 21:22, 3 October 2022 (UTC)[reply]
  4. Per Jacobellis v. Ohio, no need to get boxed in to definition that can be gamed. nableezy - 04:23, 4 October 2022 (UTC)[reply]
  5. No need to get bogged down in particulars. Mass creation can be mass creation even when done slowly. Pinguinn 🐧 04:56, 4 October 2022 (UTC)[reply]
  6. A definition for guidance is a good idea, but not rigid numbers. Thryduulf (talk) 07:40, 4 October 2022 (UTC)[reply]
  7. It's really more about vibe then about specific numbers. We should avoid at all cost the wikilawyering that will come with putting a specific number on it. Create a definition but avoid putting a box around it too strictly. --WhoIs 127.0.0.1 ping/loopback 12:33, 4 October 2022 (UTC)[reply]
  8. I support creating a general description of what "mass creation" would be. But not a rigid definition as such, because that would only lead to lawyering over details. More like a list of indicators of mass creation, like same process, high number, high rate, same source, same structure, etc. —  HELLKNOWZ ∣ TALK 16:10, 4 October 2022 (UTC)[reply]
  9. As with other editor's concerns this sounds like something that would end up being wikilawyered, whether claiming that creations do or don't meet the standard. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:33, 4 October 2022 (UTC)[reply]
  10. Pointless. "We'll know it when we see it". Might as well try to create a numerical definition of what WP:SIGCOV is, or what a stub is, or any of the other concepts which people have differing ideas around - the existence of those differing idea is not and should never be a reason not to do anything about the issues they address. FOARP (talk) 12:33, 5 October 2022 (UTC)[reply]
  11. I've seen enough "I have made only 3 reverts, so I didn't edit war". ~ ToBeFree (talk) 22:02, 5 October 2022 (UTC)[reply]
  12. Impractical. Not the sort of thing you can define as each case needs to be considered individually for impact, need, etc. BoJó | talk UTC 09:44, 8 October 2022 (UTC)[reply]
  13. Better to focus on quality rather than quantity of new articles. - Donald Albury 14:45, 8 October 2022 (UTC)[reply]
  14. The definition, if any is eventually adopted, will likely come from experience using the noticeboard. LessHeard vanU (talk) 16:08, 11 October 2022 (UTC)[reply]
  15. As mentioned previously, no. Articles should be looked at own merit unless an issue is raised elsewhere.--Ortizesp (talk) 14:43, 20 October 2022 (UTC)[reply]
  16. I oppose any and all attempts to stop people from working hard to make Wikipedia better. Unless and until WP:BOLD is repealed, we need to leave people alone, and stop having to justify the work they are doing. --Jayron32 18:28, 28 October 2022 (UTC)[reply]
  17. I'm not averse to having some guidance on this, but rigid rules are open to abuse. If you set 30 per month, creating 29 will suddenly be allowed by implication. 200 per year is way too low. That's less than one a day and is hardly overwhelming the community. SpinningSpark 18:33, 1 November 2022 (UTC)[reply]

Comments (Create definition)

Comments from Thryduulf (Q3)
  • The definition will probably have to be slightly fuzzy, but I can only see this being a good thing. 19:08, 3 October 2022 (UTC) Definitions need to be fuzzy because the real world is messy, but this appears to be being treated as rigid which is not. Thryduulf (talk) 07:40, 4 October 2022 (UTC)[reply]
Comments from Rhododendrites (Q3)

[Turned comment into 3A per suggestion on the talk page]. See my comments at Q1 for what I think a request for permission should look like. — Rhododendrites talk \\ 20:56, 3 October 2022 (UTC)[reply]

  • @Vanamonde93: mass-creation is simply creating articles without individually checking them for notability. Period - I suspect if that were the working definition, there would be consensus to completely prohibit it, but I've not seen anybody put forth a definition like that before. On the flip side, it would also allow for creating thousands of articles per year as long as you know they're notable, regardless of rate/sourcing, and those are things people are clearly concerned about. — Rhododendrites talk \\ 21:14, 3 October 2022 (UTC)[reply]
  • @Vanamonde93: politicians that unambiguously meet NPOL, legally recognized towns, etc. - I agree that we should allow some forms of mass create at scale, but you just defined mass creation as limited to creation without checking the articles for notability. Creating articles that are unambiguously notable means they've been checked for notability. :) I'm not making a point about what should be allowed -- just that that's not a helpful definition. — Rhododendrites talk \\ 21:21, 3 October 2022 (UTC)[reply]
Comments from Vanamonde
  • @Rhododendrites: I don't think there will. There's plenty of mass-creation that's quite justifiable; described scientific species, politicians that unambiguously meet NPOL, legally recognized towns, etc. These are areas in which we've had community support not only for mass-creation, but for bot creation. I think the community is upset about mass-creation of non-notable pages, which, I believe, is the consequence of the GNG-SNG mismatch I mentioned above. Vanamonde (Talk) 21:18, 3 October 2022 (UTC)[reply]
    @Rhododendrites: When I say "checked for notability", I mean topics have individually been evaluated against a criterion that requires manual evaluation. You can quibble with the word "checked" if you'd like, but there's a qualitative difference between creating pages off of a list of MPs (or towns, or cricket players) and looking to see if each meet GNG. That's the fundamental feature of mass creation. Vanamonde (Talk) 04:09, 4 October 2022 (UTC)[reply]
Comments from Editor xaosflux
Above only mentions a lower bound, however at some upper bound such an endeavour should fall to bot-flagged accounts to prevent flooding. Perhaps guidance related to when a bot task should be used is wise here. Such a mass-creation would still need to pass all other requirements, in addition to being pre-approved as uncontroversial. — xaosflux Talk 14:43, 4 October 2022 (UTC)[reply]
Comments from FOARP (Q3)

Have to be honest, I am not at all jazzed by the supporters here !voting oppose on other proposals just because this proposal (and the other attempted definitions) are not being supported. Wikipedia has many contested but nonetheless valid concepts. Take WP:SIGCOV for example - why should we develop a numerical measure of what this is? Would anyone ever agree as to what it is? Yet it is still a very valid and useful concept even without a specific numerical definition. Please address the other proposals on their own merits. FOARP (talk) 12:26, 11 October 2022 (UTC)[reply]

Comments from Paradise Chronicle

Mass creating articles is not per se a bad thing. It can also be a good thing. And we have many definitions for all sorts of things on wikipedia. And the top of the list are the ones who create the most articles.Paradise Chronicle (talk) 23:16, 29 October 2022 (UTC)[reply]

Comments from Pbsouthwood

If we want to have any chance of coming up with a way of handling or regulating article creation at scale we need to know what it is we are discussing. So far, it is fairly clear that we do not all agree on what it means, so there is little hope of consensus on how to handle it. I do not think there is a simple definition, and I have doubts that this approach is likely to have useful results. The problem is complex and I do not think it has been adequately analysed, so we have a lot of proposed solutions which are not likely to solve anything. · · · Peter Southwood (talk): 19:38, 1 November 2022 (UTC)[reply]

Comments from WhatamIdoing (definition)

User:Spinningspark, I hope you don't mind too much if I pick on you for a moment. You wrote that rigid rules are open to abuse. If you set 30 per month, creating 29 will suddenly be allowed by implication. You're far from the only person to express this sentiment, but I like the clean way you phrased it.

Let's say you're absolutely correct about both points. But:

  1. Our current rules are also open to abuse. In fact, the current rules are already being abused, to cry "mass creation" at someone who was creating one or two articles per day on 100% guaranteed-to-be-notable subjects. Do you mean that you prefer the abuse to run in the direction of stopping article creation (current rules) or do you believe there is a proposal that is clear enough that editors who believe WP:ITSCRUFT can't claim that 500 articles per year is "mass creation", but also vague enough that people can't be sure that nobody will invoke mass creation if they create 500 articles per year? I don't believe that's possible, but I would love to be proven wrong.
  2. What's actually wrong with someone creating 29 articles in a month, if we tell people not to make as many as 30? IMO we have a legitimate, practical need to keep a lid on large-scale article creation: We don't want the reviewers' workload to go out of control overnight; we do want enough time (mostly days) to figure out whether the articles are basically okay (e.g., on a notable subject) before huge numbers of them exist. This suggests a sort of speed limit. But: If we collectively decide that the maximum speed limit should be x articles a month is the biggest number that meets our practical needs, then what's actually wrong with someone creating x articles a month? Do you expect drivers to get speeding tickets for driving right under the maximum speed limit? ("But, officer, I'm sure the only reason he's driving exactly 29 miles per hour is because he's gaming the speed limit! You're supposed to drive a variable and arbitrary speed that's always at least 10% lower than the posted speed but usually at least half that speed, so nobody can claim you have 'a pattern of contributing to the encyclopedia'!")

WhatamIdoing (talk) 20:53, 1 November 2022 (UTC)[reply]

@Spinningspark, I really appreciate your answer. It causes me to think these things:
  • whether someone is behaving in a bot-like manner – We control bots because of the potential for making a very big mess faster than a human could detect and stop the bot. Under what circumstances do you think BAG would declare a human who writes one article, or even a few articles, per day to be editing in a "bot-like" manner? WP:MEATBOT suggests that the only criteria are "high-speed or large-scale" edits, and writing one article a day – adding 0.000015% to Wikipedia's existing articles each day; creating less than 0.2% of the new articles written each year – is neither high-speed nor large-scale. Even 10 articles per day (a rate that basically nobody sustains) seems unlikely to be considered "high-speed or large-scale". I don't disagree with you, but I think that this rule would not address the concerns of editors who are disgusted by the slow and steady creation of single-sentence substubs.
  • That simply cannot be done from databases and listings because such sources are not in-depth – What kind of database do you have in mind? You're obviously not thinking about database entries like https://omim.org/entry/609423, whose prose content alone would qualify as a B-class article. You're probably not thinking of the US Census, whose database we have used in three different decades to create and update many thousands of longer-than-stub articles. You're probably also not thinking of https://terrasindigenas.org.br/en/terras-indigenas/3891, a database entry that includes contains not only a substantial number of facts, but also a list of 500(!) news articles about the place the database entry is about. I wonder if your thoughts would be more accurately represented by the statement "That simply cannot be done from [...] sources [that] are not in-depth". Some databases provide enormous amounts of relevant, factual, encyclopedic information. Others don't. IMO we shouldn't treat shallow databases, or databases with basically unencyclopedic content (e.g., a database that replicated the contents of Twitter feeds from Jack Sweeney) the same way that we treat the in-depth database entries with relevant and encyclopedic information.
WhatamIdoing (talk) 01:25, 2 November 2022 (UTC)[reply]
Comments from Spinningspark (definition)

Reply moved from #Comments from WhatamIdoing (definition) –MJL ‐Talk‐☖ 21:58, 1 November 2022 (UTC)[reply]

My opinion is that mass creation is potentially problematic only when it is bot-like. The BAG group are best placed to determine whether someone is behaving in a bot-like manner. Editors who look problematic should be pointed at bot approvals. If they refuse to go through that process then it's a problem for administrators to deal with. Producing well-referenced articles is not bot-like, even when a large number are produced in a year. That simply cannot be done from databases and listings because such sources are not in-depth almost by definition. That's not to say in-depth sources can't be found and creating the stubs would be a bad thing, but that's why BAG ask to see a community discussion backing the bot proposal before allowing it to run, often with a small test run to check it is not problematic. If the basic concept of a mass creation is sound and has jumped through all the hoops, then I don't care if creation outruns our ability to review. Wikipedia as a whole outrun reviewing capability almost as soon as it was created. Frankly, allowing that to happen was the key to Wikipedia's outstanding success as well as the source of some of its problems. SpinningSpark 21:47, 1 November 2022 (UTC)[reply]

@WhatamIdoing: There is nothing intrinsically wrong with creating one-line stubs. It's only problematic if such articles are, or are likely to be, incapable of any further expansion, or are based on unsuitable sources. If you don't like that, then put forward a proposal to ban the creation of stubs; don't mix it up with the issue of mass creation. Creating at the slow rate of one or two a day doesn't strike me as a big issue. If they are problematic that gives reviewers time to address it and take action before too many are created. If they're not, then, well, it's not a problem. SpinningSpark 08:27, 2 November 2022 (UTC)[reply]

Comments from Espresso Addict (definition)

Spinningspark: "The BAG group are best placed to determine whether someone is behaving in a bot-like manner." For the benefit of the closers, strongly disagree. Espresso Addict (talk) 10:21, 2 November 2022 (UTC)[reply]

Comments from Editor X (definition)

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 3A: Alternative three-part definition

An editor is engaged in "article creation at scale" if these three criteria are all met:

  1. Rate - More than [X] new articles in the span of a month or [Y] in the span of a year (with X and Y to be determined subsequently, if this proposal finds support).
  2. Related articles - The articles are on a similar topic, similar theme, or based on the same set of sources.
  3. Manually created - Rather than the use of a bot/script/tool (which requires going through a different process, bot authorization).

Anyone who answers "yes" to all three of these is engaging in "article creation at scale" of the sort that would require abiding by the rules set forth elsewhere in this RfC (such as posting a request to a noticeboard, if Q1 gets support). Even if an editor does not think they meet the criteria, an uninvolved administrator may determine that someone's editing fits within the spirit of these requirements, and instruct them to seek permission.

Support 3A (alternative three-part definition)

  1. As proposer. — Rhododendrites talk \\ 20:56, 3 October 2022 (UTC)[reply]
  2. I think we can change the numbers later, but this is a good start at defining mass-creation. --Enos733 (talk) 21:16, 3 October 2022 (UTC)[reply]
  3. I don't think a generic "create definition" will suffice. We need to decide on a concrete definition at this stage. Scolaire (talk) 15:25, 5 October 2022 (UTC)[reply]
  4. Support this kind of criteria, but the third requirement is not necessary, as it is still article creation at scale if a script is used, just that extra controls are needed. Graeme Bartlett (talk) 03:58, 7 October 2022 (UTC)[reply]
  5. Generally support, since if we are going to have certain procedures related to article creation at scale we need a definition. I agree with Graeme Bartlett that criterion #3 is not needed and so should be eliminated. I also would remove "same sources" from #2, since it doesn't really matter and if there are multiple sources we are getting to the point where the mass creation is probably not a problem anyway. Of course, #1 needs to be defined appropriately. The proposal of 50 per month or 500 per year is really nowhere near appropriate, as we should be able to handle 2 creations a day from any editor. Something more like 25 per day, 100 per week would be more like a mass creation that would be potentially problematic (although I would be fine increasing those numbers). Rlendog (talk) 17:54, 7 October 2022 (UTC)[reply]
  6. I support having a definition of mass-creation/article creation at scale, and I support that definition being based on all three of the criteria (rather than just some). The exact numbers (which have been removed?) do not matter to me, though I would like someone to be able to create more than one article per day without being hassled over it. Editors have done 30-day challenges to create articles, and if they like it enough to keep going, why shouldn't they? WhatamIdoing (talk) 04:27, 8 October 2022 (UTC)[reply]
  7. Support, but not this definition. The Rubicon has already been crossed, here. How it will pan out I don't know. We certainly need some kind of definition, in the manner of the processes we use at the moment but I don't know what it will contain. The bot approval problem is a process we can adapt. The disruption such a editor could do at scale would be enormous, so we need some kind of defintion for selection into the approval process with policy defining that standard. The whole thing needs needs to be hightly controlled at the beginning until everybody knows whats going on. If these creations are limited to 25 per day (which seem entirely arbitary and without foundation), its possible we either drive away new editors who would be expecting to create far more, as they have that capabilty, or we might miss the benefits of the AI revolution entirely, because we don't have vision to realise it. That opinion may change, re:numbers, when we see the physical benefits. 25 a day in 9000 odd a year. That is a number of articles that is easily checked and is within our capability. That figure may be a ideal number to start. scope_creepTalk 18:43, 15 October 2022 (UTC)[reply]

Oppose 3A (alternative three-part definition)

  1. Support some guidelines in principle, but 500 a year before some community oversight kicks in is way too many. Seraphimblade Talk to me 21:19, 3 October 2022 (UTC)[reply]
  2. The method of creation is not relevant, not convinced with these numbers. Thryduulf (talk) 23:24, 3 October 2022 (UTC)[reply]
  3. I'm fundamentally opposed to a numerical definition, see comments below. Vanamonde (Talk) 04:11, 4 October 2022 (UTC)[reply]
  4. Moral support as a better direction than a rigid definition, but any numbers will inevitably lead to gaming the system and arguing over semantics rather than following the intention. Kind of like everyone is always arguing about bot-like editing. Also I should not that mass creation and assisted creation are not mutually-exclusive. Both, either or neither processes could apply to article creation. Bot approval would still ask for mass creation approval should it become a more codified requirement. Finally, one point I do agree with is that an uninvolved admin (or "mass creation clerk" or some such) should have the "authority" to label any multiple article creation as mass creation to avoid arguments about what is or isn't mass creation on individual level. —  HELLKNOWZ ∣ TALK 16:30, 4 October 2022 (UTC)[reply]
  5. I oppose criteria 1. A definition of mass creation should be based on the amount of time spent writing the article. It should not be based the time between edits or the total number of articles. For example, a single article written in one minute is far more likely to be a problem than 500 articles each of which was written in 7.3 hours by an editor working ten hours a day for 365 days [i.e. (365 days x10 hours per day)÷500 articles=7.3 hours per article]. James500 (talk) 00:28, 6 October 2022 (UTC)[reply]
  6. I can't support without having an idea of X and Y. I don't object to the idea, but if we leave this open then at some point X might get set to 3 and Y to 10 - and that's clearly nonsense (I've seen a suggestion that's not a million miles from those numbers). If a sensible lower bound were put on X and Y I might be able to support. Blue Square Thing (talk) 08:26, 8 October 2022 (UTC)[reply]
  7. As with Q3 above, impractical. BoJó | talk UTC 09:46, 8 October 2022 (UTC)[reply]
  8. Again, this is focused on quantity, I think quality is the key. - Donald Albury 14:47, 8 October 2022 (UTC)[reply]
  9. Same as above: "We'll know it when we see it". Just like with WP:SIGCOV and other similar concepts that lack a numerical definition but are nonetheless useful. FOARP (talk) 12:29, 11 October 2022 (UTC)[reply]
  10. The worse offenders will game any given quantity or timescale (or sock around them). As commented in the prior definition discussion, examples will likely be self incrimminating to the casual observer. LessHeard vanU (talk) 16:17, 11 October 2022 (UTC)[reply]
  11. Oppose, as articles can be on a similar topic without being mass created. In addition, the rate of creation doesn't matter; five hundred mass-created articles created over a week are as difficult to deal with as five hundred mass-created articles created over a year. BilledMammal (talk) 05:07, 16 October 2022 (UTC)[reply]
  12. Any numerical criteria is easily gameable. It is enough to use plain old common sense to determine whether someone is engaged in mass creation. ಮಲ್ನಾಡಾಚ್ ಕೊಂಕ್ಣೊ (talk) 07:06, 16 October 2022 (UTC)[reply]
  13. Oppose any and all attempts to stop people from doing the hard work necessary to make Wikipedia grow. WP:BOLD is still a core policy. --Jayron32 18:29, 28 October 2022 (UTC)[reply]

Comments (alternative three-part definition)

Comments from Rhododendrites (alternative three-part definition)

@Enos733 and Seraphimblade: To move this forward, I've replaced 50 and 500 with "X" and "Y". The idea is to agree to some basic rules, and then deal with specific numbers afterward, rather than altogether. Does that make sense? Hope I'm not being too bold... — Rhododendrites talk \\ 21:27, 3 October 2022 (UTC)[reply]

@Thryduulf: ^^ The numbers had already been removed by the time you commented. I'm curious what you mean by "method of creation is not relevant". We're making rules for a specific activity. Up to now, a lot of people have filed all "mass creation" under WP:MASSCREATE and WP:MEATBOT, which are specifically rules about when one should seek bot authorization. The method is important, because use of tools/scripts is often frowned upon while manual creation is typically treated differently. — Rhododendrites talk \\ 00:44, 4 October 2022 (UTC)[reply]

@Seraphimblade: The Even if an editor does not think they meet the criteria, an uninvolved administrator may determine that someone's editing fits within the spirit of these requirements, and instruct them to seek permission. part tries to do that. Is there a better way? — Rhododendrites talk \\ 01:56, 4 October 2022 (UTC)[reply]

I find the arguments that setting clear numbers (and allowing admins to go by the spirit of the numbers to avoid gaming the system) will encourage wikilawyering/gaming more than some subjective "I know it when I see it" meaning... bizarre. Like why did we implement 3RR? Because "don't edit war" means people wikilawyer over the meaning, blocks/unblocks get litigated, etc. Do people still wikilawyer over 3RR? Of course, but having a bright line sets expectations clearly, without removing the ability of admins to act on other forms of edit warring. — Rhododendrites talk \\ 16:54, 4 October 2022 (UTC)[reply]

Comments from Seraphimblade (Q3A)

Even without a particular number currently in place, I still think this is open to gaming. I would probably see a definition more along the lines of edit warring, where we effectively say "If you violate 3RR, you are pretty much certainly edit warring, but that doesn't mean you're not edit warring if you don't make more than 3 reverts. Now, here's what an edit war tends to look like." I think that would be a much more effective way of defining "mass creation" and much less susceptible to gaming. Seraphimblade Talk to me 01:46, 4 October 2022 (UTC)[reply]

Comments from Thryduulf (Q3A)

@Rhododendrites: Method of creation is not relevant because the only thing that matters to readers is the output. Tools, scripts, bots, manual can all produce output of good or bad quality. Thryduulf (talk) 07:34, 4 October 2022 (UTC)[reply]

Comments from Espresso Addict (Q3A)

I'd like to draw attention to WP:Women in Red's long-running #1day1woman initiative, which encourages daily article creation on what's arguably a single topic. Generally the outcomes have been favourable and have certainly resulted in positive press coverage. Espresso Addict (talk) 23:44, 6 October 2022 (UTC)[reply]

Question 3B: Should we create a definition of "article creation at scale"?

Proposed: Create a definition of "Article creation at scale".

This definition, once finalized, would be usable for establishing limits for the need to request consensus to create at scale, for requesting permission to create at scale, or for other discussions surrounding article creation at scale. (Details may not be finalizable here in this RfC but can be used for input for later proposals.)

Support (Q3B)

  1. If this us just "should we have a definition", that seems obvious to me. Or, well, it did seem obvious until I saw a couple people on this page argue against it. What in the world is the noticeboard people are supporting going to be for if there's no clear definition of when someone needs to use it and/or what considerations requests will be judged by? Many people have proposed things like CSD or special rules for deleting mass created pages -- how can we support or propose anything like that without starting from a clear definition? — Rhododendrites talk \\ 13:52, 4 October 2022 (UTC)[reply]
  2. I believe this would be useful in principle. I'm fairly certain it's not going to get anywhere in practice, given how wide the disparities are in the understanding of the problems we face. Vanamonde (Talk) 16:14, 4 October 2022 (UTC)[reply]
  3. We need some definition, but the placement of hard limits or certain methods will lead to wikilawyering. I think something that gives a basic definition, but only as far as the general idea, would be a better idea. We don't need a strict definition to recognise harassment, and we don't need one to recognise articles being created en masse. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:39, 4 October 2022 (UTC)[reply]
  4. We absolutely must agree on some sort of a definition. Without a definition, we're all talking past each other when it comes to mass creation. Espresso Addict (talk) 00:47, 5 October 2022 (UTC)[reply]
  5. Per ActivelyDisinterested. The definition should be set of characteristics that mass actions generally have and non-mass actions generally don't that act as guidance not rigid rules. Thryduulf (talk) 09:37, 5 October 2022 (UTC)[reply]
  6. Sure? The devil's in the details on these things, though. — Red-tailed hawk (nest) 22:12, 5 October 2022 (UTC)[reply]
  7. Support. I guess this is what the community needs so the parties don't have to explain it over and over again in discussions on editors who mass create articles.Paradise Chronicle (talk) 23:06, 5 October 2022 (UTC)[reply]
  8. Support, no real reason not to considering how much the topic is referenced. Devonian Wombat (talk) 02:17, 7 October 2022 (UTC)[reply]
  9. Yes, but make it fairly generic, and don't confuse the definition with the problems that may be there. The scale is just the rate or number, not the quality. Graeme Bartlett (talk) 03:59, 7 October 2022 (UTC)[reply]
  10. Support. We need a definition if we are going to do something about it. I think it needs to involve numbers, since only numbers can define whether the creation is "mass"; once we know we have a mass creation under the definition we can address whether the quality is adequate in any particular case. The numbers don't have to be rigid - if we say. for example, 25 per day and someone creates 24 per day for several days we can address that as mass creation, or even make the definition a bit fuzzy, like about X per [period]. Rlendog (talk) 17:58, 7 October 2022 (UTC)[reply]
  11. Support. This is the bare minimum we have to do to be able to take any action regarding article creation at scale at all. The definition might be somewhat fuzzy, but we need to have something to reference. --Aquillion (talk) 18:17, 7 October 2022 (UTC)[reply]
  12. Support. If we use numbers, then we can have a discretionary range - a minimum(e.g. more than 10 articles per week) and a maximum(e.g everyone doing more than 30 articles per week is engaging in mass-creation).Lurking shadow (talk) 19:40, 7 October 2022 (UTC)[reply]
  13. Yes, we need a definition. That definition needs to be clear enough that people who cry "mass creation" because someone created a handful of articles on subjects they dislike can be told to knock it off. We do not need another process in which there is one set of rules for me and another set of rules for thee. WhatamIdoing (talk) 04:29, 8 October 2022 (UTC)[reply]
  14. Yes, per my opposition to 3A above Blue Square Thing (talk) 08:27, 8 October 2022 (UTC)[reply]
  15. Though I still oppose numeric values being used as anything but very loose guidance (if at all), I think a clearer definition of what "mass creation" generally looks like would be helpful. Seraphimblade Talk to me 12:02, 9 October 2022 (UTC)[reply]
  16. As a statement more that a hard definition; "careless creation of quantities of low quality articles from poor sources are considered detrimental to the community cause" - if anyone can think of words beginning with c to replace article, poor sources, and detrimental, can substitute as they wish. LessHeard vanU (talk) 16:24, 11 October 2022 (UTC)[reply]
  17. Support - you can't make rules about something if you don't define your terms. ~ ONUnicorn(Talk|Contribs)problem solving 20:37, 11 October 2022 (UTC)[reply]
  18. People here are disagreeing by an order of magnitude over the threshold (and other criteria) of "mass creation". Clearly we can't just say "I know it when I see it". Ovinus (talk) 00:51, 15 October 2022 (UTC)[reply]
  19. Support, as 3A seems doomed to fail. There does need to be a definition, though I would have preferred to see a definition agreed in this RfC. Scolaire (talk) 12:44, 16 October 2022 (UTC)[reply]
  20. If we are discussing "Article creation at scale", we should know what it is. Note that a simple fixed numerical definition is pointless, it must be conceptual and, if numerical, it must be flexible/adaptative - Nabla (talk) 23:57, 16 October 2022 (UTC)[reply]
  21. This seems to be the most common sense place to start. Agree with Nabla that this shouldn't be a strict bureaucratic number but an overall spirit. (Contrast the WP:3RR with the broader spirit of WP:EDITWAR.) This doesn't have to lead to punishing or limiting mass creation, but being able to bring it into the light of discussion. Jontesta (talk) 23:14, 17 October 2022 (UTC)[reply]
  22. WP:MASSCREATE currently says that creating more than 25-50 articles on the same topic is mass creation, with no additional context about rate or article quality. Judging by a lot of the comments on this page, particularly the ones distinguishing prolific creation from mass creation, that's not an accurate description of the issue at hand. Without another definition, though, we're making a lot of assumptions and talking past each other, and it's not clear what behavior any proposals would affect in practice. TheCatalyst31 Reaction•Creation 03:58, 19 October 2022 (UTC)[reply]
  23. Support if this RfC makes rules about mass creation of articles, the definition of what makes something a mass creation is both important and IMO required. Dreamy Jazz talk to me | my contributions 23:22, 27 October 2022 (UTC)[reply]
  24. If we are going to debate whether creation of articles at scale is permissible, we really do need to have it well defined, so we can all be discussing the same thing. If we can't define it, we are wasting our time trying to regulate it. Having said that, I am also not expecting anyone to come up with a workable definition, per the Granfalloon comment below. · · · Peter Southwood (talk): 19:11, 1 November 2022 (UTC)[reply]
  25. Seems reasonable. Nabla makes a buch of good points. --Guerillero Parlez Moi 11:02, 4 November 2022 (UTC)[reply]

Oppose (Q3B)

Per my support of 3A, I don't think a generic "create definition" will suffice. We need to decide on a concrete definition at this stage. Scolaire (talk) 15:27, 5 October 2022 (UTC) Switching to Support to try to reach consensus on one "create definition" proposal. Scolaire (talk) 12:44, 16 October 2022 (UTC)[reply]
  1. Definitions are impractical and will be circumvented. A smart mass creation artist will always stop just short of the limit. Consider each case on its individual merits. BoJó | talk UTC 09:51, 8 October 2022 (UTC)[reply]
  2. "Article creation at scale" is a granfalloon. Investing time into codifying a definition would just reify a wide variety of practices into a single thing. We've got people worried about the creation of 1000 sports stubs based on single lines in statistics tables from Gray's Sports Almanac, and other people pointing out that biomedical databases are rich with more than enough material to write articles, despite the "database" moniker. This isn't one argument; it's half a dozen different arguments, stemming from deep illusions that "GNG-level" and "SIGCOV" have meanings handed down from Sinai. XOR'easter (talk) 16:51, 8 October 2022 (UTC)[reply]
  3. Defining a threshold for the scale of articles that's okay to create is meaningless. In a vacuum, any particular number of articles—whether it's 10 or 10,000—could be spam or a major contribution to the encyclopedia. If we then try to create a more vague, non-numerical definition to avoid having the policy gamed, then it's not really an effective guideline. What matters is the quality of that encyclopedic content, in terms of verifiability and notability, which we already have well-established policies about which we can use to discuss the merits of articles and whether they should be deleted or not. If you think a particular subject or type of article is inherently not notable, then propose a new notability policy or guideline to specifically address it. A threshold based on the number and size of articles doesn't help us avoid crappy articles, and that blunt of a guideline will most definitely deter helpful editors from writing good articles. Steven Walling • talk 21:17, 11 October 2022 (UTC)[reply]
  4. Pure WP:BURO. I'm sure that this is not intended as such, but this is ultimately acting as just an excuse to do nothing. Wikipedia has many contested concepts (e.g., WP:SIGCOV, WP:BLUDGEON, WP:AUD etc.) but the fact that these are contested does not make them useless or counter-productive. What is an is not mass-creation is something that should be beaten out on a case-by-case basis and need not be predetermined here. Indeed, there are projects where mass creation already has been dealt with (particularly WP:NASTRO) and curiously enough they did not bother to define it explicitly beyond referring to the already-existing WP:MASSCREATE. FOARP (talk) 09:39, 12 October 2022 (UTC)[reply]
  5. Not something that makes sense to define numerically, and should be looked at on a case by case basis.--Ortizesp (talk) 14:44, 20 October 2022 (UTC)[reply]
  6. Oppose, same as all the others. We should not be regulating hard work, merely because someone has the time to do a lot of it. WP:BOLD exists. Follow it. Leave people alone. --Jayron32 18:30, 28 October 2022 (UTC)[reply]

Comments (Q3B)

Comments from Joe Roe (Q3B)

I agree with Rhododendrites that you self-evidently have to have a shared definition of something if you're going to start creating rules and noticeboards about it. But what on earth does these questions mean in the context of an RfC? If there is no consensus to create a definition, does the ordinary English phrase "mass article creation" becomes undefined? If there is a consensus against creating a definition, is it officially decreed undefinable? In the (admittedly increasingly unlikely) event that any new processes come out of this RfC, are we prohibited from describing their scope unless we get the go-ahead here? This whole thing gives me a headache. – Joe (talk) 06:56, 5 October 2022 (UTC)[reply]

Comments from Espresso Addict (Q3B)

I'm liking the idea of a "set of characteristics that mass actions generally have and non-mass actions generally don't that act as guidance not rigid rules" per Thryduulf. I'm not sure how in this framework we might move forward and agree such a thing, particularly as I think the rules of the discussion prohibit mentioning actual cases. Espresso Addict (talk) 19:26, 5 October 2022 (UTC)[reply]

@ActivelyDisinterested: Everyone is certainly capable of recognising mass creation, the intractable problem is that we all appear to be recognising different things. Espresso Addict (talk) 01:43, 8 October 2022 (UTC)[reply]

There seems likely to be consensus for creating some form of a definition. While we're all watching this RfC, is it worth trying to draft something together? Perhaps on the talk page? Espresso Addict (talk) 01:07, 17 October 2022 (UTC)[reply]

Comments from Valereee (Q3B)

Espresso Addict, the rules aren't intended to prevent referring to a past case or cases to allow you to explain why something is necessary or reasonable or a good idea. They're primarily intended to keep people from bickering over past grudges. Please feel free to discuss past cases that are necessary to understand a proposal. Valereee (talk) 20:18, 5 October 2022 (UTC)[reply]

Comments from Andrew Davidson (Q3B)

I agree that a definition is required for practical purposes. Some examples may help in clarifying what people mean by this.

  1. The archetypal case is Rambot – an early bot that created stubs based on the US census. I suppose most everyone would agree that that was mass creation.
  2. Jesswade88 has been lauded for her steady creation of articles about women. She aims to create one every day and still seems to be going strong. I would myself say that this is not "mass" creation because the creations seem separate and discrete rather than being batched and "cookie cutter". The word I'd use for this is "sustained" rather than "mass".

I could go on but you get the idea.
Andrew🐉(talk) 17:25, 6 October 2022 (UTC)[reply]

Comments from ActivelyDisinterested (Q3B)

I think anyone is capable of recognising mass creation. If someone creates 25 well referenced articles in a day, that's a bit odd but it's fine. When a set of articles are all the same apart from small details and have one reference that is a database or list, that's obviously mass creation. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 17:49, 7 October 2022 (UTC)[reply]

Comments from WhatamIdoing 2

ActivelyDisinterested, I think we need a definition that is absolutely objective, precisely because not everyone is capable of recognizing mass creation. Under your "obvious" definition, I can't tell whether 25 well-referenced articles that happened to all be the same apart from small details is mass creation. I can't tell whether just two articles, which happen to be the same apart from small details and have one database reference, is "mass" creation. (Where's the "mass" part of mass creation, when only two articles are involved?) We shouldn't be leaving the definition up to each editor's personal opinion. That doesn't lead to sensible conversations and quick resolution of disputes. That leads to unfair enforcement: I have lots of wiki-friends, so I can create 100 weak articles on my Very Important™ Subjects, but you don't, so I'll get you stopped if you create five on your unimportant subject. That's not healthy for the community, and no matter how much we might wish it otherwise, that is the alternative to having a clear definition. WhatamIdoing (talk) 04:40, 8 October 2022 (UTC)[reply]

@FOARP, it's hard to "take any action" when we first have to (in your words) "beat" out an agreement about whether this or that situation counts as mass-creation/a situation requiring extra scrutiny. Remember, we're here precisely because editors already do not "know it when we see it". If we did, then we presumably wouldn't have half the community claiming that an average of 1.5 articles per day is "mass creation" and the other half yelling back that there's nothing "mass" about one or two articles at a time. WhatamIdoing (talk) 23:48, 13 October 2022 (UTC)[reply]
Comments from FOARP (Q3B)

Should a definition exist? Well, it already does. Indeed more than one definition already exists. I do not oppose coming up with a more formal one at some point - ideally after a period of case-by-case examination - but for Pete's sake let's not let this just become a barrier to taking any action at all in this RFC! FOARP (talk) 09:47, 12 October 2022 (UTC)[reply]


Comments from Nabla

We should have agreed on one before the RfC. - Nabla (talk) 23:57, 16 October 2022 (UTC)[reply]

Comments from Editor X

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 4: Should we prohibit the creation of articles at scale?

This proposal would prohibit the creation of articles at scale based upon a rate definition to be separately decided.

Support (prohibit)

  1. The creation of encyclopedia articles must be understood as a matter of quality, not quantity, and that the rapid creation of articles almost certainly threatens our extant processes for article triage and improvement. Chris Troutman (talk) 20:37, 3 October 2022 (UTC)[reply]
  2. The main problem affecting Wikipedia's status as an encyclopedia is the general lack of reliability. This includes lack of wikitext factual verification, lack of NPOV verification, and (where they exist) lack of citation validation. The exceptions are few, and glaring, as they inadvertently spotlight the vast majority of articles' shortcomings. As a tertiary resource of knowledge, Wikipedia falls short and cannot be trusted. Any action that does not immediately and directly deal with this problem contributes to it by extending and enhancing the status quo. This RFC, and its purpose, are prime examples. Not a single article should be allowed in Wikipedia mainspace unless it is deemed trustworthy by uninvolved editors. A first step would be that it should conform to the relevant Wikipedia mainspace policies, not an extraordinary concept. Any articles not created in this way do not contribute knowledge fit for anybody. Their provenance and accuracy are at best uncertain, and may as well be fiction. Considering the site's history and current state, it is prudent they be considered so unless proven otherwise. 65.88.88.68 (talk) 15:56, 6 October 2022 (UTC)[reply]
  3. A "moral support", but every time I've examined a set of mass-created articles, they always have serious problems. I've never seen it done well. I've seen editors assert it can be done well, but I've yet to see an example. Levivich (talk) 15:40, 26 October 2022 (UTC)[reply]

Oppose (prohibit)

  1. Much too blunt of an instrument. At the most extreme, we're saying we don't want someone to create 51 GAs in a month on various topics? Or we're assuming they're all stubs? If the latter, a more precise question might be to ask whether we want a minimum level of quality for articles created at scale. Update: I've added Q5 accordingly. — Rhododendrites talk \\ 20:52, 3 October 2022 (UTC)[reply]
  2. There are circumstances where mass-creation is quite justifiable. I'd support the more specific prohibition of mass-creation where notability is not automatic. Vanamonde (Talk) 21:20, 3 October 2022 (UTC)[reply]
  3. I think what is needed here is regulation, not prohibition. Seraphimblade Talk to me 21:24, 3 October 2022 (UTC)[reply]
  4. Per everyone above. Thryduulf (talk) 23:24, 3 October 2022 (UTC)[reply]
  5. Mass creation without preapproval should be - and already is - forbidden, but this goes beyond that. BilledMammal (talk) 00:10, 4 October 2022 (UTC)[reply]
  6. Per Rhododendrites. Pinguinn 🐧 04:58, 4 October 2022 (UTC)[reply]
  7. A non-starter. Mass creation—manual or semi-automated—has been one of the most useful tools we have for expanding our coverage of repetitive but encyclopaedic topics (e.g. geographic places, species, sports) and will continue to be for the foreseeable future. We have occasionally had problems with editors doing it badly, yes, but this would be throwing the baby out with the bathwater. It would also be a disaster for efforts to address systematic bias, because we'd be curtailing efforts to expand coverage in under-represented areas, while retaining all the mass-created articles we have now (favouring the Anglosphere/Global North). – Joe (talk) 11:06, 4 October 2022 (UTC)[reply]
  8. Mass creation of articles is OK if done properly. Hut 8.5 12:20, 4 October 2022 (UTC)[reply]
  9. Echoing everyone else. —  HELLKNOWZ ∣ TALK 16:36, 4 October 2022 (UTC)[reply]
  10. * Pppery * it has begun... 18:00, 4 October 2022 (UTC)[reply]
  11. Articles are good, bad articles are bad. We don't want to stop good articles, but minimise bad articles that have to be taken to AfD and clog up NPP. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:41, 4 October 2022 (UTC)[reply]
  12. Plenty of instances where mass creation is perfectly justified (after elections, for example). I do think some parts of this (wider) debate have lost sight of the fact that new articles are the lifeblood of an encyclopedia. Espresso Addict (talk) 01:37, 5 October 2022 (UTC)[reply]
  13. If an editor happens to find a batch of 10 notable topics in a day, and has the time to create properly sourced new articles on all 10 in a day, then that should be encouraged, not prohibited. The problem is the bad sourcing and cookie-cutter nature of some past binges of mass creation, not the volume itself. —David Eppstein (talk) 16:32, 5 October 2022 (UTC)[reply]
  14. On the throwing the baby out with the bathwater and the problem is the bad sourcing and cookie-cutter nature of some past binges rationales articulated above. Are short descriptions of individual species or once-inhabited locales the sort of thing I expect to find when I open an encyclopedia? Yes, they are. Do we benefit people by freeing information that had been locked up in specialized, harder-to-navigate references? Yes, we do. XOR'easter (talk) 16:47, 5 October 2022 (UTC)[reply]
  15. No. We are supposed to be creating an encyclopedia, and it is much closer to its beginning than the end. There is no reason to prohibit the creation of any articles. Phil Bridger (talk) 18:51, 5 October 2022 (UTC)[reply]
  16. I've seen editors who can write 10+ sourced, non-stub articles in a day. There are definitely topics where Wikipedia is missing a lot of notable articles but it's also easy to find sources (e.g. American historic sites), and we want to encourage the creation of quality articles on those topics, not prohibit it. TheCatalyst31 Reaction•Creation 03:10, 6 October 2022 (UTC)[reply]
  17. No, if the articles are good then they are welcome. Graeme Bartlett (talk) 04:00, 7 October 2022 (UTC)[reply]
  18. No reason to prohibit mass creation of good articles that meet GNG and are well sourced. Rlendog (talk) 17:59, 7 October 2022 (UTC)[reply]
  19. Decent, well sourced articles are welcome, however they come about. We don't need to restrict this in such a way Blue Square Thing (talk) 08:29, 8 October 2022 (UTC)[reply]
  20. All that is needed is the requirement to cite reliable, non-database sources which provide significant coverage. Someone might have a couple of books which talk at length about numerous topics that are needed as individual articles. Those articles would then be mass-created but the sources would be reliable and coverage in each would be broad. We would welcome mass creation in that scenario. BoJó | talk UTC 09:56, 8 October 2022 (UTC)[reply]
  21. Wrong approach, why limit the rate if articles subjects demonstrate notability and sourcing. - Donald Albury 14:50, 8 October 2022 (UTC)[reply]
  22. This will start a war regarding WP:STUB, and their creation, which is an accepted guideline. LessHeard vanU (talk) 16:28, 11 October 2022 (UTC)[reply]
  23. No as long as the articles are still acceptable, if they aren't that's the same as if they were created individually other than perhaps the way the author may be dealt with namely the fact they have created more unsuitable pages quicker may make sanctions more likely. Crouch, Swale (talk) 20:24, 11 October 2022 (UTC)[reply]
  24. ~ ONUnicorn(Talk|Contribs)problem solving 20:40, 11 October 2022 (UTC)[reply]
  25. Blanket opposition to the creation of articles is the most boneheaded idea imaginable. Our mission is to create more encyclopedic content, not prevent it. Steven Walling • talk 21:02, 11 October 2022 (UTC)[reply]
  26. Not in these hopelessly wide terms. Ingratis (talk) 06:56, 12 October 2022 (UTC)[reply]
  27. If articles are good, then what does it matter if they are created en masse? We shouldn't try to gatekeep carefully crafted meticulous articles that people spend hours on before moving to the next one. There is nothing wrong with adding a bunch of stubs (that meet the criteria) and improving on them later. Supertrinko (talk) 02:07, 13 October 2022 (UTC)[reply]
  28. Oppose - this should be dealt with case-by-case. In a theoretical scenario where someone is able to crank out a quality article daily that shouldn't be prohibited. Or if someone with expertise wants to make several stubs and improve on them over the next few days to weeks, that's not a problem either. I am of the opinion creating excessive rules will result in a barrier to entry for casual editors who may have valuable experience in a certain topic or field but don't live and breathe Wikipedia (I probably fall into that category). Connor Long (talk) 06:28, 15 October 2022 (UTC)[reply]
  29. No, of course we can imagine a productive case for the above.--Ortizesp (talk) 14:44, 20 October 2022 (UTC)[reply]
  30. Mass creation can help in some cases. --Jax 0677 (talk) 23:26, 20 October 2022 (UTC)[reply]
  31. As previous editors have said, this should be dealt with on a case-by-case basis. Robert McClenon (talk) 05:29, 27 October 2022 (UTC)[reply]
  32. Said better above. Dreamy Jazz talk to me | my contributions 23:23, 27 October 2022 (UTC)[reply]
  33. Not only should we NOT prohibit it, we shouldn't even pay attention to it. WP:BOLD is still policy, and people who have the time, energy, and resources to improve Wikipedia should not be bothered. --Jayron32 18:32, 28 October 2022 (UTC)[reply]
  34. Quantity should not be an issue. More good articles is desirable, it does not matter how many or at what rate, if the content is good, we want them. What we don't want is garbage, and it doesn't matter at what rate the garbage is produced, we don't want it anyway. To be tolerant of the learning needs of new editors, we try to be more tolerant of substandard material from people who appear to be working and learning in good faith, what we are really trying to deal with here are mass creations of substandard articles on unsuitable topics, largely because they drain our limited resources for no benefit. For substandard articles on unsuitable topics, we do not need a definition of mass creation or creation at scale as it does not affect the innate undesirability of the articles. We could just tell people who still create unsuitable articles after a reasonable trial period to stop creating articles because they are not competent. They can still contribute in other ways. This brings up a more relevant issue of how we assess whether a person shows competence in article creation, and one way is the ratio of surviving patrolled articles they have created to the total number hey have created (this should be something a bot could keep track of, and record it somewhere, like in Xtools). If someones articles have 90% survival rate after patrolling, then the person is probably doing adequate work, if the survival rate is 10%, they have a lot to learn, and may not be able to learn it. When the numbers are more evenly balanced, we should consider the residual value of the deleted articles - Why were they deleted? Have they been merged or deleted entirely? I would not hold it against an editor's record if the inadvertently created an occasional valid article that was later found to be a content fork under a plausibly acceptable alias, or found to be more suitable as a section in a broader article, as long at the topic is notable. When the majority or a large number of someone's articles have not been patrolled, a red flag should be raised. It may be necessary for them to slow down a bit or explain why they should not need to stop. · · · Peter Southwood (talk): 08:01, 1 November 2022 (UTC)[reply]

Comments (prohibit)

Comments from JPxG

@Levivich: My guess as to why you have not seen an example of successful mass creation is that this RfC (and most discussions about the issue) have been based on noticeboard discussions where problems are being brought up with editor conduct. Situations where no catastrophe occurs, of course, are rarely brought to noticeboards, and even more rarely brought to arbitration. In order to balance the discussion, I've volunteered an example at Wikipedia_talk:Arbitration_Committee/Requests_for_comment/AfD_at_scale#Mass_creation_of_island_articles_from_GNIS -- my own formulaic mass-creation of island geostubs in 2020 (prior to the current slew of discourse on the issue). None of them have been nominated for deletion, and 25 have subsequently been expanded into GAs (with one of these later losing GA designation upon becoming a FA). jp×g 16:35, 26 October 2022 (UTC)[reply]

Comments from 50.74.109.2

Wikipedia has failed to apply standing policies regarding verification, neutrality etc. years after an uncounted number of offending single articles has been created. In light of such failure, codifying batch creation of articles is a fool’s errand. Despite this, the comments opposing prohibition seem to presuppose/imply that article batches will be somehow validated a priori. Without any proof other than wishful thinking like, "if a bunch of articles are ok [how?] let's publish 'em". This is inherently inapplicable: there is no corresponding tool for validation of article batches. It isn’t likely that any such tool can ever exist. Entities/people who deal professionally with validation and true fact-checking would be already using it if there was one. In the end, all articles in a batch will have to be validated singly, and as pointed in the beginning, anyone can see how this has worked out.

This RfC surreptitiously serves to make Wikipedia’s unreliability more palatable, by obscuring the real problem behind useless, minutely detailed procedure. Well-argued opinions regarding utopian proposals that ignore simple realities will never make sense no matter how politely they are expressed, and will not magically transform these proposals into workable solutions. I’m afraid this is just so much more pompous nonsense. 50.74.109.2 (talk) 01:32, 9 October 2022 (UTC)[reply]

Comments from WhatamIdoing (prohibit)

User:Supertrinko, for sufficiently high volumes of articles, we'd end up with one person flooding the review queues. Even if every article you write looks like it is ready to be nominated at Wikipedia:Featured article candidates from the very first revision, if you do that once a minute round the clock, that's going to cause problems for the New Pages Patrol and the RecentChanges reviewers. We don't really want to surprise the reviewers with the new-article equivalent of an Email bomb.

Previous discussions (years ago) suggested that a rate of less than one or two new articles per hour (i.e., 25 to 50 per day per editor) was likely sustainable for these groups. I don't know what the number is now, or how likely anyone is to reach it in practice, but there probably is a threshold beyond which the reviewers would feel overwhelmed and discouraged, and we would risk them either clearing or rejecting articles en masse just to get them out of the queue. WhatamIdoing (talk) 23:58, 13 October 2022 (UTC)[reply]

I have to disagree with such limitations as one or two per hour. 50 articles per day per editor is so, so small. I feel it is on New Pages Patrol to adapt to the number of articles that are coming in, not on them to limit the number of new articles allowed. Supertrinko (talk) 05:19, 14 October 2022 (UTC)[reply]
@Supertrinko, I believe that, during the last year, only about 25 editors ever managed to make more than 50 articles in a single day, and none of them did it frequently. This is a quite unusual achievement. If you are looking at the average number of articles created over a long time period, I believe the record for all of last year was averaging around three or four articles per day, and the record for a single month was around 15 articles per day during that month. There are links to some statistics in Wikipedia:Arbitration Committee/Requests for comment/Article creation at scale#Statistics for mass creation if you'd like to get the real numbers instead of relying on my memory. WhatamIdoing (talk) 05:55, 14 October 2022 (UTC)[reply]
If we encourage and enable mass article creation, then that number would be higher. Supertrinko (talk) 08:06, 14 October 2022 (UTC)[reply]
@Espresso Addict and @Levivich, about "[E]very time I've examined a set of mass-created articles, they always have serious problems": I think the question in play here is "What constitutes a serious problem?" A few years back, an editor created a large number of articles about different medical conditions. They were all sourced to a high-quality medical textbook. He created hundreds of them, and only a small percentage were either merged or deleted.
But: they are all very short, and most of them cited just a single source. Some people automatically consider short articles to represent "serious problems". Others will not. We don't have a shared understanding about what constitutes a serious problem. WhatamIdoing (talk) 20:49, 30 October 2022 (UTC)[reply]

Comments from Espresso Addict (prohibit)

Thanks for the reminder of the BilledMammal's stats, WhatamIdoing; I took a look at the recent creations of one of the most prolific creators, and found that the bottom line is a bit deceptive; a lot of the total is disambiguations -- I'm guessing they can't be excluded? Averaging two articles a day when one is a bio article and the other a disambiguation for that surname doesn't seem all that disruptive. Espresso Addict (talk) 06:17, 14 October 2022 (UTC)[reply]

@Levivich:: "[E]very time I've examined a set of mass-created articles, they always have serious problems" I think this is the definition of mass creation vs whatever we want to call unproblematic creation of multiple articles. Espresso Addict (talk) 20:45, 26 October 2022 (UTC)[reply]

Question 4A: Should we restrict creation of articles at scale?

Proposed: Mass-creation is permitted for a group of topics when all members of that group meet a criterion directly granting notability. All other mass-creation is prohibited unless supported by consensus at the mass-creation noticeboard (see Q1).

Addendum for clarity, not part of the proposal: criteria that have traditionally been held to grant notability include the GNG (of course), but also SNGs such as NPOL, NPROF, CREATIVE, GEOLAND, and a few others. Topics that meet these standards are not typically also required to meet GNG. The same is not true of other SNGs (notably NSPORTS). The distinction being made in the proposal is between these two categories of notability criteria.

Support (limit)

  1. As proposer, and per comments above. Mass-creation is a problem only when the notability of each created item is not demonstrated. There are cases where groups of topics do meet this criterion (under NPOL, for instance, or GEOLAND) where it isn't a problem. Vanamonde (Talk) 21:28, 3 October 2022 (UTC)[reply]
  2. That seems fair enough. Several SNGs are understood to be independent of the GNG (for example WP:NPROF: This guideline is... explicitly listed as an alternative to the general notability guideline). If the subject of an article meets one of those then there is no need to show that the subject passes the GNG. If you don't think any SNGs should work that way then make a proposal to change the SNGs which don't. Hut 8.5 12:20, 4 October 2022 (UTC)[reply]
  3. I'd See it as fair that if one creates 5 articles a day that they come under review and if they fairly elaborated articles, they should be allowed to create them on, but if they create one-two-three line stubs (with maybe an infobox) they should temporarily be only allowed to create drafts and submit them for review to see if they pass the AfC requirements.Paradise Chronicle (talk) 14:15, 5 October 2022 (UTC)[reply]
  4. Limiting mass creation allows time for proper scrutiny, especially with regards to SNGs that don't necessarily mean GNG is also met. Der Wohltemperierte Fuchs talk 22:07, 5 October 2022 (UTC)[reply]
  5. If the do meet the inclusion criteria, then that removes a problem. I also think that the articles created en masse should also prove that they meet the criteria. Graeme Bartlett (talk) 04:03, 7 October 2022 (UTC)[reply]
  6. Some restriction on article creation at scale is clearly necessary, and this is at least a good start. Most of the opposes seem answered by the fact that approval can be sought at the mass-creation messageboard.--Aquillion (talk) 05:50, 28 October 2022 (UTC)[reply]

Oppose (limit)

  1. Opposing mainly because I don't understand. Following on the proposer's support !vote, how does passing a SNG like GEOLAND or NPOL not make something notable? I mean, there's the "presumed" language of those guidelines, but that's even in the GNG. What's the distinction being made here? If this boils down to "articles created at scale need to be notable", that's pretty uncontroversial because all articles already have to be notable. If it's that articles have to meet the GNG, it would be clearer to propose that explicitly. I'd also disagree that notability is the only problem with mass creation. Sourcing and rate are the issues I see come up most often. — Rhododendrites talk \\ 22:58, 3 October 2022 (UTC)[reply]
    Adding post-addendum: Thanks for clarifying. I'd still go by the last part of what I wrote: all new articles have to be notable, and I don't think it particularly matters which notability guideline applies. What matters is more about sourcing and rate, so I'm not inclined to introduce a distinction that gives thumbs up/thumbs down just based on notability-related criteria. — Rhododendrites talk \\ 12:41, 24 October 2022 (UTC)[reply]
  2. Per Rhododendrites. Thryduulf (talk) 23:27, 3 October 2022 (UTC)[reply]
  3. Mass creation is a problem even when the articles meet SNG's, as SNG's are subject to change, and the mass created articles on the basis of SNG's are almost always low quality. BilledMammal (talk) 00:12, 4 October 2022 (UTC)[reply]
  4. SNGs should not be "granting" notability to start with, only pointing out when it is likely to exist. I certainly don't want anything that would further encourage that. Seraphimblade Talk to me 01:57, 4 October 2022 (UTC)[reply]
  5. Either it's notable or it isn't. SNGs are there to help clarify when articles should be created, not mandate what articles are created. Pinguinn 🐧 05:02, 4 October 2022 (UTC)[reply]
  6. Notability would be better handled by ensuring articles have some minimum referencing in place. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:48, 4 October 2022 (UTC)[reply]
  7. Per WP:CREEP. This seems to be pointless rule-making. I don't see any situation where making this into any kind of policy or guideline would change anything we do about mass creation. —David Eppstein (talk) 16:33, 5 October 2022 (UTC)[reply]
  8. This is the same question as above, just with a carve-out for SNGs. I don't think this would solve our problem, since (a) a lot of problematic mass creation has been in areas that ostensibly were covered by an SNG, like Carlossuarez46 and GEOLAND, and (b) a lot of productive mass-creation isn't covered by a formal SNG because the articles literally never get deleted at AfD anyway (species, sites on the National Register of Historic Places, etc.) TheCatalyst31 Reaction•Creation 03:14, 6 October 2022 (UTC)[reply]
  9. Per David Eppstein. Also, there is no reason to prohibit or even limit mass creations that are well-sourced and meet notability criteria, regardless of the status of other members of the "group". Rlendog (talk) 18:06, 7 October 2022 (UTC)[reply]
  10. Decent articles that are well sourced are decent, well sourced articles. We shouldn't restrict this in such a way - it might be that only 10% of <insert name of large topic> articles are notable; we should have those articles - and if they're mass created then that's fine Blue Square Thing (talk) 08:31, 8 October 2022 (UTC)[reply]
  11. Same as for Q4 above. Sources are what matter. This idea is making a rule for the sake of making a rule. Keep things simple. BoJó | talk UTC 09:59, 8 October 2022 (UTC)[reply]
  12. Same as above. Why limit the creation of acceptable articles? - Donald Albury 14:52, 8 October 2022 (UTC)[reply]
  13. Weakly for my reason above. Crouch, Swale (talk) 20:25, 11 October 2022 (UTC)[reply]
  14. Creating a rule where you have to ask permission to create encyclopedia articles is a surefire way to deter people from volunteering their time to expand and improve the encyclopedia. Steven Walling • talk 21:04, 11 October 2022 (UTC)[reply]
  15. There's no reason mass article creation should be allowed in some areas, but not others. If the article is notable, then it can be improved later. Supertrinko (talk) 02:09, 13 October 2022 (UTC)[reply]
  16. I can't see a reason why you wouldn't welcome it. scope_creepTalk 18:09, 14 October 2022 (UTC)[reply]
  17. No, there's cases where this would be fine and even encouraged.--Ortizesp (talk) 14:45, 20 October 2022 (UTC)[reply]
  18. Hard to understand, but appears too restrictive. Robert McClenon (talk) 05:31, 27 October 2022 (UTC)[reply]
  19. Same as all of the other, leave people doing good work alone. Let them be. Someone does not have to get permission to work at Wikipedia, and we shouldn't be hounding people who do the most work. --Jayron32 18:32, 28 October 2022 (UTC)[reply]
  20. Per above. My first edits were all new articles. --Rschen7754 01:32, 2 November 2022 (UTC)[reply]

Comments (limit)

Comments from Vanamonde (limit)

This looks like it's headed for failure already, because the variation in standards among the SNGs is simply being ignored. Some SNGs are written such that they confer notability independent of GNG. Others are not. If a person meets NPOL, they are not required to meet GNG for an article to exist. The same is not true of NSPORTS. Seraphimblade, I'm particularly confused by your position; because not only do some SNGs already grant notability (to the same extent as GNG), mass-creation is currently being justified on the basis of SNGs that do even less. This proposal is intended to restrict that, not enable it. Vanamonde (Talk) 04:16, 4 October 2022 (UTC)[reply]

  • @TheCatalyst31: I appreciate that the proposal won't solve the whole problem, believe me, but you imply that it would address some pieces of the problem; so why is it a bad thing? Vanamonde (Talk) 14:57, 6 October 2022 (UTC)[reply]
  • @Rhododendrites: I added a clarification above in response to your comment, but realized you may have missed it in all the back-and-forth; does that address your concerns at all? Vanamonde (Talk) 17:35, 7 October 2022 (UTC)[reply]
  • Rlendog The proposal doesn't do what you say it does though; mass-creations that meet GNG, for instance, are explicitly exempt. Vanamonde (Talk) 21:08, 7 October 2022 (UTC)[reply]
Comments from Seraphimblade (limit)

For clarity, my oppose is on the grounds that we never should have had SNGs like the "professor" and "populated place" ones, which purport to confer notability even in the absence of substantial good quality source material about the subject. I'm already seeing a growing amount of discontent regarding that, and don't want something which legitimizes that practice further. I hope relatively soon there, we will see a change in those areas and a cleanup of them like we once had to do with fiction and more recently have been doing with sports biographies. (Of course, not having it on a separate page entirely doesn't mean we shouldn't have anything about it; most populated places, for example, would fit quite nicely on a page like "List of populated places in Example County, Somestate".) Seraphimblade Talk to me 07:31, 4 October 2022 (UTC)[reply]

Comments from Editor Paradise Chronicle (limit)

Having reflected on my vote, I'd like to add that the very basic stubs (if properly sourced) are of course also a valuable contribution of the mass creators to the Wiki project, but maybe (probably) their edits would be more appreciated at Wikidata than Wikipedia.Paradise Chronicle (talk) 09:51, 6 October 2022 (UTC)[reply]

Comments from TheCatalyst31 (limit)

@Vanamonde93: To be clear, I meant the problems with the original Question 4 as much as I did the problem of mass-creation. Excluding topics with SNGs would still leave this policy as a blunt instrument that would prevent productive mass creation (if a bit less so), but it would also move it away from addressing the core issue at hand. TheCatalyst31 Reaction•Creation 01:01, 7 October 2022 (UTC)[reply]

Comments from FOARP (limit)

Vanamonde - Mass-creation is ABSOLUTELY a problem for GEOLAND, precisely because of the supposed automatic notability it creates. Please look at this ARBCOM discussion, this AN discussion, and these AFDs 1 2 and this entire project that has been operating for years now to clean up the mess of mass-created GEOLAND articles JUST IN CALIFORNIA. Sorry to go all-caps but really, it's an issue. FOARP (talk) 08:31, 7 October 2022 (UTC)[reply]

Joe - The global south is possibly even more likely to be the victim of failing mass-created articles than the north. It's just easier to say that an entire class of things on a database all deserve articles when the number of people who will understand that they don't is much more limited. Carlossuarrez46 was only able to get away with mass-creating articles about Iranian "villages" that were not actually villages because no-one who spoke Farsi was there to correct him. Ditto Lugnut's articles about India/Pakistani cricketers. We are not helping people in the Global south by loading up Wiki with content-less articles of this kind. FOARP (talk) 08:31, 7 October 2022 (UTC)[reply]

Comments from Andrew Davidson

We don't have clarity on the definition yet. But supposing that we're talking about bulk creation of the Rambot sort then this might be restricted along similar lines to the existing procedures used at WP:RFBOT, WP:AUTOPATROL and WP:AWB in which editors are given trusted status to do high volume edits. Article creation is just a special case of editing in general. Andrew🐉(talk) 10:41, 7 October 2022 (UTC)[reply]

Comments from Joe Roe (limit)

@FOARP: I don't think anyone here is going to say that Carlossuarrez46 is someone to emulate. But hard cases make bad law and he didn't really "get away with it" in the end, did he? What I'm saying is that at this point we have articles on, for example, pretty much every single populated place in North America and Western Europe; I simply cannot see how we will ever approach that level of coverage for other parts of the world without some form of mass creation from databases. To cut off that possibility now, after using it so successfully to seed our coverage of the affluent, English-speaking world, would be fundamentally unjust. – Joe (talk) 11:59, 7 October 2022 (UTC)[reply]

Comments from Rlendog.

Vanamonde93 The proposal as written does not explicitly exempt mass creations that meet GNG. In any case, whether they meet GNG can often be subjection - hence many debates at AfD. Which is why I think (per my comment in #2) the criteria should be at least one source that plausibly meets GNG. Rlendog (talk) 15:43, 8 October 2022 (UTC)[reply]

Question 5: Minimum article quality when created at scale

Articles created at scale should be required to meet a certain level of quality in addition to minimum sourcing requirements (see Q2).

For example: minimum number of sentences, article size, assessment, ORES score, etc.).

If you support this, you may suggest qualitative or quantitative standards, but a separate question will be required to find consensus for specific requirements.

Support (minimum quality)

# Support. Articles that: 1)were disapproved by the community at the noticeboard for article creation at scale or the Bot Approvals Group or 2)contain no sources or 3)contain only: deprecated sources or sources that are easily editable by unqualified people should be speedily deletable. 1) is equivalent to a deletion discussion, 2 and 3 suggest that the author didn't have any standards whatsoever and their creations cannot be trusted.Lurking shadow (talk) 21:19, 3 October 2022 (UTC) Moved down.Lurking shadow (talk) 07:16, 5 October 2022 (UTC)[reply]

  1. Support. Mass created articles (in vast majority stubs) mainly live from their creators (or are notable for their creators) and then after, are seldomly read or expanded for years. I suggest to enable the draftification of stubs (sourced or not) per wikipedia guideline and if there is interest by their creators (or the community) they'll expand to meet the AfC requirements.Paradise Chronicle (talk) 07:36, 5 October 2022 (UTC)[reply]
  2. Tentative support, looking at the ideal that an article should be more than a database entry. The support is tentative given the potential difficulty in finding an actionable formulation for what is quite a qualitative consideration. CMD (talk) 08:24, 5 October 2022 (UTC)[reply]
  3. Mass-created articles are always undesirable unless they are of a certain quality. Requiring one or two RSs alone does not guarantee that articles will be encyclopaedic. Scolaire (talk) 15:33, 5 October 2022 (UTC)[reply]
  4. THe articles should be useful to a reader. They should also have enough referencing to show that they meet inclusion criteria. One sentence could be enough though, if it is useful to a reader. Graeme Bartlett (talk) 04:04, 7 October 2022 (UTC)[reply]
  5. See my essays at Wikipedia:Substantive content, @FOARP: would that kind of guidance work? Crouch, Swale (talk) 20:27, 11 October 2022 (UTC)[reply]
  6. Articles in general should meet a minimum standard of quality, but this especially applies to mass-created articles that are harder to address in mass. We are beyond the point where hundreds of new one-liners can be expected to receive actual improvement later. Reywas92Talk 17:24, 12 October 2022 (UTC)[reply]
  7. Support Absolutely with cherries. What is the point of it, otherwise. To create 40k of half-cut, partially damaged articles, or missing crucial information. Why would you do that? You want complete articles that are functionally perfect and contain as much as information about the subject as possible, with as many sources as needs to WP:V the content. Otherwise what is the point? Are you planning to do multiple passes. Who does that (although I'm not sure, its an unknown). Your not going to create 40k of articles and then do a second pass, with some new configuration with your software, adding other sources for example. Your going to do all of it, upfront. I have strong feeling I don't think people realise the change that going to come, with these automated tools, particularly AI. If someone created 40k of article, nodoby is going to update them, perhaps even not the person who creates them. So they need to be as complete as possible at the beggining. scope_creepTalk 17:07, 14 October 2022 (UTC)[reply]
  8. Support. And I like @Paradise Chronicle:'s idea of stubs that are never viewed or improved be moved into draft format. WomenArtistUpdates (talk) 00:59, 21 October 2022 (UTC)[reply]
  9. Support. Some sort of standards or guidelines are necessary to evaluate when mass-article-creation is appropriate, otherwise any other restrictions (which are sorely needed) would just be spinning their wheels. --Aquillion (talk) 05:51, 28 October 2022 (UTC)[reply]
  10. Support. Creating a large number of badly written articles is disrespectful of those who clean up the mess. If creating large numbers of articles, the creator should be sufficiently competent to do it properly, and comply with MoS, and reasonably correct spelling and grammar. · · · Peter Southwood (talk): 18:58, 1 November 2022 (UTC)[reply]

Oppose (minimum quality)

  1. IMO the basic requirement of containing a source with SIGCOV (as well as any rules that apply to non-mass-created articles) is sufficient. * Pppery * it has begun... 18:00, 4 October 2022 (UTC)[reply]
  2. Per Pppery. —David Eppstein (talk) 16:34, 5 October 2022 (UTC)[reply]
  3. There's no way to quantify "quality" that can't and won't be gamed. Nor is there a meaningful notion of "quality" that could extend over all topics on which articles can be created, beyond vague platitudes. XOR'easter (talk) 16:36, 5 October 2022 (UTC)[reply]
  4. Quality's a bit of a hard thing to systematically measure. There have been sanctions to restrict users so that they are only allowed to create articles of minimum length (such as 500 words), and maybe some concrete proposal along these lines would be rational. But, at the same time, there are some topics that are notable under current guidelines (such as legally recognized populated places) that can be quite easily created and described at a small prose size while not being terrible. I'm not principally opposed here, but I'm not able to do anything but oppose without a good mechanism for judging the nebulous character of "quality". — Red-tailed hawk (nest) 22:19, 5 October 2022 (UTC)[reply]
  5. Per XOR'easter and Red-tailed hawk. The number of sentences, and article size, are not necessarily a measure of quality. The grading of an article does not necessarily reflect the quality of that article due to human error in applying the (less than completely objective and unambiguous) criteria at WP:ASSESS. The ORES score of an article does not necessarily reflect the quality of that article due to the limits of articlequality model. James500 (talk) 01:23, 6 October 2022 (UTC)[reply]
  6. Per above. Measuring quality and setting a minimum standard of it is difficult and can be gamed. Basic requirements that apply to normal articles should be sufficient for mass created ones too. The problem we need solving is not that mass created articles are incomprehensible. The problem is that such articles (normal or mass created) would've been deleted or not deleted based on the inherent notability of the "subject", and that should be maintained no matter the rate of creation. —CX Zoom[he/him] (let's talk • {C•X}) 12:56, 6 October 2022 (UTC)[reply]
  7. Even a sourced one line stub can be beneficial to the encyclopedia, if for nothing else than a starting point for expansion. Rlendog (talk) 18:05, 7 October 2022 (UTC)[reply]
  8. This sort of thing can work as a sanction at ANI etc... when a problem editor is identified. It's not suitable as a one size fits all approach Blue Square Thing (talk) 08:33, 8 October 2022 (UTC)[reply]
  9. Quality is a matter of opinion and all articles can be improved. The key factor is sourcing. BoJó | talk UTC 10:03, 8 October 2022 (UTC)[reply]
  10. Per WP:STUB and WP:NOTABILITY, it is the hard to define attempt to simply create a mass of articles to inflate an editors creation listing that it the issue that needs addressing. LessHeard vanU (talk) 16:34, 11 October 2022 (UTC)[reply]
  11. We already have clear guidelines for notability and verifiability that cover all possible articles, and we also already have processes for developing a consensus about when articles need to be deleted. The last thing we need is more rules. Steven Walling • talk 21:06, 11 October 2022 (UTC)[reply]
  12. Articles can always be improved later. There is no reason to put this rule on mass created articles specifically. Supertrinko (talk) 02:10, 13 October 2022 (UTC)[reply]
  13. I'll put my opposition here just in the narrow sense of species articles. Per WP:SPECIESOUTCOMES as long as a species has been described in reliable sources, it gets an article, whether at scale or not. Usually those articles do end up as stubs initially. I don't have any other strong opinions on the rest of this RfC, but that is one aspect that needs to be followed regardless. The only way for such an article to be deleted is if it was a made up article. If "minimum quality" is taken to mean "it exists in sources" in this instance, then consider my oppose a neutral/meh. KoA (talk) 17:49, 13 October 2022 (UTC)[reply]
  14. If something is acceptable as a manually created article, the same thing should be acceptable as part of a mass creation too. Not necessary to have a seperate quality requirement for mass created articles. ಮಲ್ನಾಡಾಚ್ ಕೊಂಕ್ಣೊ (talk) 07:31, 16 October 2022 (UTC)[reply]
  15. No, no inherent issues with stubs or even microstubs depending on context.--Ortizesp (talk) 14:46, 20 October 2022 (UTC)[reply]
  16. Again, this is too restrictive, and should be on a case-by-case basis. Robert McClenon (talk) 05:33, 27 October 2022 (UTC)[reply]
  17. Again, as with every other proposal, we shouldn't be putting the breaks on people who are doing good work. This entire train of thought that just because someone does a LOT of work, they or their work are under suspicion automatically is beyond rude. --Jayron32 18:34, 28 October 2022 (UTC)[reply]
  18. Stubs and redlinks attract editors to Wikipedia. We need more of both. WaggersTALK 11:48, 1 November 2022 (UTC)[reply]
  19. Oppose minimum standards above what qualifies for CSD A1. --Rschen7754 01:30, 2 November 2022 (UTC)[reply]

Comments (minimum quality)

Comments from Rhododendrites (minimum quality)

Adding this question because it comes up frequently and would be useful to resolve one way or the other. Not supporting or opposing at this time. — Rhododendrites talk \\ 21:07, 3 October 2022 (UTC)[reply]

@Lurking shadow: I agree with those SD criteria for articles created as scale. Moving forward with creation despite rejection at the TBD venue seems straightforward but isn't really related to quality. My thinking with this question was "the community will only approve requests to mass create articles that meet X, Y, Z criteria for minimum quality," and specifically "in addition to minimum sourcing requirements" set forth in Q2. I imagine 2 and 3 would be covered by Q2, no? — Rhododendrites talk \\ 21:25, 3 October 2022 (UTC)[reply]

Comments from Pinguinn (minimum quality)

I don't see how this would be workable. Of the four suggested criteria, the first and second lend themselves to poor writing, as any student tasked with meeting a length requirement on an assignment can attest. The third is based on purely subjective criteria meant to be used by individual reviewers. The fourth is based on an algorithm based on machine learning, not objective criteria. Though there are undoubtedly other ways to measure articles, I strongly suspect they'll all fall short as either not objective or not indicative of article quality. Pinguinn 🐧 05:22, 4 October 2022 (UTC)[reply]

Comments from Abductive (minimum quality)

The problem is not an undefinable "quality" or lack thereof, but whether the articles are "encyclopedic". In the case of species, there are often users who mass-create stubs which are basically copies of the database entries that they are sourced to. These creations will meet (or can be made to meet) any of the above definitions of notability or quality, but are still useless. Why? Because they don't provide any more information to the reader than the database item that they would find if they Googled a species with no Wikipedia article. Fundamentally, the problem is users who are unclear on the scientific rationale for us humans calling something a species, and on the purpose of an encyclopedia. But this cannot be codified for even a straightforward topic such as species, let alone for all possible mass-creation endeavors. Abductive (reasoning) 14:08, 4 October 2022 (UTC)[reply]

Comments from ActivelyDisinterested (Q5)

I'm not going to vote on this yet, as proposed it looks bureaucratic. However I wonder if something is needed in regard to WP:BLPs, were mass creation of stubs could be problematic. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:54, 4 October 2022 (UTC)[reply]

Comments from Espresso Addict (Q5)

I feel this is the direction we ought to be heading, but I can't see how a definition can be made that won't just lead to deliberate bloating. That's certainly been my experience with creators wanting to get over the 1500-character DYK threshold. Espresso Addict (talk) 19:43, 5 October 2022 (UTC)[reply]

Comments from TheCatalyst31 (Q5)

I'm interested in this proposal, if only because it's similar to how WikiProject National Register of Historic Places solved its own mass creation problem years ago. The project essentially took a hard line against two-sentence stubs with only a database source that didn't explain why a historic site was historic, and while it took some effort and an ArbCom case to informally put it into practice (and we still haven't cleaned up all the substubs created beforehand), it's at least helped us cut down on new articles that don't state their importance. That being said, I would like to see how this would be implemented before I support; it's easier to do this for a single topic area than the entire project, and short stubs are more valid for some topics than others. TheCatalyst31 Reaction•Creation 03:36, 6 October 2022 (UTC)[reply]

Comments from WhatamIdoing (minimum quality)

The devil is in the details. In principle, I would support a low minimum quality (e.g., two sentences + two sources) but I would never support a minimum quality that is higher than the quality of the median Wikipedia article. The median article is a stub, and although we don't actually have exact stats yet (I asked for one recently), my guess is that the median article is about 8 sentences and 3 inline citations (including non-independent sources). A new article should never be required to be better than 50% of existing articles. WhatamIdoing (talk) 04:49, 8 October 2022 (UTC)[reply]

Comments from Scolaire (minimum quality)

Twenty years ago many articles were created as stubs and grew to be sizeable articles; this happens very rarely now. Stubs tend to be just left; in the case of creation at scale, they are left by the article creator who doesn't care that users are not going to volunteer to expand them. This harms the project, I believe. In the FritzpollBot poll in 2008 I said: "A paper encyclopaedia is biased towards what the editors want us to know. Wikipedia is biased towards what we ourselves want to know. Correction of that bias by creating articles about what we don't want to know serves no educational purpose while creating the potential for harm (accidental or deliberate misinformation etc.)." The result of that poll was that the bot was approved, but it never accomplished its task, to the relief of many. Mass-created substubs are doing the same thing in 2022, with the same undesirable consequences. Scolaire (talk) 15:06, 10 October 2022 (UTC)[reply]

Comments from Editor X (minimum quality)

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 6: New mass creator permission

Withdrawn

A new permission would be established to be able to create articles at speed. This would be allocated by an administrator, on the basis of the candidate having a history of appropriate creations as well as clear evidence of understanding relevant guidelines. A 7-day window from application to granting of permission would be required so that other editors can contribute evidence relating to the candidate's competence. Inappropriate use of this permission would be raised at the mass creator noticeboard (if established) or at the administrator's noticeboard, with a relatively low threshold of proof being necessary for revoking the permission. Withdrawing. Espresso Addict (talk) 19:50, 5 October 2022 (UTC)[reply]

Support (New mass creator permission)

  1. Support as nominator. Espresso Addict (talk) 01:57, 5 October 2022 (UTC)[reply]

Oppose (New mass creator permission)

  1. I cannot conceive of any way that this will be useful, if even workable. Thryduulf (talk) 09:40, 5 October 2022 (UTC)[reply]
  2. Bad idea, do not see how this solves any issue, likely to just result in it being seen as a permission just to create large amounts of failing articles the same way some people see autopatrolled. FOARP (talk) 12:35, 5 October 2022 (UTC)[reply]
  3. I'm opposed to having hard limits to define mass creation, as it will only lead some to make sure they just scrap past that minimum and other to argue that they haven't. Without those hard limits I don't see how this would be possible. If those limits are put in place, then I expect we'll see the issue seen with autopatrolled as mentioned by FOARP. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 13:28, 5 October 2022 (UTC)[reply]
  4. This seems like another hat for hat-collectors to collect, and will only encourage them to mass-create to justify having the hat. —David Eppstein (talk) 16:35, 5 October 2022 (UTC)[reply]
  5. I don't see the purpose unless and until we limit the rate of article creation far more than we already do. Vanamonde (Talk) 19:36, 5 October 2022 (UTC)[reply]

Comments (New mass creator permission)

Comments from Joe Roe (Q6)

What would the permission allow you to do that ordinary editors can't? – Joe (talk) 06:45, 5 October 2022 (UTC)[reply]

@Espresso Addict: (Oh, isn't this so much better than threaded discussion?) Maybe something like the AfC pseudoright or the redirect autopatrol list? Or come to think of it, could this be somehow attached to autopatrolled? As far as I know it can't really be a permission if there isn't an associated technical ability. – Joe (talk) 07:01, 5 October 2022 (UTC)[reply]
Comments from Espresso Addict (Q6)

Joe Roe: The idea is that only those editors with the permission would be able to do whatever it is we eventually define as mass creation without sanctions. Clearly I need to make that clearer somehow, though not sure how I am meant to do that. Espresso Addict (talk) 06:52, 5 October 2022 (UTC)[reply]

@Joe Roe:: I'd envisaged that once we come up with a definition for mass creation in terms of numbers of articles per unit of time, editors lacking the permission who exceeded the threshold would be forced to wait to create new articles in mainspace. I imagine it would often be assigned with autopatrolled but I was not envisaging it as a subset of that. If we fail to come up with a machine-codeable definition, then it would work as a personally assigned exemption to semi-automatic sanctions for engaging in mass creation. Espresso Addict (talk) 07:22, 5 October 2022 (UTC)[reply]
Comments from Thryduulf (Q6)

The definition of what is and isn't "mass creation" is of necessity going to be blurry and not reducible to numbers. Even if it were it wouldn't be useful as what matters is not who is mass creating but what is being mass created. 09:43, 5 October 2022 (UTC)

Comments from Editor X

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 7: Should we adopt a new speedy deletion criterion that relates to mass-created articles that lack any sourced claims of importance?

This proposal is appropriate for the RfC on AfD; discussion at WT:ACAS. Has been added to a section at WT:ADAS.

This proposal would create a new speedy deletion criterion, A12, as follows:

A12: No reliably sourced indication of importance (mass-created articles).

This criterion applies to any mass-created article that does not have a reliably sourced indication of importance. This would apply to any mass-created article that does not indicate why its subject is important or significant. This is a lower standard than notability. If the sourced claim's importance or significance is unclear, you can improve the article yourself, propose deletion, or list the article at articles for deletion.

Support (proposal name)

  1. This establishes a bare minimum for mass-created articles to avoid speedy deletion, which is that they contain a source that verifies their claim of significance. It differs from WP:A7 both in terms of what sorts of articles are in its scope and its requirement for a bare minimum amount of sourcing rather than merely including a credible claim of significance. This is minimally burdensome on article writers and, in my view, avoids the problems in the framing of Question 2 that several editors have objected to. This proposal is not mutually exclusive with any other proposal made thus far and can serve in its own right as a remedy or as a supplement to other policy changes proposed in this RfC. — Red-tailed hawk (nest) 04:11, 5 October 2022 (UTC)[reply]
  2. Support. You should not have to repeat the same argument again and again for essentially identical articles. We should not accept the fiction that these articles (e.g., Carlossuarez46's abadi and GNIS articles) which are written out simply by filling spaces in a template based on the same database source are actually different articles - if one fails they all fail. Doing otherwise means allowing mass-creators to establish a fait accompli that no-one can challenge. FOARP (talk) 12:38, 5 October 2022 (UTC)[reply]
  3. Espresso Addict makes a good point but at the moment all that work falls on AfD and NPP, which has been shown to be unsustainable. Creating a new way this could be handled is needed. Some framework could be put round this, that it is only for obvious candidates and misuse being subject to sanction. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 13:02, 5 October 2022 (UTC)[reply]
  4. Support as one option. Some mechanism to actually enforce our mass-article creation criteria is necessary to prevent WP:FAIT situations; currently it is much easier to mass-create articles than to reverse that action. So we need some sort of CSD for the most straightforward cases, and this is one valid option for that. --Aquillion (talk) 15:04, 5 October 2022 (UTC)[reply]
  5. Support - Even if this will be difficult to use, having one option for handling poor mass creation is better than having no options. -Indy beetle (talk) 08:34, 13 October 2022 (UTC)[reply]

Oppose (proposal name)

  1. Oppose. As an admin who works in speedy deletion, it is already very hard to see consensus between admins over what constitutes an A7-able article. Broadening to all types of articles and adding in the requirements (1) to check that the article forms part of a mass creation set, and (2) that the source for the claim not only exists but is reliable, in my opinion takes this well beyond the limited amount of consideration that speedy deletion is meant to encompass. Espresso Addict (talk) 04:23, 5 October 2022 (UTC)[reply]
  2. Oppose. I'd like to be at Support, but Espresso Addict is right.Lurking shadow (talk) 07:12, 5 October 2022 (UTC)[reply]
  3. Oppose per Espresso Addicit and my comments below. Thryduulf (talk) 09:46, 5 October 2022 (UTC)[reply]
  4. Strong oppose without ensuring we have a clear definition of "mass creation" to work by. Also, this is the sort of thing the next RfC is for (after we get the basic guidelines for mass creation settled). — Rhododendrites talk \\ 13:10, 5 October 2022 (UTC)[reply]

Comments (proposal name)

Comments from Thryduulf (Q7)

Determining whether an article is part of a mass creation set may mean analysing months of contributions which cannot be done by admins patrolling speedy deletion categories, also one reason A7 is topic-limited is that it is not possible for a single admin to reliably determine what is and isn't a claim of significance in all subject areas. This is also out of scope for this RfC. 09:49, 5 October 2022 (UTC)

Comments from Pbsouthwood (Q7)

Importance is subjective. It is not an appropriate criterion. · · · Peter Southwood (talk): 18:44, 1 November 2022 (UTC)[reply]

Comments from Editor X

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 7a:

For AfD RfC; Has been added to a section at WT:ADAS.

Should we instead introduce the following speedy deletion criterion:

A12: Unsourced or obviously unreliably sourced mass-created articles.

This criterion applies if the mass-created article has been unsourced in all of its history, or is only sourced to obviously unreliable or deprecated sources in all of its history.

Support (proposal name)

  1. SupportThis is more narrow. If the author doesn't put this bare minimum of effort into the articles then the mass-creation cannot be trusted.Lurking shadow (talk) 07:12, 5 October 2022 (UTC)[reply]
  2. Support for the same reasons discussed above. FOARP (talk) 12:40, 5 October 2022 (UTC)[reply]

Oppose (proposal name)

  1. It is impossible for a single editor patrolling a speedy deletion category to reliably determine whether an article is part of a mass creation set. Thryduulf (talk) 09:51, 5 October 2022 (UTC)[reply]
  2. Strong oppose without ensuring we have a clear definition of "mass creation" to work by. Also, this is the sort of thing the next RfC is for (after we get the basic guidelines for mass creation settled). — Rhododendrites talk \\ 13:09, 5 October 2022 (UTC)[reply]

Comments (proposal name)

  1. Comment -- I'm generally supportive of this criterion, but I think that a general blanket assertion of notability (either GNG or a subject specific notability, guideline such as WP:PROF) should suffice to avoid speedy deletion. For instance, (from my WP:PROF guideline work) if someone has access to an out-of-copyright set of biols of 19th century American university presidents and makes stub articles for those without articles on the basis of WP:PROF's presidents of respected universities are almost certainly notable guideline, and asserts all the articles from the same source, I'd be in favor of keeping (or at least not speedy deleting). But if someone uses their own university phonebook to create articles for every professor there, I'd think speedy is warranted. It's all about "What's the minimum criteria for citation in the source for the articles?" that we should consult before mass deleting; systemic bias and all. -- Michael Scott Asato Cuthbert (talk) 09:41, 8 October 2022 (UTC)[reply]
Comments from Editor X

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 8: Delete ranking of editors by created articles

Delete the list of editors by numbers of articles created, Wikipedia:List of Wikipedians by article count.

Support (delete ranking)

  1. Support as nominator. One motivation for bulk article creation appears to be gaining perceived prestige as an editor. Much of this is difficult to combat, but simply deleting the ranking is one simple step that might help to change the culture without obvious negative effects. I note (which I had not realised) that in April 2021 the order of the top 100 editors was randomised in an effort to discourage mass stub creation. Espresso Addict (talk) 05:58, 5 October 2022 (UTC)[reply]
  2. This is a bold idea, and I find myself drawn to it. If needed, we already have a marker of an experienced article creator, the autopatrolled right. The list stems from a nice bit of code so I wouldn't want the information to be inaccessible, but certainly it doesn't need to be so prominent. CMD (talk) 08:29, 5 October 2022 (UTC)[reply]
  3. Reading other the many argumentative threads around mass creation this is a thought I've previously had. It seems a ridiculous measure, but the fact that action has previously been taken about it, as noted by Espresso Addict, hints towards this being necessary. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 13:09, 5 October 2022 (UTC)[reply]
  4. Creating new articles should be for the sake of the project, not for gaining plaudits. Scolaire (talk) 15:38, 5 October 2022 (UTC)[reply]
  5. Support for many reasons. One of them is the one brought forward by Espresso addict.Paradise Chronicle (talk) 16:07, 5 October 2022 (UTC)[reply]
  6. We already scramble the top 100, and while there are plenty of excellent and valuable editors in that group, there are also a number of editors who are banned for various reasons, including a few whose contributions were part of why we're having this RfC in the first place. Deleting the page isn't going to solve the problem on its own, but it seems worth doing. TheCatalyst31 Reaction•Creation 03:26, 6 October 2022 (UTC)[reply]
  7. Support - The ranking only encourages editors who are here to get a "high score" rather than build a quality encyclopedia, as shown by the number of Top 10 creators who have been banned and left a huge pile of stubs for others to sort through. We're better off without it. –dlthewave ☎ 03:56, 6 October 2022 (UTC)[reply]
  8. The top 100 is already scrambled as it is, making the page not even effective at its current job. Thus its only purpose right now is to send the wrong message about what Wikipedia is about. —pythoncoder (talk | contribs) 04:05, 6 October 2022 (UTC)[reply]
  9. I don't think this has any direct relevance to mass-creation, but I would go further and remove any ranking of editors (whether by number of articles created, edits made, DYKs, or anything else) that could in any way be thought to be "official". Such rankings make editors concentrate on quantity rather than quality. We recently had an editor whose reply to any concern about his editing was simply to point to the number of articles he had taken to GA. That sort of thing should be stopped. Phil Bridger (talk) 08:13, 6 October 2022 (UTC)[reply]
  10. Support — In academia, emphasis on citation count has led to a publish or perish culture, where researchers are encouraged to publish as many ground-breaking articles as possible, without regard to quality. This process has resulted in many cases of scientific misconduct and other questionable research practices, such as plagiarism and data fabrication, in an attempt to cut corners where possible. One of the most extreme cases is that of Yoshitaka Fujii, a Japanese anesthesiologist who was found in 2012 to have fabricated data in at least 183 papers.
    Any quantity-based status is prone to similar abuses. Several users near the top of the closely related Wikipedia:List_of_Wikipedians_by_number_of_edits, such as Rich Farmbrough, Koavf, and Lugnuts have specific sanctions targeting mass edits, such as unnecessary cosmetic changes and low-quality stubs. That does not mean that this other list should be deleted, since there are large-scale editing patterns with little risk of error due to their systematic nature — for example, the top-ranking user, Ser Amantio di Nicolao, is mostly involved in very large-scale category changes and talk page creeations using AWB or Cat-a-lot. However, mass creation of articles requires more careful discretion, such as ensuring that the database used is reliable and can be used to support WP:GNG; and is much more damaging to Wikipedia than say, mass talk-page creation, when it goes wrong, because we often have to scrutinize a large fraction of those articles which don't have obvious problems. –LaundryPizza03 (dc̄) 09:40, 11 October 2022 (UTC)[reply]
  11. Support Stat-padding, as it is known in other places on the Interwebz, is the bane of online gaming communities. By removing ranking it may well demotivate those who mass produce low quality articles, while hopefully those with the encyclopedias best interests at heart will pay it no need. Anyone can put the their stats up on their homepage if they feel the need. LessHeard vanU (talk) 07:32, 12 October 2022 (UTC)[reply]
  12. Support I am of the belief this may help and can't hurt. The goal should be quality over quantity, and this is a metric that inverts this goal Connor Long (talk) 06:19, 15 October 2022 (UTC)[reply]
  13. It can't hurt. Quantity of edits isn't something we should be highlighting versus quality. Der Wohltemperierte Fuchs talk 18:34, 24 October 2022 (UTC)[reply]
  14. There should be fewer competitive activities in Wikipedia in general. Ranking by number of articles rewards quantity at the expense of quality. Robert McClenon (talk) 05:36, 27 October 2022 (UTC)[reply]

Oppose (delete ranking)

  1. I don't disagree that there's the potential for harm, but it's already randomized to the point of being pretty useless and anyone can get an unrandomized list through a quick Quarry query (and widely publicize that query) exactly like those at the top of this page. Maybe having it in projectspace makes it look a little too official/endorsed/valued, in which case a move might be in order. Meh. — Rhododendrites talk \\ 13:08, 5 October 2022 (UTC)[reply]
  2. I don't think that the real cause of mass-creation is the fact that the list exists. Such numbers are also available on XTools and WikiScan, so if the point is to censor the amount of articles one has created so as to not allow someone to be overly prideful, then this is not really an effective implementation anyway. — Red-tailed hawk (nest) 16:25, 5 October 2022 (UTC)[reply]
  3. Pointless proposal. I'm an admin on pi:, and found several bot created articles, with a single word or sentence and nothing else. There is no bot generated report there, why did the creator do it? Furthermore, even though I'm a complete novice in coding, I was easily able to create a list of editors by move count at Quarry in about an hour's time. I could do it just as easily for article creation count. Someone who really wants to know their article creation count can find it directly from databases. I don't think, this proposal is going to change a single editor's mind let alone solve the problem of mass creation. —CX Zoom[he/him] (let's talk • {C•X}) 13:11, 6 October 2022 (UTC)[reply]
  4. No point, as that list would not be a motivation for creating substubs rather than useful pages. Graeme Bartlett (talk) 04:14, 7 October 2022 (UTC)[reply]
  5. I think rankings are silly and can incentivise unhelpful behaviour, but I'm not sure deleting this one would be a net benefit. First, because the list can be useful: for the overall statistics or as a maintenance list (if someone is very high up there, then their creations will most likely need scrutiny). Second, I don't think this particular list has a great influence on editor behaviour: my impression has been that, to a larger extent than with other rankings, people have tended to ignore the report and just kept their own manual counts on their user page. – Uanfala (talk) 13:00, 7 October 2022 (UTC)[reply]
  6. We have a process for deleting articles and it does not include 'arbitrary decisions taken on subpages of subpages of pages completely unrelated to the page in question'. If there's a policy-based reason to delete Wikipedia:List of Wikipedians by article count, it should be nominated at MfD. At best, a consensus here should be taken as a reason to have that MfD. Nobody's even bothered to notify the page's creator or watchers that this discussion is happening. – Joe (talk) 13:24, 7 October 2022 (UTC)[reply]
  7. Seems bizarre to me to remove a list of created articles. If there are issues with people mass producing articles, then this list would be helpful to know who's making the most and fix the issue. Without a list - there are no ideas. Lee Vilenski (talk • contribs) 14:01, 7 October 2022 (UTC)[reply]
  8. Disclaimer: list creator. Thank you Joe for the notification. There is no evidence this page is contributing to mass stub creation. There was evidence at one time involving a single user, who has since been permanently banned. That incident led to the top 100 scramble, which appears to be effective. Many users track their contributions this way, who are not bad actors. I don't mind moving it out of project space if so decided. And if that doesn't work, I can move it off-site as I have a wiki that can host it. There is demand for this list by many who are not bad actors. Some of the delete votes appear to dislike the egotistic nature of these lists which is fine they are not for everyone but these lists can also motivate and reward legitimate and positive behavior. I also don't think it's a good idea in principle to give up control of the list to offsite actors. -- GreenC 14:39, 7 October 2022 (UTC)[reply]
  9. Unlikely to make a difference, particularly now that the top 100 are randomized. (I also agree with the sentiment that RfCs are not for things covered by the deletion process, although that's an auxiliary issue.) Extraordinary Writ (talk) 17:49, 7 October 2022 (UTC)[reply]
  10. Don't see a need for this. Rlendog (talk) 18:01, 7 October 2022 (UTC)[reply]
  11. The rankings tool is a helpful way to find the most prolific article creators, whose work we should watch.—S Marshall T/C 18:56, 7 October 2022 (UTC)[reply]
  12. It seems pointless to remove something that would clearly just move to some other server and still get used in the same sort of way Blue Square Thing (talk) 08:40, 8 October 2022 (UTC)[reply]
  13. A case of "it ain't broke, so don't fix it". BoJó | talk UTC 10:06, 8 October 2022 (UTC)[reply]
  14. Oppose, that page was created to reduce the load on Wikiscan and doesn't cause people to go cray-cray with the stub-creation-ay. Abductive (reasoning) 18:33, 9 October 2022 (UTC)[reply]
  15. Oppose - Not needed. --Jax 0677 (talk) 23:33, 9 October 2022 (UTC)[reply]
  16. I highly doubt anybody is creating loads of articles in order to rank more highly on that rather obscure list. Hut 8.5 11:45, 11 October 2022 (UTC)[reply]
  17. Oppose as not necessary given the scrambling. Hut 8.5 - I know that a number of mass-creators were regular visitors to the talk-page at that page and obviously checked it regularly, so even applying AGF, not a stretch to say they were at least interested in who was ranking highly on it. FOARP (talk) 14:27, 11 October 2022 (UTC)[reply]
  18. No, its useful and interesting for other editors who may need it for research purposes. Crouch, Swale (talk) 20:29, 11 October 2022 (UTC)[reply]
  19. Oppose No need for this. Although perhaps you would need a seperate list. Once the process is stable, effectively there is no reason why your AI engine couldn't create 40k articles, for example on the worlds bays. They editor would shoot right to the top of the rankings, so perhaps a seperate list or some indication they are mass articles creators so to differentiate them, in a simple manner. scope_creepTalk 16:57, 14 October 2022 (UTC)[reply]
  20. Now that it's randomized, this specific page is not an impetus for mass creation. Editors count articles themselves (see, for example, Lugnuts's user page). If there were more evidence that people were generally making hundreds of low-quality articles simply to climb this specific ladder, I'd support. Ovinus (talk) 01:06, 15 October 2022 (UTC)[reply]
  21. Deleting this would add burden to shared resources like quarry, slowing things down for everyone due to interested editors repeatedly running queries. Having it in a centralised page as now is a better use of resources. ಮಲ್ನಾಡಾಚ್ ಕೊಂಕ್ಣೊ (talk) 07:42, 16 October 2022 (UTC)[reply]
  22. Why? Who cares?--Ortizesp (talk) 14:47, 20 October 2022 (UTC)[reply]
  23. Oppose: I'm sure some editors create perverse incentives for themselves using the list, but the same can be done without it. I don't believe this list is a significant contributor to the phenomenon of mass creation. The list parallels the edit count list, which I would view as similarly harmful (perhaps moreso), but I still hold the view of an infobox I made in 2014. — Bilorv (talk) 13:23, 21 October 2022 (UTC)[reply]
  24. Oppose and get rid of the scrambling as well. Users should know at this point that having created a ton of stubs isn't particularly prestigious and we don't need to protect people from themselves like this. This is an interesting statistic and it should be easily available. Elli (talk | contribs) 18:50, 24 October 2022 (UTC)[reply]
  25. Oppose if people are this concerned about their position on this list, they would just a Quarry query or other method to find out their exact number. Dreamy Jazz talk to me | my contributions 23:27, 27 October 2022 (UTC)[reply]
  26. Meh. --Jayron32 18:34, 28 October 2022 (UTC)[reply]
  27. I'm going to copy User:S Marshall's reasoning verbatim: The rankings tool is a helpful way to find the most prolific article creators, whose work we should watch. WaggersTALK 11:52, 1 November 2022 (UTC)[reply]
  28. Oppose as unlikely to have any measurable effect until someone measures the effect. · · · Peter Southwood (talk): 18:47, 1 November 2022 (UTC)[reply]
  29. If someone really wants this, send it to MFD. Meanwhile someone could just create a tool to do it anyway. --Rschen7754 01:29, 2 November 2022 (UTC)[reply]

Comments (delete ranking)

Comments from XOR'easter (delete ranking)

I kinda feel that anyone who really gets their jollies from racking up their stub creation count would get much the same feeling from, e.g., listing their created stubs on their User page. Meh. I'm having a hard time seeing that there would really be a benefit or a loss for eliminating this list. XOR'easter (talk) 16:32, 5 October 2022 (UTC)[reply]

Comments from Espresso Addict (Q8)

XOR'easter: I'm sure most editors get satisfaction from listing their creations in their user space but that does not make them readily findable (especially by new editors) nor easily comparable to others, nor does it send a meta-message that one's value as an editor can be measured in how many articles one creates. It also has the benefit of reminding the editor that they started that article, and might want to come back to update or improve it, which the ranking does not. I don't think the effect will be large but I think it might be a small positive step. Espresso Addict (talk) 19:00, 5 October 2022 (UTC)[reply]

Rhododendrites: I think most editors don't know how to use Quarry or similar. Having an "official" ranking available in project space and linked from various help guides seems to give it too great a prominence. Espresso Addict (talk) 19:10, 5 October 2022 (UTC)[reply]

Comments from Editor Paradise Chronicle (Q8)

Going with the logic that the creators list encourages the creation of stubs, an expanders list could be thought of. I know the GA and FA list exists, but one for raising stubs to start articles or adding several sections to an article would help as well.Paradise Chronicle (talk) 04:52, 7 October 2022 (UTC)[reply]

@Lee Vilenski it is no about removing the list of created articles, but about removing the ranking Wikipedia:List of Wikipedians by article count. Maybe you wanted to vote in another discussion where this is discussed? And I'd be interested where this I discussed, I would also oppose the removal of a list of created articles.Paradise Chronicle (talk) 10:54, 8 October 2022 (UTC)[reply]

Comments from FOARP (Q8)

I hate to be go WP:BURO, but this is really a discussion to have at WP:MFD and not here. FWIW I have plenty of concerns about this page which I've discussed a number of times on the talk page there, but I don't get how the scrambling hasn't already addressed this.FOARP (talk) 08:47, 7 October 2022 (UTC)[reply]

Comments from (Your Name)

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 9:Should mass-creators be forced to respond to issues?

If there is a discussion on the mass-creation noticeboard, should the mass-creator be required to address the concerns on that noticeboard if the mass-creator is editing elsewhere?

Support (respond to issues)

  1. Support. People are responsible for their edits. It is much easier to spot issues if the author informs you about what and why they did it.(Communication is required). And mass-creation has the potential for huge damage to the encyclopedia.Lurking shadow (talk) 10:10, 5 October 2022 (UTC)[reply]
  2. Support in principle, although some refinement of what it means to "respond" and what happens if you don't require elaboration - I don't think we can literally force people to respond, but we can establish that a failure to respond means that editors can make negative inferences about their intent and process, and that this is sufficient to justify sanctions preventing them from mass-creating articles in the future. This is necessary because of the danger of WP:FAIT - there is too much risk that someone could mass-create a bunch of articles and then try to stall out the process by refusing to engage, hoping that the difficulty of reversing their action leads to it remaining in place. --Aquillion (talk) 15:03, 5 October 2022 (UTC)[reply]

Oppose (respond to issues)

  1. Oppose - Our current noticeboards work just fine without this requirement. If an editor doesn't show up to explain/defend themselves, we make decisions based on the information that we have, and there's no need to make a written rule that failure to comment may be viewed negatively. –dlthewave ☎ 21:37, 5 October 2022 (UTC)[reply]
  2. Per WP:NOTMANDATORY. We cannot compel ordinary editors to contribute to a particular noticeboard or thread; the only people that we expect this of are holders of advanced permissions, who are generally required to maintain the trust and confidence of the community. On the other hand, the decision to actively edit and not respond to an WP:ANI thread can at times indicate that a user has little or no interest in working collaboratively, and we should not prohibit the community from analyzing a user's conduct along those lines. — Red-tailed hawk (nest) 22:11, 5 October 2022 (UTC)[reply]
  3. The noticeboard whose existence is implied in this question ideally shouldn't exist. That said, if it is actually created against the concerns voiced at #Oppose_(Create_noticeboard), there's no need for a specific ADMINACCT-like requirement to be codified for that one noticeboard. Users are already expected to respond to community concerns about their editing (WP:DISRUPTSIGNS: "Does not engage in consensus building", "Rejects or ignores community input"), and WP:NOTCOMPULSORY limits this RfC's ability to go beyond this. ~ ToBeFree (talk) 22:13, 5 October 2022 (UTC)[reply]
  4. Per WP:NOTCOMPULSORY. We cannot force editors to make edits. James500 (talk) 01:35, 6 October 2022 (UTC)[reply]
  5. Oppose If they don't show up, they'll probably show up if the decision taken was perceived negatively by them. Also, mass creators are not new editors, they are rather experienced editors and know how noticeboards and talk-page notifications function.Paradise Chronicle (talk) 09:39, 6 October 2022 (UTC)[reply]
  6. If they don't respond, that is not necessarily a serious problem. But if they continue with causing a problem with article creation, they can then be blocked, perhaps after AN/I discussion. There could be use of a noticeboard, or the user's talk page. Graeme Bartlett (talk) 04:12, 7 October 2022 (UTC)[reply]
  7. Oppose - We cannot force behaviour and should not try. The one thing I would request is that mass-creators at least do not battle to prevent clean-up - BTW this is typically the reason that mass-creators get banned from Wikipedia: not because of their mass-creation per se, but because they then get very angry when others start questioning the value of their X-thousand articles on Y-subject created manually from Z-database and engage in uncivil and disruptive behaviour against the people doing the clean-up. This then results in months of unpleasantness for the people doing the clean-up ending at ARBCOM (because ANI simply isn't up to the task of banning these typically long-tenured and influential editors). It is striking that only one of the top-ten article-creators (Ser Amantio di Nicolao) is still an active editor today, with all the rest being indeffed (3) retired under a cloud (3) semi-retired/just not very active (3), and that that one remaining editor has admirably stayed out of attempts to clean-up their articles (particularly the mass-created ones about non-notable geographical features in Antartica based solely on the GNIS database created back in 2008 when standards were more relaxed).FOARP (talk) 09:20, 7 October 2022 (UTC)[reply]
  8. Better dealt with at ANI and through a voluntary approach via sanction negotiation etc... Blue Square Thing (talk) 08:41, 8 October 2022 (UTC)[reply]
  9. You can't FORCE people to do anything. If they won't respond, WP:ENGAGE applies. BoJó | talk UTC 10:09, 8 October 2022 (UTC)[reply]
  10. Oppose - Oppose per WP:NOTMANDATORY. --Jax 0677 (talk) 23:36, 9 October 2022 (UTC)[reply]
  11. Oppose Being involved in a scandal is just humiliating — the only involvement one should have in cleaning up their act is to make defending statements to the public or in the courtroom. And we don't want to shame people to the point of a catastrophic mental health breakdown, or else we risk getting sued in real life. Carlossuarez46 quit after the mass-creation scandal about Iranian abadis broke out, declining even to comment at the resulting Arbcom case. –LaundryPizza03 (dc̄) 09:48, 11 October 2022 (UTC)[reply]
  12. Exactly the same rules as anywhere else. mass-created articles do not need to be treated differently. Supertrinko (talk) 02:14, 13 October 2022 (UTC)[reply]
  13. Oppose Completely unworkable. We should not be relying on the false hopes that an editor can mass create garbage and then we can just make them fix it. We have to find avenues for resolving this situation without them, since many I'm sure will not be cooperative or may simply stop mass creating and move on to better pastures (and no longer doing anything bad is generally acceptable enough on Wikipedia to not be blocked). -Indy beetle (talk) 08:38, 13 October 2022 (UTC)[reply]
  14. Truly problematic and disruptive cases of this (i.e., editors who keep on mass creating after been apprised of others' concerns) are already covered under WP:IDHT et al. Ovinus (talk) 00:47, 15 October 2022 (UTC)[reply]
  15. This isn't a job. --Rschen7754 02:01, 15 October 2022 (UTC)[reply]
  16. Oppose, out of scope for our roles as editors.--Ortizesp (talk) 14:49, 20 October 2022 (UTC)[reply]
  17. Unworkable, and redundant. 1) We already require people to communicate with others, so I don't see why we are singling out people who work harder than the average Wikipedian. 2) We shouldn't assume that just because someone creates a lot of content, they are automatically under suspicion. --Jayron32 18:36, 28 October 2022 (UTC)[reply]
  18. Per everyone else here WaggersTALK 11:54, 1 November 2022 (UTC)[reply]
  19. We literally cannot force anyone to respond to communication. If they do not respond, however, they can be blocked for disruptive behaviour, and there are good arguments that creating large numbers of unencyclopedic articles is disruptive. Blocking from article creation only would also be an appropriate response. · · · Peter Southwood (talk): 18:39, 1 November 2022 (UTC)[reply]

Comments (respond to issues)

Comments from BeanieFan11 (Q9)

You can't force someone to comment on a page. BeanieFan11 (talk) 18:20, 5 October 2022 (UTC)[reply]

Comments from ActivelyDisinterested (Q9)

You can't force editors to do something, but if they continue with something that is considered distruptive then there is WP:ANI. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 17:52, 7 October 2022 (UTC)[reply]

Comments from Editor X (respond to issues)

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 10: add mass-creation as a reason to WP:BUNDLE

Next RfC; + to WT:ADAS

Should we add mass-creation of articles by the same editor using substantially the same sources and format as a reason for bundling multiple articles into a single AFD at WP:BUNDLE?

Support adding mass-creation as a reason to WP:BUNDLE

  1. Support as proposer - This would make it easier to discuss multiple articles in a single AFD, speeding up clean up of these mass-created articles. At present you would need to show that the articles are actually hoax/spam articles, or have similar titles, or are about a range of manufactured products (which is a sub-set of mass-creation) whereas they may simply be multiple notability-fails. The AFD could still be unbundled if other editors think it is a train-wreck.FOARP (talk) 12:49, 5 October 2022 (UTC)[reply]

Oppose adding mass-creation as a reason to WP:BUNDLE

  1. Isn't this exactly what the next RfC is supposed to be about? The whole point of having this one first, if I understand, is to have a working definition of "mass creation" (and some clearer guidelines in place) before talking about what to do with articles created in that manner. I certainly wouldn't support adding this without having a clear definition to work from. — Rhododendrites talk \\ 13:02, 5 October 2022 (UTC)[reply]
  2. Exactly per Rhododendrites, but see also my ideas at Wikipedia:Mass action review that I hope to workshop into shape as part of the next RfC. Thryduulf (talk) 13:10, 5 October 2022 (UTC)[reply]

Comments on adding mass-creation as a reason to WP:BUNDLE

Comments from ActivelyDisinterested (Q10)

I support this, but as other have said this is probably best discussed in the AfD part of the RFCs. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 13:21, 5 October 2022 (UTC)[reply]

Comments from Wjemather

This is an appropriate proposal for the next RFC, i.e. "Article deletion at scale", but not this one. wjematherplease leave a message... 13:48, 5 October 2022 (UTC)[reply]

Comments from Editor X

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 11: Should we allow (or disallow) mass-created articles that lack any sourced claims of importance?

This proposal establishes a principle that all mass-created articles must contain a reliably sourced claim of importance. This would apply to any mass-created article that does not indicate why its subject is important or significant. This is a lower standard than notability. This proposal does not imply that all mass creations of articles that contain reliably sourced claims of importance are acceptable; it merely identifies that mass creations that do not adhere to this principle are generally not acceptable. This is non-exclusive with any option from Question 2; should there be a rough affirmative consensus in favor of both this proposal and a proposal in Question 2, both shall take effect.

Support (sourced claims of importance)

  1. This establishes a bare minimum for mass-created articles, which is that they contain a source that verifies their claim of significance. This is non-exclusive with any particular option in Question 2 because this does not make any comments regarding sources that would necessarily contribute to GNG. This principle is eminently reasonable and more or less any mass creation that fails this would probably require a good bit of legwork to clean up if not resulting in much of the mass creation eventually being deleted. Failure to follow this principle could either be enforced as a simple disruptive editing case or, should the community decide in another RfC or question, through some modification to WP:DEL-REASON and/or the creation of a CSD criterion. — Red-tailed hawk (nest) 22:00, 5 October 2022 (UTC)[reply]
  2. I note that the heading here is ambiguous. But a claim of importance should be required, and the information should be sourced. A claim of importance may include SNG evidence. Enough good references also show GNG and importance. If no claim of importance then speedy delete criterion A7 may apply, and if its out of scope then claims will avert being an obvious AFD candidate. — Preceding unsigned comment added by Graeme Bartlett (talk • contribs) 04:19, 7 October 2022 (UTC)[reply]
  3. Support. We need to have some sort of bare-minimum standard for mass-created articles or discussions over them are just going to spin wheels endlessly. And this is a reasonable extension of WP:A7, which can be tweaked to be more in-line with it if necessary. --Aquillion (talk) 18:21, 7 October 2022 (UTC)[reply]
  4. Articles in general should have a cited claim to importance, not just mass-created ones. Reywas92Talk 17:21, 12 October 2022 (UTC)[reply]
  5. Never understood why the community limited WP:A7 to only certain types of articles. It should apply to all articles. This is a step in the right direction. Levivich (talk) 20:08, 17 October 2022 (UTC)[reply]
  6. I agree with Levivich that A7 should apply to all articles, though A7 is less strict as it does not require a source. But if an article does not contain a reliably sourced claim of importance ("importance" meaning what it does in A7) then it is not useful to the encyclopedia, at least not in its presented state. — Bilorv (talk) 13:30, 21 October 2022 (UTC)[reply]

Oppose (sourced claims of importance)

  1. Without any definition of what constitutes a "sourced claim of importance" that is objective and applicable to all subjects this is just a recipe for arguments. Thryduulf (talk) 15:13, 6 October 2022 (UTC)[reply]
  2. "Sourced claim of importance" is one of those things that seems unambiguous but actually at the coalface is rather hard to agree on. Is being a populated place a claim of importance? A listed building? An athlete who competed at the Olympics? Or was in their national top 10? An academic with two papers with >100 citations? Also database sourcing is often (though not always) reasonably reliable, it's the "significant coverage" that is lacking. Espresso Addict (talk) 23:54, 6 October 2022 (UTC)[reply]
  3. The opposite of what is required. Sets the bar way too low and potentially re-opens the door to mass-creation of (for example) sports biography stubs based on a single appearance somewhere, sourced only to a database, that was closed in WP:NSPORTS2022. wjematherplease leave a message... 12:40, 7 October 2022 (UTC)[reply]
  4. This is attempting to apply the standard of WP:CSD#A7 to mass article creations, with the added requirement of a source for the claim. A7 has been restricted so that it's usually obvious whether an article has a claim of importance or not, and attempts to expand it usually fail because it wouldn't necessarily be clear what qualifies. However any admin is likely to give the benefit of the doubt to the article creator, and certainly any article which indicates the subject might meet an SNG or which includes some non-terrible references is unlikely to be deleted under this criterion, so it probably wouldn't make much difference anyway. Hut 8.5 16:40, 7 October 2022 (UTC)[reply]
  5. Oppose. What is a "claim of importance"? If it meets GNG it's fine. As per #2, I think if it has one decent source at creation, it should be considered to meet minimum standards for creation. Rlendog (talk) 18:03, 7 October 2022 (UTC)[reply]
  6. Importance is like significance – a value judgement – and different people have different values. For example, I would suppose that being included in a national register of historic buildings is implicitly a claim of importance but some seem to expect something more. As another example, consider a forthcoming FA: Daglish railway station. There doesn't seem to be a clear claim of importance for this – it all depends whether you think ordinary railway stations are important or not. And railway stations are the sort of topic that tend to be created systematically, en masse. We have lots of articles about them, such as this FA, and so they seem to be generally acceptable. We should be going by such precedents rather than requiring them to be proven afresh, every time, in a subjective way. Andrew🐉(talk) 18:24, 7 October 2022 (UTC)[reply]
  7. I assume this is about CSD's concept of Wikipedia:Credible claim of significance, which explicitly does not apply to many subjects (e.g., animal species). It's nice when a Wikipedia article gives me a reason to care about the subject, but subjects can be notable, and articles about them can be useful, even if I don't care. If I'm looking up the subject, I already know why I want to know what it is; I don't need the article to tell me nearly as much as I need the article to have a decent first sentence. WhatamIdoing (talk) 04:55, 8 October 2022 (UTC)[reply]
  8. Who judges the claim of importance? I've seen this used to defend articles about obviously non-notable subjects before. Not, in my view, the way to go Blue Square Thing (talk) 08:44, 8 October 2022 (UTC)[reply]
  9. We already have notability. Something important to one person is unimportant to another. BoJó | talk UTC 10:13, 8 October 2022 (UTC)[reply]
  10. Oppose Conflicts with WP:STUB and WP:Notability, which are well established guidelines. LessHeard vanU (talk) 07:34, 12 October 2022 (UTC)[reply]
  11. I'm satisfied that the existing notability guidelines are sufficient. Supertrinko (talk) 02:16, 13 October 2022 (UTC)[reply]
  12. "Sourced claim of importance" is vague. Existing WP:Notability is enough. ಮಲ್ನಾಡಾಚ್ ಕೊಂಕ್ಣೊ (talk) 07:50, 16 October 2022 (UTC)[reply]
  13. "Importance" is too subjective. --Rschen7754 04:05, 17 October 2022 (UTC)[reply]
  14. Oppose. What policy requires a topic to be important? Important to whom? Notability is difficult enough to manage, and it is much more objective than importance.· · · Peter Southwood (talk): 18:30, 1 November 2022 (UTC)[reply]

Comments (sourced claims of importance)

Comments from BeanieFan11

How do we determine what counts as a "claim of importance"? Would playing in the NFL count? What about participating in the Olympics or playing MLB? Or does it have to be something much higher to count as "important"? BeanieFan11 (talk) 14:13, 6 October 2022 (UTC)[reply]

Comments from Thryduulf (Q11)

A7 is subject-limited in part because it is often not possible for someone unfamiliar with a topic to reliably judge what is and isn't a claim of significance or importance and many of the SNGs are about giving guidance about this, but disagreement and arguments about interpretation of SNGs is one of the reasons we are here so this is likely to be both duplicative and unproductive. Thryduulf (talk) 15:20, 6 October 2022 (UTC)[reply]

Comments from Wjemather (sourced claims of importance)

@BeanieFan11: My understanding would be that (for example) participating in the Olympics is a credible claim of importance even if it is already well established that it is not a credible claim of notability. wjematherplease leave a message... 12:46, 7 October 2022 (UTC)[reply]

Comments from ActivelyDisinterested (Q11)

Claims of importance or notability just make the issue to clouded. We need a proposal that mass created articles must have at least one, non-database, reference that is not from an unreliable source. This should be the requirement for any stub article, but that is a different discussion. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 17:57, 7 October 2022 (UTC)[reply]

Comments from Robert McClenon

The requirement for a sourced claim of importance is more restrictive than A7 because A7 does not require that the credible claim of significance be verifiable. If the credible claim of significance is not verified, that is AFD, not A7. Robert McClenon (talk) 05:40, 27 October 2022 (UTC)[reply]

Comments from Editor X

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 12: Editors who mass-create stubs have a duty to expand the articles later

It's okay, and sometimes beneficial, to create a lot of stubs, but it's not okay to leave them languishing in the stub state for years on end. Other editors are unlikely to be interested in expanding them past the stub state; therefore, editors who mass create stubs are expected to expand those articles within some reasonably generous timespan. If they don't, we can require them to stop creating stubs until they have expanded most of the stubs they have previously created. 05:22, 8 October 2022 (UTC)

Support (duty to expand)

  1. support statement

Oppose (duty to expand)

  1. There's no way to enforce this. Stubs as seeds for more developed articles are, to me, a reasonable way forward so long as there's actually some way of developing them. And they're more likely to be developed if they exist as stubs than if they don't. Blue Square Thing (talk) 08:46, 8 October 2022 (UTC)[reply]
  2. The idea that articles which remain stubs are unacceptable is just wrong and not consistent with the view of the community. Just over half of our articles are stubs, and most of them have remained that way for years. A proposal to delete half of the articles on Wikipedia would be dead on arrival. Hut 8.5 09:05, 8 October 2022 (UTC)[reply]
  3. Nothing at all wrong with any stub that has a reliable source. Why pick on stubs when there are volumes of "start-class" articles that have been "expanded" by conversion of a useful stub into an incoherent mess? BoJó | talk UTC 10:22, 8 October 2022 (UTC)[reply]
  4. Its a volunteers project.Paradise Chronicle (talk) 10:39, 8 October 2022 (UTC)[reply]
  5. This can't be enforced, and is a more complicated version of requiring a bit more substantiveness at the point of creation, which is much easier to assess. CMD (talk) 04:26, 9 October 2022 (UTC)[reply]
  6. Oppose generally. There can be exceptions such as proving notability and ensuring compliance with copyright laws and WP:BLP but not for everything. Emir of Wikipedia (talk) 20:32, 9 October 2022 (UTC)[reply]
  7. Oppose - Oppose per WP:TOOSHORT. --Jax 0677 (talk) 23:37, 9 October 2022 (UTC)[reply]
  8. No as long as at the time of creation they are acceptable. Crouch, Swale (talk) 20:30, 11 October 2022 (UTC)[reply]
  9. You can't require someone to edit. We're all volunteers here. Steven Walling • talk 21:07, 11 October 2022 (UTC)[reply]
  10. Reluctatnt oppose, because this focus' upon the article and not the creator - who is the problem. We have criteria for the speedy deletion of poor quality content, while struggle to deal with problematic editors. LessHeard vanU (talk) 07:42, 12 October 2022 (UTC)[reply]
  11. Strongly oppose this, it will only discourage editors from contributing. There's no rush to have a stub improved, and it can be done at anyone's discretion. Supertrinko (talk) 02:17, 13 October 2022 (UTC)[reply]
  12. Oppose because its meaningless. Nobody is going to update mass-created articles. The numbers will be too big. Questions like who is going to validate them are perhaps more important, when you have 200k mountains and hills of the Africa articles, for example. Who validates these to prove they are good? The whole idea about stubs is a busted flush, its all last decade thinking. AI might be able to update a lot of them to a basic standard as certainly nobody is really interested on working on the 10,000's of stubs we have at the moment. I don't think we have the people for that. Many of them are never seen. Google and Microsoft can already mechanically create stubs of better quality than many of the stubs that are created on here. So the question is moot. AI is going to make that look like an amoeba at work. There should be some kind of minumum quality, specifying what should be in the article. Domain specific knowledge for that article series. Even a soft framework, where you suggested what should be in them, would be useful. Your not aiming for stubs with automated tools. You want beautiful, visually exciting articles of a sufficient length and detail. Think of the something of the Book of Kells, done at scale with lots of diagrams, animations, maps and lots and lots of visual details. scope_creepTalk 18:57, 14 October 2022 (UTC)[reply]
  13. Per scope creep. >95% of Wikipedia stubs will remain that way forever, and that's fine. Ovinus (talk) 00:45, 15 October 2022 (UTC)[reply]
  14. This isn't a job. --Rschen7754 02:00, 15 October 2022 (UTC)[reply]
  15. Oppose from my own experience with making a bunch of stubs about lichen I can tell you that there isn't a ton of stuff out there so a stub is about as long as it's gonna get. Dr vulpes (💬 • 📝) 04:59, 16 October 2022 (UTC)[reply]
  16. This is contrary to WP:VOLUNTEER. Different editors have different skills, forcing someone to do something they don't like doing with is not a good use of voluteer time. ಮಲ್ನಾಡಾಚ್ ಕೊಂಕ್ಣೊ (talk) 07:58, 16 October 2022 (UTC)[reply]
  17. Oppose as unenforceable, therefore pointless. · · · Peter Southwood (talk): 18:23, 1 November 2022 (UTC)[reply]
  18. Oppose Wikipedia is a volunteer organization, no one has a duty to do anything. --Jayron32 11:50, 2 November 2022 (UTC)[reply]

Comments (duty to expand)

Comments from WhatamIdoing (duty to expand)

This proposal is based on some comments from multiple editors in previous discussions. Note that 3.7 million out of our 6.5 million articles are rated as stubs. I think that argues against this idea, but others may think it argues for it. WhatamIdoing (talk) 05:22, 8 October 2022 (UTC)[reply]

@Scope creep, please look at Familial Amyloidosis, Finnish Type, Acrospiroma, Diffuse neonatal hemangiomatosis, and Cantú syndrome. All of them started off as single-sentence, single-source substubs during a mass-creation campaign. It is not actually true that "Nobody is going to update mass-created articles". WhatamIdoing (talk) 21:51, 30 October 2022 (UTC)[reply]
Scope creep: It never happens.
Me: Here are four counterexamples that I could find easily.
SC: It still never happens.
Do you expect me to agree with you on the basis of plain assertions?
I could readily agree that it doesn't happen often enough or fast enough to satisfy some editors, but could we please move away from the already-proven-wrong claims that it doesn't happen at all? WhatamIdoing (talk) 23:19, 30 October 2022 (UTC)[reply]
Comments from Espresso Addict (duty to expand)

I'm conflicted on this one. I wish people who had sources in their hand would expand stubs more than they do, but one can't really make an editor do anything (except go away). There's a fundamental problem that's been nagging at me that articles do, actually, have an owner, because the notification system sends notifications to the original creator. This means that if I were to create 100 "articles" of the form "XYZ is a listed building in ABC.<ref_Listing, ref_Pevsner>", I might just get away with it, but I would stellarly annoy anyone who had been planning to create one of them when they'd just amassed enough sources to write a "proper" article. On the other hand, someone might read it and think, "I pass XYZ on the way to work every morning, I'll take a quick photo" or even "I used to live near XYZ, surely the listing says more than that [click]." Espresso Addict (talk) 06:08, 8 October 2022 (UTC)[reply]

Comments from Scope_creep (duty to expand)

Hi @WhatamIdoing. I think it probably is true, to our detriment. Certainly the evidence I have is mostly anecdotal because there is no research to really determine what exactly is going on, but having looked at thousands of articles through NPP/AFC and through manual searches over the years, I know its not happening. Its a function of the volunteer diven WP. It is certainly true that there is articles that folk are working on, that slowly evolve from tiny wee things into large well-written, well-structured articles, written line by line and built up by multiple editors, but that kind of mechanism, while worthy, fails most and you hardly really see it they days. I'm suprised when I come across an article that's like that. I've looked at thousands of articles that are critically important, for example, folk who have absolutely changed the world from one day to the next but they don't get updated. It like the dark ages and its been like that for a decade, at least. Even the level 1 to 5 most important article haven't been completed yet. I'm working on the Joseph Lister article. I was asked by a Glasgow academic and a London academic to write the article when I was looking for info on another article and I contacted them for help and they both mentioned it, offhand and they we discussed and fine I'll do it. They'd waited for 15 years to see a proper article but they're was no movement on it. And that is one of the most important people of the 19th Century, saved billions from an early death. AI may help to update them, if they can be properly checked. scope_creepTalk 23:03, 30 October 2022 (UTC)[reply]

Comments from Editor X (duty to expand)

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 13:

Neutral description of proposal.

Support (proposal name)

  1. support statement

Oppose (proposal name)

  1. oppose statement

Comments (proposal name)

Comments from JoelleJay (proposal 13)

I'd like to workshop this proposal before putting it to !vote. How about a baseline question: "Should there be a per-editor limit to the number of articles created after [X date] that do not contain sufficient sourcing to meet a relevant notability guideline?"

Further details could be:

"A single editor cannot have more than [Y number] articles on GNG-governed subjects created after [X date] that did not contain [a source of SIGCOV|2 sources of SIGCOV] at the time of creation. Surpassing [Y number] will trigger escalating warnings leading to a p-block from article creation. [Y number] can be offset by adding GNG sources to articles they created, or by adding GNG sources to other articles that have never met GNG. The latter option is available to all editors regardless of the status of their own creations, and can be built up as their [Z number]."

There is no need to hammer out what is "SIGCOV" in this RfC; if we don't need a clear definition for AfC/AfD/NPP/etc. to work, we don't need one here either.

Honestly, I think we should be strongly discouraging putting articles into mainspace that don't have sources demonstrating notability, and having a limit any one editor can have of such articles would help. So I think this should be evaluated at the creation time. And while adding any sourcing to any article is great, for the purposes of restricting undersourced article accumulation we should stick to "Z credits" being earned only by bringing non-GNG-demonstrating articles to demonstrating-GNG status. 00:03, 9 October 2022 (UTC)
Comments from Espresso Addict (proposal 13)

JoelleJay:

  • "at the time of creation": suggest within 48 hours/5 days of creation; many editors don't put all their sources in in the first edit. Also need clarity re creation in mainspace vs userspace or draftspace; it's ok to create drafts there that don't comply with some mainspace policies.
  • I suspect that commenters will want more information on what [Y number] is in advance of voting. 1? 5? 10? 100?
  • "to other articles that have never met GNG": is it necessary to limit it to these? Any sourcing work is valuable.
  • There's going to be a major problem with how technically feasible this all is to determine. No bot is going to be able to determine what a GNG-compliant source is, and editors will need access to the source, which might be offline. Eg, if I add a source to one of the Pevsner guides, who's to say whether there's enough coverage therein? It could create a lot of work for experienced editors. Espresso Addict (talk) 21:55, 8 October 2022 (UTC)[reply]
Comments from WhatamIdoing (proposal 13)
If an article currently cites:

☒N no sources at all

but the real world has:

checkY hundreds of non-trivial newspaper reports, magazine articles, and books

then the subject is notable.

JoelleJay, I think this needs more work. To start with, sourcing doesn't make a subject notable, and the absence of sourcing doesn't make a subject non-notable. I proposed the box on the side here for the relevant section of WT:N, but editors were alternately concerned that it would encourage editors to be lazy about citing sources (I guess these editors think that if citing sources isn't absolutely mandatory on penalty of article deletion, then nobody will do it), or that it might be possible to have literally hundreds of non-trivial newspaper reports, magazine articles, and books written about a subject, but still somehow the subject wouldn't be notable (I'm not sure what they're thinking).

So the first thing is that you'd have to change the wording to be about demonstrating notability, rather than being notable: "Should there be a per-editor limit to the number of articles created after [X date] that do not contain sufficient sourcing to meet demonstrate compliance with a relevant notability guideline?"

Second, I think you might want to consider a much more fundamental proposal: "As of <date>, all new articles must name at least one Wikipedia:Independent source." Right now, only BLPs are required to cite any sources at all. You could move User:WhatamIdoing/Christmas candy to the mainspace, and the worst that anyone could do to it is to slap an {{unref}} tag on it. Per User:Espresso Addict, there should be a sensible grace period (WP:BLPPROD gives you seven days after being tagged, so it should be the same or longer), but if you want all new articles to identify a specific source, we actually need a rule that requires this. WhatamIdoing (talk) 23:25, 8 October 2022 (UTC)[reply]

Comments from Editor X (proposal 13)

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 14: Prohibit mass creation of (GNG) articles sourced only to non-prose-containing databases

Mass creation of articles on GNG-dependent topics cannot be based on (EDIT: "sourced only to" or "easily sourceable only to") databases that do not contain significant secondary prose coverage authored specifically on the subject.

Support (Q14: databases)

  1. Databases that do not contain original prose written specifically about the subject are comparable to specialized web browsers or news aggregators: they may be reliable, but they are not synthesizing anything specifically for any entry, they are merely tools to autocurate and present facts without further analysis. Some of those facts might be from a secondary publication, but they are also likely to be cited to a primary research article or even uploaded directly by a researcher. Scientific databases especially are likely to contain millions or billions of entries (quoting from WP:NASTRO) with thousands of attributes, but merely existing in a (non-notability-granting) database with many severable "facts" ascribed to it is not sufficient for the database to count towards GNG for a given object. For the facts that are referenced to secondary sources in the database or have secondary coverage elsewhere, those publications should be the basis of the article rather than the database. If a GNG topic can only be referenced to a database (or to other primary documents), it should not have a standalone article. See also this discussion. JoelleJay (talk) 00:18, 13 October 2022 (UTC)[reply]
  2. Sourcing only to databases is an inherent failure of GNG. Even in a category with relaxed requirements, this may leave little to create a substantive article about. Any sort of bulk creation must have approval to ensure that a mass of articles that is impossible to deal with once created is using sources that are highly reliable and will lead to expansion without undue synthesis. Reywas92Talk 02:41, 13 October 2022 (UTC)[reply]
  3. Articles sourced only to databases violate WP:NOTDATABASE, as without OR they cannot put the data in context with explanations referenced to independent sources. This means that creating articles sourced solely to them is already forbidden, but clarify this would be useful. BilledMammal (talk) 03:40, 13 October 2022 (UTC)[reply]
  4. As to me, articles should not be created for creating articles but because there is a certain interest/knowledge on them. If there is a some knowledge, they'll probably also have an other source for the article they plan to create other than a basic database. Before, creating articles was good, but maybe its time for Wikipedia to shift the focus on quality instead of quantity. To prohibit articles based on databases, would shorten many deletion discussions.Paradise Chronicle (talk) 06:54, 13 October 2022 (UTC)[reply]
  5. The early days of Wikipedia are past us, we no longer need to enable rapid expansion via database scraping. Copying databases is less helpful to the reader than letting them just use the original database in question, and we ought to be focusing on things that have SIGCOV in secondary sources. -Indy beetle (talk) 08:42, 13 October 2022 (UTC)[reply]
  6. I mean, this is 4A come again, so yes, I support. If GNG is to be used to justify an article's existence, then GNG needs to be demonstrated when it's created, meaning that a database scrape is by definition insufficient. Some SNGs would justify database scraping (NPOL, for instance) and there GNG isn't in the picture usually. Vanamonde (Talk) 15:51, 13 October 2022 (UTC)[reply]
    I would prefer "sourced only to" per ONUnicorn below, but I believe that's a wording issue that is fixable when implement given the comments here. Vanamonde (Talk) 16:24, 13 October 2022 (UTC)[reply]
  7. Support These databases are full of errors that are not validated upon use, just accepted. Sloppy practice. They are already the bane of WP and using them at scale is just asking for trouble. We should be seeking high-quality sources at every level. Personally I don't they will be needed to be used by automated tooling, as by point most archives will already be open for AI access. scope_creepTalk 18:17, 14 October 2022 (UTC)[reply]
  8. This is wise. I understand the position that "if it's notable, it's fine", but even ignoring the concerns of lowered quality standards, there is plenty of potential for database errors, that requiring additional sources would make far less likely. Ovinus (talk) 01:10, 15 October 2022 (UTC)[reply]
  9. Database entries are not evidence of notability. If it is a reliable database with good information, relevant information should be put onto a useful article related to the topic of the database rather than squirreled away to some hapless substub. CMD (talk) 03:26, 15 October 2022 (UTC)[reply]
  10. per WP:NOT. One article sourced only to a non-prose-containing databases is a problem; mass-creation of such articles even moreso. Levivich (talk) 20:06, 17 October 2022 (UTC)[reply]
  11. I've being going back and forth on this one. But Levivich wording clarifies the issue for me only to a non-prose-containing databases. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:47, 19 October 2022 (UTC)[reply]
  12. Per WP:NOTDATABASE. Avilich (talk) 18:36, 20 October 2022 (UTC)[reply]
  13. Support articles sourced to non-prose-containing databases really only can be used to verify snippets of information and that the subject exists. Having one prose containing source would help combat this as it would ensure that mass creation through some kind of database crawl gets a better review by the creator as to whether it's notable. Dreamy Jazz talk to me | my contributions 23:32, 27 October 2022 (UTC)[reply]

Oppose (Q14: databases)

  1. Similar rhetoric is being used to weaken the notability of road articles at Wikipedia:Reliable sources/Noticeboard#Google Maps and I am concerned that the implications of this are not well understood. --Rschen7754 00:45, 13 October 2022 (UTC)[reply]
  2. As demonstrated in multiple venues around this issue, the proposer & I just fundamentally disagree on the utility of databases as sources. Espresso Addict (talk) 01:00, 13 October 2022 (UTC)[reply]
  3. I have no desire to limit the creation of articles simply because the source of the information is a database. If it meets the notability guidelines, then it deserves a place here. Supertrinko (talk) 02:19, 13 October 2022 (UTC)[reply]
  4. There seem to be cases where databases are reasonable as the main source for an article at creation - as discussed elsewhere. Of course, it'd be lovely to be able to expand that article to include plenty of other sources, but I can see cases where at creation this it is reasonable to simply use database sources at times. There also seem too many clauses open to interpretation here as well - which concerns me a little. I can certainly see topic areas where something similar to this would apply, but others where it should be reasonable to allow it - it seems a little too broad brush Blue Square Thing (talk) 06:12, 13 October 2022 (UTC)[reply]
  5. Not all databases are created equal. Moreover, there is a difference between prohibiting "articles sourced only to non-prose-containing databases" (which I would be more likely to support) and prohibiting articles from being "based on databases that do not contain significant secondary prose". You can have the database as the foundation of the article, and then have additional sources to flesh it out and I think that would be fine - but such an article would still be based on the database. To be clear, I would hesitate to support the "sourced only to" language - I'm still a little uncomfortable with that because I think some databases are better than others. But I absolutely oppose the "based on" language. ~ ONUnicorn(Talk|Contribs)problem solving 16:01, 13 October 2022 (UTC)[reply]
  6. per ONUnicorn. Not all databases are the same and they are used differently. While I recognize that users can use a database to create articles at scale, that doesn't inherently mean that we should throw out all databases. Consider a list/table of all federal judges, or of all species - in these cases the compiled data provides a start to an article of a subject that the community considers notable. Yes, editors can and should search for and expand the article, but unlike others here, I have less concerns with stub articles and more concerns with copyright violations, unverified information and WP:NPOV issues. --Enos733 (talk) 20:13, 13 October 2022 (UTC)[reply]
  7. Prose vs. not prose is not a measure of the reliability of the source material. A rambling blog post I write purely from my own opinion can be all prose... and it's definitely not as reliable for citations as a fact checked structured database, depending on the subject matter. Steven Walling • talk 04:40, 14 October 2022 (UTC)[reply]
  8. This area is already covered sufficiently by Question 2A * Pppery * it has begun... 15:06, 14 October 2022 (UTC)[reply]
  9. If the source is reliable, then it may be used. --Jax 0677 (talk) 18:57, 15 October 2022 (UTC)[reply]
  10. Per ONUnicorn. ಮಲ್ನಾಡಾಚ್ ಕೊಂಕ್ಣೊ (talk) 08:05, 16 October 2022 (UTC)[reply]
  11. GNG does not require sources to be written in "prose", nor does it preclude "databases". The problem is not "non-prose databases" per se. James500 (talk) 20:11, 16 October 2022 (UTC)[reply]
  12. No, non-prose providing database websites can be valuable, and data can be extrapolated to create valid articles. Not sure why those sources in particular are demonized.--Ortizesp (talk) 14:51, 20 October 2022 (UTC)[reply]
  13. I do not understand the focus on presentation style rather than reliability and content. Prose/not-prose is simply the wrong criterion to base such a drastic judgment upon. XOR'easter (talk) 17:58, 28 October 2022 (UTC)[reply]
  14. There's a wide range of databases out there, and I don't see how we can generalise about them. Inclusion in certain databases (say, accepted plant species in Plants of the World Online, or languages in Glottolog) pretty much guarantees a GNG pass. The requirement for prose is a red herring: how the information in a database happens to have been presented is not relevant. The records in many databases will have enough data to allow the writing up (in prose) of at least a Start-class article (for a random example, see what information is available in the various tabs of this USDA Plant Database record.) Many databases will also provide bibliographies for each entry, and these will contain the sources that could be used for any subsequent in-depth expansion (we're not requiring a B-class level of in-depth treatment from newly created articles, are we?). And for those quoting WP:NOTDATABASE above: I'm surprised that this needs pointing out, but that guideline says that Wikipedia itself is not an indiscriminate database, there's nothing there that limits the use of databases as sources. – Uanfala (talk) 12:20, 29 October 2022 (UTC)[reply]
  15. There's a lot of generalising about what kind of information you can find in databases above. To add to the list of counter-examples: terrasindigenas.org.br is a comprehensive database of Indigenous territories in Brazil. It doesn't have any prose, but a typical entry contains: the name, location extent, and ecological makeup of the territory; a list of peoples that live or lived there; a timeline of its legal status; a population history; and a list of links (usually at least a few dozen) to related news articles. You could easily create a decent start-class article from that, and each of the ~720 entries are about populated, legally-recognised places (i.e. notable topics), 99% of which are red links. Compare that to List of Indian reservations in the United States – all blue. As with other proposals here, this would needlessly hamper efforts to expand our coverage of the Global South to match what we already have for comparable topics in the North (thanks in large part to mass creation). – Joe (talk) 15:22, 31 October 2022 (UTC)[reply]
  16. Oppose making rules that rely on assumptions that are inadequately supported. Databases vary in range and reliability of content, as well as how it is presented. · · · Peter Southwood (talk): 18:15, 1 November 2022 (UTC)[reply]
  17. Meh. --Jayron32 11:52, 2 November 2022 (UTC)[reply]

Comments (Q14: databases)

Comments from JoelleJay (Q14: databases)

Another potential proposal: Definition of a "database" for the purposes of GNG. "Databases/statistical websites that do not contain significant prose text, written by a human, discussing the subject in detail, do not contribute to GNG". The issue obviously isn't with database pages that also contain substantial secondary prose coverage clearly written by humans, for example OMIM entries like https://omim.org/entry/609423, which has a list of the individual people responsible for writing that specific entry. The issue is with databases that do not contain any human prose written specifically about the subject: that is by definition a primary source and can never contribute to GNG. 20:32, 8 October 2022 (UTC)

@Espresso Addict, what is wrong with the OMIM example? I included it specifically because it has been used as an example of a "database that contains SIGCOV"; defining "database" for our uses to exclude anything where there is significant secondary prose coverage on the subject would eliminate the issue. 23:48, 8 October 2022 (UTC)
WhatamIdoing, if a database entry also has its own reliably published embedded video recordings or audio files or some visualization of ASL or whatever that would normally count towards GNG then that's obviously fine. 01:01, 9 October 2022 (UTC)[reply]
Espresso Addict, this isn't some random distinction I came up with just for this RfC, it's directly coming from our P&Gs. Just because one can cobble together verifiable facts from a source doesn't make the source secondary, which is required for GNG and OR. And secondary, per our policy, means analysis of the subject by an author. The database entry you linked clearly would not be sufficient basis for an article because it contains NO commentary on the subject: it is strictly the output of a database query. Just as much material could be collected for any high school athlete in the US.
WhatamIdoing, the key point is being discussed by other people in secondary RS. And I'm confident the vast majority of editors do NOT interpret GNG as if significant coverage in reliable sources that are independent of the subject means "significant coverage achieved by adding together multiple RS that are not necessarily all independent or secondary". That would be nonsensical since just about every topic has some secondary, independent coverage and some primary, non-independent SIGCOV. 04:09, 9 October 2022 (UTC)
Boca Jóvenes, the distinction between primary and secondary that we use is in the sense of whether information has been analyzed and discussed by someone directly (see the wording in WP:SECONDARY). So raw data that has been "interpreted" by tools (e.g. software that automatically identifies protein domains from a sequence; a db query that returns specific facts in a UX-friendly table), or even the output of human manipulation of data (someone manually compiling an athlete's performance stats) are still primary because we do not have anything that has been said about that information, it's ultimately merely been reproduced. If just curating or annotating data was sufficient to make it secondary, all experimental publications would be considered secondary for the novel research they present. Without people actually commenting on the subject directly, even when there are bountiful verifiable facts attached to it, we cannot discern the importance or encyclopedic merit of it (or of any of its facts). An editor can't just write an article on the relationship between mothers against decapentaplegic and Smurf based solely on string-db.org spitting out their predicted binding interactions; someone has to report that relationship specifically in RS, and then those findings must be discussed by an independent researcher before any of that info can be used on WP. JoelleJay (talk) 23:02, 9 October 2022 (UTC)[reply]

@Supertrinko, Espresso Addict but if the subject demonstrably meets the notability guidelines then there's no issue? Any db entry that has enough DUE information in it to kludge together an article of single-sentence facts should also have the secondary SIGCOV sourcing to actually write a real, NPOV, non-OR article--which negates the need to source the whole article to the database in the first place. But those databases where inclusion has been empirically shown to reflect notability are largely already covered in various SNGs (like the Messier catalog for NASTRO), and so this GNG-based criterion would not apply. Meanwhile, the vast majority of other databases that have not been evaluated as notability-granting (i.e. are indiscriminate) contain thousands to billions of entries with no info beyond routine parameters reported in a primary source, and therefore could never be notable. But even among indiscriminate databases, if a particular entry contained significant secondary original prose coverage it would be exempted from this criterion. Every single case of problematic mass creation has been exclusively based on databases. If we can't ensure that all entries in a database actually meet GNG, then we are permitting bulk creation of articles that, by definition, should not exist. And preventing that is the whole point of this RfC... JoelleJay (talk) 06:38, 13 October 2022 (UTC) @Blue Square Thing I would be interested to know which GNG-based topics should license mass creation of articles without the assurance that at least one piece of secondary SIGCOV exists? JoelleJay (talk) 06:50, 13 October 2022 (UTC)[reply]

Moved reply from Atsme to own section. –MJL ‐Talk‐☖ 20:48, 1 November 2022 (UTC)[reply]
  • @Enos733, this wouldn't affect subjects governed by non-coverage-based criteria (like federal judges). This restriction would be for databases where inclusion does not indicate notability by itself (and would therefore require multiple pieces of SIGCOV in SIRS to demonstrate notability). JoelleJay (talk) 23:56, 13 October 2022 (UTC)[reply]

Uanfala, if inclusion in a particular database is considered an indication of further GNG coverage elsewhere 95% of the time, then that database can be discussed at the appropriate venue as a reliable-enough notability predictor for mass creation purposes. However, GNG is not itself demonstrated by sources that only "indicate" other coverage is likely to exist. For a reliable source to contribute to GNG it must itself contain independent, secondary SIGCOV; a primary/tertiary database, such as one that is automatically populated with information scrubbed from keywords in primary research papers or even secondary review papers, without that information being contextualized in the database entry with discussion of it by a human, by definition cannot meet the "secondary" criterion. There is zero way to determine which of potentially thousands of pieces of data in an entry are DUE in a WP article if we do not have someone else (and per OR, this definitely cannot be a specialist WP editor) interpreting it directly. For example, in this database entry it would be impossible to choose which info should be summarized in prose on WP because it just contains all published information without any analysis. On the other hand, this database entry on the same protein contains much of the same information, but it has been regularly filtered, interpreted, and contextualized by specific named contributors since 2001. That is the distinction the prose requirement is meant to provide. JoelleJay (talk) 21:40, 29 October 2022 (UTC)[reply]

@WhatamIdoing, I'm not saying we have to ban such sources, they just should not be used for mass creating articles if a) there is not a consensus that inclusion in the database is extremely predictive of GNG (i.e. it could be used as a criterion in a GNG-predicting SNG) or b) the database source itself cannot contribute to GNG (by virtue of being primary/tertiary). JoelleJay (talk) 22:25, 30 October 2022 (UTC)[reply]
@Joe, no one should be mass creating articles from a database where inclusion has not been shown to predict GNG with high accuracy or does not itself contain enough secondary analysis (or links thereto) to count towards GNG. It doesn't matter how many severable facts a database has if they're all autopopulated from primary reports or non-independent uploads; otherwise we'd be allowed to have articles on all 58,502 wastewater treatment plants in the HydroWaste database (25 "facts" each) or all 11,000+ clones in the Ahringer RNAi library (22+). Without secondary discussion of the topic editors cannot know which facts are DUE in an encyclopedia. Especially if each entry in a database that does not predict GNG is guaranteed to contain data for the same set of parameters (e.g. area, range, etc.), writing articles on each entry based only on that info is pretty much by definition violating INDISCRIMINATE. JoelleJay (talk) 05:36, 2 November 2022 (UTC)[reply]
@WhatamIdoing, this issue kind of goes hand-in-hand with whether a source can contribute to GNG and is predicated on the assumption that all the articles in question are actually GNG-notable. Therefore sourcing that goes beyond basic primary-sourceable, fill-in-the-template parameters must exist somewhere, and for that we require the ability to write an NPOV/non-OR article. So in the case where we have topics that are not already determined to be notable via an SNG/GNG predictor, my argument is that database-derived mass creation of articles on those topics should not proceed unless the database itself (or the citations it contains for each article yada yada) can at least halfway contribute to GNG--and that means the source should already contain significant secondary prose discussion of the salient facts. If all that exists on the school in your example are the simple infobox facts you listed, then it clearly fails GNG and shouldn't be a standalone. JoelleJay (talk) 05:08, 3 November 2022 (UTC)[reply]
@WhatamIdoing, by sourcing that goes beyond basic primary-sourceable, fill-in-the-template parameters must exist somewhere and If all that exists on the school in your example are the simple infobox facts you listed, then it clearly fails GNG and shouldn't be a standalone. I am directly referencing the fact that GNG sources must still exist even if they aren't in the article.
Why should we approve mass creation of a bunch of GNG-dependent articles based on a database source where inclusion has not been established to predict GNG and that does not itself demonstrably meet GNG at least halfway? This RfC is meant to address the major problem with mass creation identified in the arb case; any remedy, regardless of which definition of "mass create" is decided on, will necessarily involve some type of preemptive restriction or way to identify mass creation. I am simply suggesting something we a) could use should we opt for a source approval-based method for mass creation, and b) could employ for identifying past cases of mass creation. If you disagree that mass creation of articles on subjects of unclear notability is, was, or could be a problem, then just say that and I won't keep trying to adjust this proposal's parameters to accommodate you. But I'm of the opinion that such an activity is harmful to the project and if reining it in means slightly modifying how NEXIST is applied in certain circumstances and introducing the tiniest inconvenience to database-driven mass creators, then it's worth it. JoelleJay (talk) 21:38, 3 November 2022 (UTC)[reply]
Comments from Espresso Addict (proposal 14)

JoelleJay: I wouldn't include the OMIM example; that isn't an edge case!! WhatamIdoing might be able to help? I don't think it should be limited to online databases; print almanacs or gazetteers are also problematic, perhaps even more so as it is more difficult to see what coverage they have. Espresso Addict (talk) 22:06, 8 October 2022 (UTC)[reply]

JoelleJay: I can't do word counts on this machine but the OMIM entry you linked must be >>1000 words with 62 refs. It's actually possible to make a viable article based on something as sparse as [1], which has no prose at all at the top level, though there's a number of clickthroughs. I'm beginning to think the key is reliability (expert/authorative authorship) not amount of prose. Espresso Addict (talk) 01:28, 9 October 2022 (UTC)[reply]

Boca Jóvenes: "The problem is invariably the database's own programming which they use to produce statistical summaries of various kinds and I would argue that these outputs are potentially, if not probably, unreliable." I don't understand this at all. I've worked with databases (as an amateur) and the problem is data input errors, not output problems? I asked my OH, who has expert professional experience in this area, who agreed with my assessment. Espresso Addict (talk) 00:35, 10 October 2022 (UTC)[reply]

Comments from WhatamIdoing (proposal 14)

User:JoelleJay, why does coverage need to be in prose? Are you hoping to ban videos as sources? WhatamIdoing (talk) 00:17, 9 October 2022 (UTC)[reply]

It sounds like the key points for JoelleJay are "secondary" and "clearly written by humans"; prose per se is not the key point.
We never have been able to agree on whether the WP:SIRS rule (a single source, considered in isolation, must have all the qualities that suggest notability [independent, secondary, significant coverage, etc.]) should be applied generally, or if you should look at the independent sources in combination to decide whether, taken as a group, there is enough information to write an article (e.g., a page here about his family, a short article there about his early career, a brief analysis of how he compares to two rivals over there...). If you lean towards the latter, then adding one bona fide secondary (but perhaps brief) source to one detailed (but non-secondary) database entry would indicate notability. WhatamIdoing (talk) 01:36, 9 October 2022 (UTC)[reply]
Joelle, every time I've tried to get the folks at WT:N to agree that all sources must be secondary and individually contain SIGCOV for GNG's purposes, we've been unable to reach consensus. We can reach consensus that all sources must be independent to count (as long as we are fairly superficial with the definition of independence; find my talk page if you want to talk about the impossibility of finding scientific sources covering experimental drugs that are truly independent from the would-be manufacturer of a drug candidate), but we have historically been unable to reach consensus that only sources individually containing SIGCOV should count at all, or that only sources that are primarily/entirely secondary should count at all. If this is the sort of obvious truth that No true Wikipedian would ever disagree with, then we must have a lot of fake Wikipedians. ¯\_(ツ)_/¯
@Boca Jóvenes, you might be interested in the concept of secondary data, which is not exactly the same thing as a secondary source. It is at least true that some databases are primary sources; it might be true that all databases are (or at least all of the "isolated data point" parts are) primary.
More generally, the question of effort might be relevant. If someone published an article that says "I have taken the list of basketball players in this group, along with their heights and the points they each scored. I have manually calculated the statistical contribution of height to points scored. I conclude that taller players score more points, and for players with an above-average height, I list the percentage of points scored that can be attributed to height for each individual", we would be happy to include this the athletes' articles. But if you dump it in a database (or a spreadsheet), and the software does this simple calculation for you, then the same information is suddenly not okay? There was still a human deciding that this was something worth analyzing. There was still a human making the analysis happen. I think we should not encourage a technology-specific understanding of analysis. WhatamIdoing (talk) 02:56, 10 October 2022 (UTC)[reply]
@BilledMammal, I think your rationale might want to be re-written. NOTDATABASE says that Wikipedia is not a database; it does not say that you can't or shouldn't cite them. Also, it's really easy to put information into context with a database. Context is what databases are good at. Context looks like "<Subject> (<birth>–<death>) was a <nationality> <profession>". That places the subject in the context of a particular time, place, and profession. You can easily get that kind of basic statement of context from a database entry while fully complying with WP:NOR. WhatamIdoing (talk) 01:37, 14 October 2022 (UTC)[reply]
Is "Elizabeth II (Elizabeth Alexandra Mary; 21 April 1926 – 8 September 2022) was Queen of the United Kingdom and other Commonwealth realms" a database entry in prose form? That's the kind of sentence that can be created from a database entry. WhatamIdoing (talk) 02:46, 14 October 2022 (UTC)[reply]
@Scope creep, what exactly are "these databases" in your claim that "These databases are full of errors that are not validated upon use, just accepted"?
Here's a link to a database: https://icd.who.int/browse10/2019/en Is this database full of errors? Are the entries in this database unvalidated?
Here's a link to another database: https://data.census.gov/cedsci/table Is this database also "full of errors", "unvalidated", etc.? WhatamIdoing (talk) 21:58, 30 October 2022 (UTC)[reply]
Scope creep, if it's "absolute madness" to think that databases could make "for good sources, in any instance", then what do you propose we use for statements like "The name for this medical condition under the ICD-10 is _____" or "According to the 2022 US Census, the population of Mulberry, Kansas was 409 people"? We normally take that information straight from databases – databases that have been validated and checked, and are widely trusted by other reliable sources.
The kind of data problems that you're describing are the ones that have resulted in me receiving spam for my neighbor under her prior name. That's not the kind of database that editors are generally looking at. WhatamIdoing (talk) 01:18, 31 October 2022 (UTC)[reply]
I suggest to you that if "Within reason" were operating on the very emotional subject of whether other people are allowed to clutter of Wikipedia with pages about subjects some editors don't care about, this RFC wouldn't even exist. WhatamIdoing (talk) 14:46, 31 October 2022 (UTC)[reply]
@JoelleJay, your comments ("impossible to choose which info should be summarized", etc.) suggest that you are looking for sources, added at the time of the article's creation, that would be sufficient on their own to write a whole article. Thus you look at the Ensembl genome database project entry for Zbtb7 [2] and say "Oh, that's not enough all by itself; therefore we should ban this". Someone else might look at the same thing and say "Well, I can't write a whole article, but I can start an accurate page that says 'Zbtb7 is a human gene on Chromosome 19 that has been called the pokemon gene'." What's wrong with starting an article that way? WhatamIdoing (talk) 22:11, 30 October 2022 (UTC)[reply]
Okay, so instead of just creating the article, you think I should get written permission to create the article? And that permission depends not on whether the subject is notable, but instead on whether everyone agrees that the source I plan to cite (e.g., Ensembl) only (within a 95 percent confidence interval) contains information about notable subjects? WhatamIdoing (talk) 23:27, 30 October 2022 (UTC)[reply]
JoelleJay, you wrote Without secondary discussion of the topic editors cannot know which facts are DUE, but I don't think that's necessarily true at the stub level.
Some years back, during the endless debates about WP:SCHOOLOUTCOMES, I asked a group what they thought about the notability of a particular 19th-century school in California. What's known about the school is: (a) the approximate dates of operation, (b) the location of the building, (c) why it closed, and (d) whether it ever issued a high school diploma. A few editors firmly declared that if it ever issued a high school diploma, then it deserved an article. They were stumped, though, on the question of what to name it, because the name of the school (including whether it formally had a name at all) is unknown.
Do you really think that I **need** a secondary source and to consider questions of DUE to start an article about that school? Pretty much every school article should start in the same way: "That School was a school in Somewhere, California. It was open from 187x to 187y."
I agree that without secondary material – also, without independent sources and without multiple sources – DUE becomes a significant barrier when you want to expand the article past the basic definition. But just for starting an article on a routine subject, I don't think DUE normally matters much, if at all. WhatamIdoing (talk) 02:30, 3 November 2022 (UTC)[reply]
JoelleJay, WP:N has a whole section dedicated to declaring that notability doesn't depend on whether the cited sources demonstrate notability at all.
This must be true:
  • The subject must be qualify for a separate, stand-alone article.
  • The subject's notability must be provable (e.g., if the article ends up at AFD) through at least two suitable sources.
This would be convenient for other editors, but has never actually been required by any policy or guideline:
  • The evidence of the subject's notability that you would present at AFD (if it ever became necessary) must be cited in the article when you create the page.
Your comments seem to assume that the latter is actually a rule, or should be made a rule. WhatamIdoing (talk) 14:59, 3 November 2022 (UTC)[reply]
JoelleJay, I think the gap between our thinking might be this: I'm talking about starting an article on a subject that the editor believes to qualify for a separate article. You seem to be talking about proving that the editor's belief is correct through a particular mechanism (citing at least one source that contains a substantial amount of prose).
I'm not sure if the requirements for starting a lot of articles need to be different from starting a couple of articles. I'm not sure if the sourcing requirements for starting articles need to be higher than the practical requirements of citing the content that you've actually added. At the moment, I'm leaning towards yes to the first question (but only when it's a lot of articles in a short period of time – more than the 25 to 50 per day [or week] that MASSCREATE seems to be talking about) and towards no to the second question. It sounds to me like you answer yes to both. WhatamIdoing (talk) 16:58, 4 November 2022 (UTC)[reply]
Comments from Blue Square Thing (proposal 14)

I'm not sure that the problem is actually the use of databases, rather the inappropriate use of individual database entries with no prose at all as the sole source. This is probably more important for the deletion RfC. You'll struggle to include very many databases if you want to define them at a entire source level - Olympedia clearly has some human created significant prose entries. Blue Square Thing (talk) 07:42, 9 October 2022 (UTC)[reply]

@JoelleJay: taxonomy, for example, seems reasonable, but if you want an absolute "assurance" rather than a "reasonable assumption" then I'd actually say almost anything that has a reasonable notability standard. Blue Square Thing (talk) 17:11, 14 October 2022 (UTC)[reply]
Comments from Boca Jóvenes (proposal 14)

Interesting one but I don't agree that databases are primary sources. They are definitely secondary. In the first place, the typical database has human input but it probably lacks adequate oversight and so transcription errors occur. That in itself doesn't mean they are unreliable because all sources contain errors. The problem is invariably the database's own programming which they use to produce statistical summaries of various kinds and I would argue that these outputs are potentially, if not probably, unreliable. I think the human input content should be welcome even if there is an occasional human error, but the database's auto-generated outputs should be treated with caution and, perhaps, classified as unreliable. BoJó | talk UTC 11:46, 9 October 2022 (UTC)[reply]

Comments from ActivelyDisinterested (Q14)

I have concerns regarding using a single database without any other sources, but I believe that's a bigger issue than this specific RFC. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 01:22, 13 October 2022 (UTC)[reply]

Comments from ElDubs (Q14)

Reply moved from #Comments from JoelleJay (Q14: databases) –MJL ‐Talk‐☖ 20:48, 1 November 2022 (UTC)[reply]

Again I'm not in favour of a rule specific to databases here. What you're highlighting is an issue with using a single source to create an article. That's already a guideline that exists that should be enforced in such situations. We're also able to add SIGCOV after the fact at some point, so I'm quite in favour of mass creation from a database of notable subjects. Supertrinko (talk) 09:17, 13 October 2022 (UTC)[reply]
Comments from BilledMammal (Q14)

@WhatamIdoing: A database entry in prose form is still a database entry. BilledMammal (talk) 02:07, 14 October 2022 (UTC)[reply]

The required context with explanations referenced to independent sources doesn't have to be in the same sentence to comply with WP:NOTDATABASE, but it does need to be in the same article. BilledMammal (talk) 03:41, 14 October 2022 (UTC)[reply]
Comments from Scope creep (Q14)

I know from experience there seems to be a spectrum in terms of quality of database sources runing from the websites that don't have money to update them, gaming and sports and sports companies, who don't care if they're accurate, right up to academic database sources, where accuracy is prized and they're continually checked and updated. They're not secondary sources though, unless there is some kind of analysis taken place, where is the human aspect? Academia follows that standard. They would never use them in that manner as secondary sourcing. If they are deemed to be accurate, then using them to accurately cite date ranges, specific events for example in BLP's is one good way of using them, but to use them cite the article to define notablity or criteria that involve some kind of deep historical analysis, is an entirely different thing. It's simply mot accurate and there is no depth to them. There is nothing there. You can't build deep historical articles or even modern contemporary articles using these. Its impossible. We should be aiming for reasonably decent, i.e large high-quality articles that are well sourced, structured properly and so. It may be a case that automated tools like AI for example may be used in a manner that builds a kind of framework article, as a kind of helper mechanism, then the editors ties it all together with proper analysis, but doing it with types of sources is going to stymie the whole vision very quickly. Its probably a truism that really poor sources means really poor low-quality articles, and we have seen that already. We should be seeking a grander vision and its not this. scope_creepTalk 21:22, 17 October 2022 (UTC)[reply]

@User:WhatamIdoing. Databases are everywhere. Without them the internet wouldn't work and they are full of errors and mistakes. Companies spent a lot to fix it: Staggering cost of data errors. Depending on who your referencing, will define how reliable it is, but its not automatically reliable in every instance. Big corporates and academia spend a lot of money to make sure their data accurate. For example the last big multinational corporate I worked for, I heard they'd spent $255million fixing, verifying and cleaning their data. Its a spectrum, so at the low-end, the data will have more errors in it, as the company doesn't have the money to keep the data in order. That is physical fact. It is absolute madness to suggest that database driven data makes for good sources, in any instance. scope_creepTalk 23:36, 30 October 2022 (UTC)[reply]
@User:WhatamIdoing Within reason like everything else. Not as a secondary source and not all the article sourced to databases. And with a proper policy in place. scope_creepTalk 08:57, 31 October 2022 (UTC)[reply]
Comments from Rhododendrites (Q14)

So which is it? Based on? Sourced only to? Easily sourceable only to? Oppose/support/[???] I guess. — Rhododendrites talk \\ 12:36, 24 October 2022 (UTC)[reply]

Comments from Editor X

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 15: How about to award the expanders of articles (stubs) with the notifications that usually only the creator receives?

Support (awards)

  1. as nominator. If it would only count for stubs or articles in general could be thought about, but the expanders of the articles are likely the ones who are also interested in maintaining the articles.Paradise Chronicle (talk) 23:28, 9 October 2022 (UTC)[reply]
  2. Support - --Jax 0677 (talk) 23:38, 9 October 2022 (UTC)[reply]
  3. I'd been thinking of trying to codify a proposal to suggest this. I don't know whether the underlying software can handle it, but there seems no reason in principle why anyone can't ask to receive notifications on any article. Espresso Addict (talk) 00:39, 10 October 2022 (UTC)[reply]
  4. Support some kind of additional incentive for expansion - Expansion of stubs needs to be encouraged more. The plain truth is that simply creating stubs does not seem to encourage their expansion into real articles in any kind of reasonable time-frame even where the content exists for such an expansion (and especially if it does not). FOARP (talk) 09:22, 11 October 2022 (UTC)[reply]
  5. Support Seems reasonable, why not do what is already in this instance. It engenders community collaboration and definition by community consensus as well as drawing folk in. scope_creepTalk 18:52, 14 October 2022 (UTC)[reply]
  6. I don't understand why this is being described as an "award", but it will quite clearly be an improvement if substantial contributors to an article could count as "creators" for various purposes: for example, when Twinkle sends out an AfD notice for the article, or when an incoming link is created. – Uanfala (talk) 12:29, 29 October 2022 (UTC)[reply]

Oppose (awards)

  1. This will only end up being gamed for the same reasons. Also I can't think of anything that would put me off developing stubs more than this. If people really want bonus points they can take articles to DYK Blue Square Thing (talk) 06:35, 10 October 2022 (UTC)[reply]
    It wouldn't put me off creating stubs, but I will note that stubs rarely get improved after their creation. Abductive (reasoning) 11:35, 10 October 2022 (UTC)[reply]
  2. We have WP:DYK. --Rschen7754 04:07, 17 October 2022 (UTC)[reply]
  3. No, you don't get awards for creating great articles, why would you get one for expanding one.--Ortizesp (talk) 14:53, 20 October 2022 (UTC)[reply]
  4. Oppose as useless for solving the problem we are supposed to be addressing here. If the stubs are on unencyclopedic topics they literally cannot be improved by editing. · · · Peter Southwood (talk): 18:09, 1 November 2022 (UTC)[reply]

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Comments (awards)

Question from WhatamIdoing (awards)

Paradise Chronicle, what do you mean by "to award with notifications"? Is this like "winning extra spam"? (Here's a reminder to others that Special:Preferences#mw-prefsection-echo is worth looking into.) WhatamIdoing (talk) 02:59, 10 October 2022 (UTC)[reply]

@Espresso Addict, you (currently) can't get notifications such as "Page link" ("please bother me every time someone adds a link to any article the software thinks I created") for pages you haven't created. However, I posted that link for the benefit of experienced editors who would like to get fewer notifications. WhatamIdoing (talk) 04:52, 10 October 2022 (UTC)[reply]
Comments from Espresso Addict (awards)

WhatamIdoing: I've just looked at that -- it doesn't look as if you can get notifications for articles that you haven't created? Espresso Addict (talk) 03:29, 10 October 2022 (UTC)[reply]

WhatamIdoing: I quite like them -- it reminds me when I'm on wikibreak that I ought to be getting back. It's also interesting to see which of "my" creations continue to get links, and might be worth trying to put some more effort into. Espresso Addict (talk) 05:11, 10 October 2022 (UTC)[reply]
Comments from Paradise Chronicle (awards)

I am rather grateful to receive the notification if a wl is added to an article I created. It encourages to expand the articles accordingly. Also over the user views one can get notified if some event is related to an article one has created. An option would be adding an other symbol for user views and wls beside the blue star for the watchlist. Could be made over a tool.Paradise Chronicle (talk) 20:33, 10 October 2022 (UTC)[reply]

WhatamIdoing, the notifications only concern on changes of MY creations (among other mentions and thanks etc.) not the ones of articles I am potentially interested in. For example if one expands an article from a stub, the one who created the stub gets the notifications, not the one who expanded the article.Paradise Chronicle (talk) 20:33, 10 October 2022 (UTC)[reply]
@Rschen7754 I guess stubs are not mass created for DYK, they are often created for being first or being very productive. I would like to find something which allows the average Wikipedian to get recognition and not only the ones who aim for DYK. Like the ones who are interested in the topic, are expanding or have expanded an article considerably but are not yet confident enough to go for a GA.Paradise Chronicle (talk) 16:27, 28 October 2022 (UTC)[reply]
@Ortizesp and I see this list as an award/recognition. And others see it similarly. It is also the amount of articles the media reports about, usually not the quality of the articles. Democracy, the consensus at times is also noted. Would be great if the amount of articles would get less notability/coverage and focus would shift to working together, consensus, quality efficiency/usefulness in everyday life of Wikipedia.Paradise Chronicle (talk) 17:28, 28 October 2022 (UTC)[reply]
Comments from Rhododendrites (awards)

I've never heard of notifications as being an award (or even particularly positive), but it seems technically challenging and kind of out of scope for this RfC IMO. — Rhododendrites talk \\ 12:32, 24 October 2022 (UTC)[reply]

Comments from Rlendog

I don't understand Abductive's comment that "stubs rarely get improved after their creation." I have seen many stubs get improved after their creation, and in some cases well after their creation. Rlendog (talk) 16:55, 13 October 2022 (UTC)[reply]

Comments from Blue Square Thing (awards)

To clarify my view above, I honestly can't think of anything worse then being given automatic credit for stub development (something I do quite often). I just want to edit. I don't need a prize for doing that - which is why I've only ever sent one article to DYK and am very unlikely to ever send am article to GA let alone FA. If anything automatic like this exists then at least give me the option of opting out if it - it would, quite seriously, put me off editing. Blue Square Thing (talk) 17:21, 14 October 2022 (UTC)[reply]

Comments from Abductive (awards)

Reply moved from #Comments from Rlendog –MJL ‐Talk‐☖ 20:18, 1 November 2022 (UTC)[reply]

  • @Rlendog:, I have created about 1700 non-list, non-disambig articles. About 1500 are still stubs. 200 or so have never been edited since my last edit, even to make minor corrections such as formatting or adding categories. Moreover, 3,749,266 out of 7,449,754 pages are assessed as stubs. This is 57% of the 6,562,906 articles on Wikipedia. Of course stubs get improved, and this is the hope of stub-creating editors such as myself (but not, I think, those completionist users who are only migrating databases onto Wikipedia). But as you note, some stubs don't get improved until "well after their creation," which suggests that the improvements are not motivated by any abstract urge to simply improve stubs. So I guess that is my definition of "rarely". Abductive (reasoning) 16:50, 16 October 2022 (UTC)[reply]

Question 16: Develop metric to measure editor's engagement in article improvement

Develop a metric that attempts to measure the degree to which an editor significantly improves articles created by other editors, either by adding sentences or paragraphs of sourced text or by adding reliable references to source existing text. It might be as simple as a count of articles that the editor had improved, or attempt to differentiate between adding one reference vs ~100-folding the article & getting it featured. This would complement existing metrics such as edit count, number of articles created, number of GAs reviewed, number of DYKs, and would facilitate the creation of user boxes or similar stating "I've improved n articles on Wikipedia" and/or "My [article improvement index] = y"

Support (article improvement metric)

  1. Support as nominator. Espresso Addict (talk) 01:02, 10 October 2022 (UTC)[reply]
  2. Support a constructive and cooperative approach to the mass creation of articles.Paradise Chronicle (talk) 08:20, 10 October 2022 (UTC)[reply]
  3. Support in principle, although I wonder about the practicalities. Ingratis (talk) 07:31, 12 October 2022 (UTC)[reply]
  4. A metric to praise those who improve articles is a fantastic idea. It sounds difficult to implement, but I support it in principle. Supertrinko (talk) 02:21, 13 October 2022 (UTC)[reply]
  5. Support There's a lot of talk of quality rather than quantity so let's have a measure of it. Such a value added metric would help in nudging editors to add quality and value to the project. Currently, too much weight is given to simpler metrics such as edit count, which seems to encourage huge numbers of low-value edits. For example, by participating in the discussion we are boosting our edit count but doing nothing to improve articles. The editors who are creating and improving content in a detailed and uncontroversial way need more recognition and reward but we can't do that if we can't measure their effort.
  6. Support It will be extremely disruptive at the begining and its important to understand who is doing what and why. Either that or its uncontrolled and we have already seen what that leads to. If its permission based, then assuming an editor comes in and creates 2k of broken articles and doesn't fix, it needs to be known if they are fixing them, never mind expanding them, so it can be controlled, at least initially. scope_creepTalk 16:33, 14 October 2022 (UTC)[reply]
  7. Support. This is a thing I would line to see, because it would be a cool thing to have, but I do not think it will have any useful effect on the problem we are addressing here. · · · Peter Southwood (talk): 18:03, 1 November 2022 (UTC)[reply]
  8. Support if practical as it will give editors incentive to improve articles. However, I don't think it will solve this problem since articles that can't be improved beyond one line cited to a database won't be, and articles that can be improved were never problematic in the first place as there is no deadline to improve them. As such, it may be out of scope for this RfC, but Wikipedia is not a bureaucracy and this is where the discussion is happening, so I'm not going to oppose it on that reason alone. Smartyllama (talk) 12:55, 2 November 2022 (UTC)[reply]

Oppose (article improvement metric)

  1. This will only end up being gamed. And, as above, I have little interest in creating stats such as this. Blue Square Thing (talk) 06:36, 10 October 2022 (UTC)[reply]
  2. Oppose - This is pure solutionism. FOARP (talk) 07:24, 10 October 2022 (UTC)[reply]
  3. Oppose - While article improvement is commendable, it doesn't solve the underlying problem of articles that can't be improved because significant coverage doesn't exist. Editors often skip over these articles when they can't find adequate sourcing, with the assumption that someone else will. We need solutions that specifically address the creation of unsourceable articles. –dlthewave ☎ 03:10, 12 October 2022 (UTC)[reply]
  4. Oppose, duplicates WP:Assessment scale, which is neglected anyway. Abductive (reasoning) 13:49, 29 October 2022 (UTC)[reply]
  5. Oppose Hate this. See Goodhart's law. --Jayron32 11:53, 2 November 2022 (UTC)[reply]

Comments (article improvement metric)

Comments from Espresso Addict (Q16)

No idea how feasible this is technically. Might be useful for evaluating editors, too. Espresso Addict (talk) 01:02, 10 October 2022 (UTC)[reply]

@Rhododendrites: If mass creation seeded articles that were worth having it would be less problematic; however it currently appears rare for others to significantly improve the microstubs resulting from mass-creation efforts, and mass creators rarely return to their microstubs. Motivation for improving articles started by others is a major problem -- I do it from time to time and it feels a bit ignored and thankless compared with starting my own. A metric that would permit editors to track how much work they have done in this area, and perhaps to work on increasing their score, might be just one means of addressing this. Espresso Addict (talk) 16:51, 24 October 2022 (UTC)[reply]

@WhatamIdoing: I'm entirely happy with sourced informative stubs. I'm not so fond of "A is a village near town in county, country" or "A b is a species of plant/animal in the genus A" microstubs. I do think that there is little or no external encouragement to improve (micro)stubs, or to source unreferenced articles (of which we have 135,570). As a lot of editors appear to be motivated by metrics I'd like to explore creating one that editors could track that would tempt them towards these useful activities. I don't think we need to encourage editors to create stubs, there seems to be adequate encouragement already. Espresso Addict (talk) 23:05, 30 October 2022 (UTC)[reply]

Comments from Rhododendrites (Q16)

Sounds good, but how does that help us define and/or resolve mass creation issues (i.e. seems out of scope for the RfC). — Rhododendrites talk \\ 12:29, 24 October 2022 (UTC)[reply]

Comments from WhatamIdoing (metrics)

User:Espresso Addict, are stubs not worth having? Imagine that you get a text message: "The doctor says it's <something you've never heard of>". Are articles like Fibrofolliculoma or Universal angiomatosis, which are tiny but sourced and accurate, actually "worthless" in such a situation? Or could they be better than nothing, given that they provide you with some basic information (e.g., the first involves little bumps and the second has bleeding problems) and can lead you to more information? WhatamIdoing (talk) 22:39, 30 October 2022 (UTC)[reply]

But "A is a village near town in county, country" is informative. In fact, if I'm looking up a small place, that's usually the only information I'm looking for. WhatamIdoing (talk) 23:21, 30 October 2022 (UTC)[reply]
Comments from Editor X

Please open your own section with username in the heading

Question 17: Amend WP:MASSCREATE

  • 1. Amend WP:MASSCREATE to explicitly recite WP:MEATBOT, and explicitly clarify that mass-creation through repetitive editing by hand is not different for policy purposes to automated/semi-automated mass-creation.
  • 2. Make getting consensus for creation prior to mass creation per WP:MASSCREATE mandatory.

Support amending WP:MASSCREATE

  1. Support both 1 and 2 as proposer - An argument repeatedly deployed by mass-creators was that they were not mass-creating, even when it was at the rate of dozens of articles per day through simply filling spaces in a template, as they were performing their mass-creation through repetitive editing by hand. They were able to argue this due to a perceived ambiguousness of WP:MASSCREATE/WP:MEATBOT as to whether they included mass-creation through repetitive editing by hand. Point 1 is intended to clarify that mass-creation by hand is also mass-creation.
    Point 2 is designed to promote engagement with mass-creators and the wider community at an early stage. I feel strongly that editors spending many hours of their day, every day, robotically creating articles is not good for their mental health or co-operative engagement with other editors, and is path that we should steer people away from where possible. Instead editors should be encouraged to discuss their planned mass-creation at an early stage with other editors, get feedback from them, and possibly even explore options for using tools/bots to speed up the process where there is a consensus to do so. This discussion can be in any appropriate forum (e.g., a relevant project). I think an earlier intervention in the cases of the mass-creators for whom we have had ARBCOM cases in recent years might have avoided the sad outcome of these proceedings. I think this is a minimalist solution based on existing policies and is not mutually exclusive with the other proposed solutions above.
    The enforcement mechanism for Point 2 would be a subject for the next RFC.FOARP (talk) 11:07, 11 October 2022 (UTC)[reply]
  2. Support. Problematic mass-creation is already semi-automated through the use of spreadsheets, scrapers, or templates, but it is a good idea to remove any ambiguity. I also support Point 2; at the moment this is an implicit requirement (Wikipedia:Bots/Requests for approval only grant approval for mass creation tasks when there is a broader consensus) and making it an explicit requirement also helps to remove ambiguity. BilledMammal (talk) 13:58, 11 October 2022 (UTC)[reply]
  3. Support both. Good and bad editors both can over-indulge in editing Wikipedia, but the issue is the content by bad editors. Give them an early warning that bot like editing may have consequences, so as to limit damage to the project. The analogy is WP:MEATPUPPET and WP:DUCK. LessHeard vanU (talk) 13:03, 12 October 2022 (UTC)[reply]
  4. Support for both proposals. If the issues at AfD are going to be solved mass creation needs to not be the norm. By getting edutors to discuss mass creation first AfD won't be clogged up later having to deal with any issues. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 15:27, 12 October 2022 (UTC)[reply]
  5. Support- Ideally we would have one policy and approval process that covers all mass creation tasks, automated or not, as the effect is the same. This would also eliminate the loophole for editors who copy-paste the output of offline/local automated processes and insist that they're doing it all by hand. –dlthewave ☎ 15:55, 12 October 2022 (UTC)[reply]
  6. Support MASSCREATE already includes article creation manually by humans, and would not be amended to include that, as suggested by the first oppose. Stub creation that is high-volume can in fact be higher risk, so permission should be required and this should be explicit in the policy. Reywas92Talk 17:20, 12 October 2022 (UTC)[reply]
  7. Support, in the understanding that anything resembling automated creation off of a list would require consensus. This is really what I tried to argue above and on the talk page, but apparently I'm not that good at getting my point across. Vanamonde (Talk) 23:37, 12 October 2022 (UTC)[reply]
  8. Support, especially if coupled with a prohibition on using non-prose databases as the sources for such articles. JoelleJay (talk) 00:21, 13 October 2022 (UTC)[reply]
  9. Support - Getting some agreement beforehand would solve a lot of the issues mass creation has been at the root of. -Indy beetle (talk) 08:44, 13 October 2022 (UTC)[reply]
  10. Support both— when doing any large-scale series of edits, including creations, getting a double-check from the community should be the common-sense thing to do anyhow. This feels like it harmonizes fully with our existing prohibitions against botlike edits without approval. Der Wohltemperierte Fuchs talk 12:12, 14 October 2022 (UTC)[reply]
  11. Support 2 2 is critical I think to keep control, certainly. For point 1, they are not the same in any kind of way, even if your creating hundreds of atrticles manually, or even thousands, its not the same scale. Not now. They are fundamentally different and there is more and more evidence of this. I've not seen the WP:MASSCREATE policy before. I suspect it probably needs to be updated in some way, in the near time frame, but not sure how. There is so many unknows. scope_creepTalk 17:28, 14 October 2022 (UTC)[reply]
  12. Weak support Mass-created (semi-)manually written articles, if not done with additional sourcing to non-database sources, are identical to bot-created articles, and should be treated as such. I'd be okay with something like, "create 10 articles in mainspace to show what you mean", and then ask for permission if you want to make more. I disagree with the citation of being bold below, as mass creation is inherently controversial (WP:CAREFUL). Ovinus (talk) 00:42, 15 October 2022 (UTC)[reply]
  13. Support per above. Levivich (talk) 20:04, 17 October 2022 (UTC)[reply]
  14. Weak support the placement in relation to bots is odd, but the spirit of the suggestion is constructive. Jontesta (talk) 23:16, 17 October 2022 (UTC)[reply]
  15. Support both 2 in particular is important to prevent an unmanageable number of low quality stubs from being created in the first place. Avilich (talk) 18:48, 20 October 2022 (UTC)[reply]
  16. Support, as above. wjematherplease leave a message... 10:37, 24 October 2022 (UTC)[reply]
  17. Support. For many of these articles, a bot would actually be preferrable (lower error rate, bot flag). It shouldn't be an extra hurdle, rather an encouragement to do it right, as Ovinus, above. AKAF (talk) 06:44, 3 November 2022 (UTC)[reply]

Oppose amending WP:MASSCREATE

  1. Expanding WP:MASSCREATE to include article creation manually by humans is totally out of the scope of a bot policy page. Telling editors who want to do manual editing to go ask permission at a bot noticeboard based on a bot policy is going to make no damn sense at all. Any reasonable person's reaction would be... well, I'm not a bot, so why are you treating me like one? And they'd be right to react this way. Don't tell editors making manual edits to ask prior permission to do low risk things like create stubs. If they need to be deleted later, we can do that. Creating a policy of asking prior permission to edit goes against our fundamental guideline to be bold in editing. Steven Walling • talk 16:35, 12 October 2022 (UTC)[reply]
  2. A bot page clearly isn't the right place to do this, and there's no indication of where else an editor would ask to create an unspecified (25? Over how long?) number of articles. I have some sympathy for the idea behind the proposal, but am concerned that it's so open that it can be used to write something really very restrictive. I also have serious and significant concerns that this is question 17 - we're back at that stage of an RfC where people lose interest and we end up "passing" something that has wide ranging impacts that aren't properly considered. I could support the idea of taking the first point of this proposal and discussing it as a standalone proposal at relevant places where it will be widely publicised and can be properly considered by a larger number of interested editors. I'm not sure that the second point of this proposal is something I could support - it goes too far, seems unnecessary. and is pointing at a really obscure part of , err, an essay a procedural polcy. Blue Square Thing (talk) 06:02, 13 October 2022 (UTC)[reply]
  3. per Steven Walling. There is a substantial disconnect between bot policy and editors doing tasks manually. Rlendog (talk) 17:53, 13 October 2022 (UTC)[reply]
  4. Um, what? Why does MASSCREATE need sentences like "it is irrelevant whether high-speed or large-scale edits...are actually being performed by a bot, by a human assisted by a script, or even by a human without any programmatic assistance" or "However, merely editing quickly, particularly for a short time, is not by itself disruptive." I really wonder whether the supporters have read either WP:MASSCREATE (which ends with the words "WP:MEATBOT still applies") or WP:MEATBOT recently. Perhaps this proposal would benefit from suggesting exact changes, in the "change X to Y" format preferred for edit requests.
    The problem with point 2, of course, is getting editors to agree on what constitutes mass creation of articles. We do not agree. MASSCREATE suggests 25 to 50 articles, but we don't agree on whether that threshold is meant to be per day (likely, if you read the discussion those numbers come out of) or per lifetime. FOARP, looking at your other comments, you want to "Make getting consensus for creation prior to mass creation mandatory", but you say that there's no need to agree on what constitutes mass creation, because "we'll know it when we see it". However, you can't "see it" until after the editor has already mass-created the articles. Taken as a whole, your proposals amount to:
    1. Someone creates 20 articles.
    2. You "know it when you see it", and you declare that the editor has culpably mass-created articles.
    3. You demand that the editor time-travel to get permission before creating the articles.
    This is not a good plan. You're going to have to pick one or the other. Either you tell editors, in objective terms when they do (and don't) need to get permission in advance, or you don't get to complain later that they didn't know that they should have gotten permission in advance. Refuse to write down the rules and them blame them later for not guessing the secret rules in your head (or subjecting them to drama when multiple editors have different rules in their own heads) is unfair. WhatamIdoing (talk) 02:09, 14 October 2022 (UTC)[reply]
  5. Plumping here formally, per my comments below. I could support something broadly related to this, but we'd need some sort of an admin-understandable definition of "mass creation", and we need to sort out what is the most appropriate venue [if the noticeboard proposal does not succeed, I'd suggest wikiprojects, unless there's no active wikiproject, in which case the reliable sources noticeboard (as it comes down to whether the proposed source is good enough) or possibly the village pump, but not the bot noticeboard, it's offensive to good-faith editors]. Espresso Addict (talk) 04:11, 14 October 2022 (UTC)[reply]
  6. Much article creation seems quite repetitive but the trouble is that this might be done at a variety of levels. For example, recent FAs include Fort Vancouver Centennial half dollar and Hurricane Sergio which are part of a series of similar articles. You can't really tell whether such work is problematic until you've got a number of examples. If someone had to ask for permission before they had started, then they would tend to talk ambitiously and hopefully about their plans, which might sound plausible. And, even if their first efforts were weak, they might improve as they get familiar with the task. I think it's better to let people have a go per WP:BOLD and monitor what they do. If lots of their creations are getting flagged at NPP then that's the time to start an audit/review process. Andrew🐉(talk) 08:56, 14 October 2022 (UTC)[reply]
  7. Per Steven Walling and WhatamIdoing. The proposed amendments of the bot policy are outside the scope of the bot policy, and are not compatible with WP:BOLD. Point 1 of the proposal is not a mere clarification: WP:MEATBOT actually refers specifically to edits that are contrary to consensus, or that cause errors an attentive human would not make, which does not include all "mass-creation through repetitive editing by hand". James500 (talk) 20:07, 16 October 2022 (UTC)[reply]
  8. This flips WP:BOLD on its head. I've said elsewhere in this discussion that I do think that it's a good idea to discuss mass creation with other knowledgeable editors beforehand, but making prior affirmative consensus mandatory is a de facto ban on mass creation: those opposed to mass creation on principle will simply filibuster each request. – Joe (talk) 09:30, 17 October 2022 (UTC)[reply]
  9. Similar to Espresso Addict, I don't mind the spirit of this proposal, but it's too ill-defined for me to support it. There isn't a firm definition of what mass creation is yet, much less what constitutes a "highly similar" article, and I worry that this would just lead to arguments over what constitutes a mass creation before anything even got to a noticeboard. (I've seen editors argue that two-line species stubs aren't mass creations, and I've seen editors argue that articles with 20+ distinct sources are cookie-cutter; good luck reconciling those opinions.) And that's before we even get to the issue of what forum to discuss these at; the bot noticeboard isn't meant to handle human editing, any forum specifically for mass creations will attract regulars with blanket opinions on them, and review by wikiprojects can lead to issues with walled gardens and less active projects not reaching consensus. And if a mass creation proposal fails, does that leave the proposer with a de facto topic ban on creating new articles in a topic area, or can they still create new articles as long as they stay below some threshold or reach a certain level of quality? All details I'd like to see worked out before I can support. TheCatalyst31 Reaction•Creation 01:44, 18 October 2022 (UTC)[reply]
  10. Oppose, don't see the value in this.--Ortizesp (talk) 14:54, 20 October 2022 (UTC)[reply]
  11. Sets a bad precedent. What is next? Rapid minor edits conducted with the assistance of copy-paste and browser tabs? --Rschen7754 00:31, 21 October 2022 (UTC)[reply]
  12. Per the unresolved issues in my comment below. Implementing a new requirement while at the same time failing to find consensus [in this RfC] for what it should apply to (and why) doesn't seem like a good idea. :/ — Rhododendrites talk \\ 12:28, 24 October 2022 (UTC)[reply]
  13. Oppose. This doesn't address the problem. If someone is creating many articles - even short articles - a day that are appropriately referenced to multiple reliable sources, there is no problem and no need for them to ask permission. Rlendog (talk) 20:23, 24 October 2022 (UTC)[reply]
  14. Oppose Sorry, I'm just not seeing what is being solved here. I also share the concerns articulated above: To what is this meant to apply? Who says that a particular set of pages qualifies? Where is the discussion supposed to happen? Doesn't this invert the be bold! spirit? XOR'easter (talk) 17:53, 28 October 2022 (UTC)[reply]
  15. Oppose Don't see the need to quote pages that already are linked from other pages. --Jayron32 18:38, 28 October 2022 (UTC)[reply]
  16. Oppose. Does not fix the problem, which is mass creation of unencyclopedic stubs. This is a competence issue. Withdraw the right to create articles from those who misuse it, whether maliciously or incompetently. · · · Peter Southwood (talk): 18:00, 1 November 2022 (UTC)[reply]

Comments (amend WP:MASSCREATE)

Comments from FOARP (amend WP:MASSCREATE)

As a simple policy point, mass-creation by bot is probably actually better than mass-creation by hand, since it at least reduces the chance of human error and takes up less of the editor's time. At the very least there is no reason why one should be considered "bad" and the other "good". One of the editors in the previous ARBCOM case (I cannot remember who) had it very right when they said that if we really wanted to create the Turkish Mahalle articles that were the subject of the first big ANI about Lugnuts we could have simply contacted the Turkish government for their data directly and just run a bot to create the articles - there really was no reason at all to just go through a low-information database by hand to do it. FOARP (talk) 11:07, 11 October 2022 (UTC)[reply]

Espresso Addict - I tend to agree that BOTPOL is not the best place for this. Just like many of our other policies, MASSCREATE started at a particular place because it was the first place to actually tackle with the problem of mass creation (if I'm not mistaken, it was RAMBOT and Dr. Blofeld's mass-creation-through-Bot attempts), but it need not stay there and can easily be moved somewhere else or spun out as its own article - would you support this proposal with the proviso that this happens? I think the main point here is we avoid making radical changes, and instead develop on what's already there. FOARP (talk) 07:41, 12 October 2022 (UTC)[reply]

Espresso Addict - I'm afraid this whole thing about definitions - something that really should be allowed to develop organically and beaten out case-by-case - has become just a reason to oppose any proposal at all. The forum for such discussions can also be allowed to develop organically. Well-received serieses of articles are unlikely to have any problems clearing this bar, but repeated flouting of a required process in creating large amounts of low-quality stubs for which their is no consensus by hand would be a behavioural issue, just as much as if someone repeatedly tried to use a bot to do so. We can see that indeed WP:NASTRO already implemented something like this - based explicitly on WP:MASSCREATION - when they cleaned up minor planets and is appears to work just fine now. Oddly, they did not feel the need to create a specific definition to achieve this. FOARP (talk) 09:22, 12 October 2022 (UTC)[reply]

Honestly, really do not understand people opposing this out of supposed dire consequences. This is already policy and has been for a decade+, the proposed amendments are minor and for clarity. Particular opposing "until there is a definition" is something I don't understand - MASSCREATE includes a definition, whatever you think about it, it is already policy. FOARP (talk) 15:34, 14 October 2022 (UTC)[reply]

  • TheCatalyst31 - "what happens to a proposer if their proposal for a mass creation fails? Are they de facto topic-banned from creating new articles in a topic area..." - This is the kind of catastrophising I referred to on the talk page. I'm not sure how you arrived at the conclusion that this proposal could ever have had this outcome when the present policy does not have this outcome on anyone who fails to get consensus for what they are planning to do. What happens if someone who seeks to get consensus for mass-creating articles fails to get consensus for mass-creating them in a particular way is they cannot proceed with mass-creation in the way they proposed, this does not apply to article creation outside of that (though other policies likely will). So in an example:
  • Mass creator: "can I go through this context-free database of 10000 single-line entries with no SIGCOV and turn every one of them into an article for which there are not additional sources?"
  • Community: "No, LOL"
  • Mass creator: *creates one or two articles as they proposed*
  • Community: "This doesn't break the consensus about mass-creation but the articles are likely garbage and going to get AFD'd if no additional sources can be found. Thank you for playing." FOARP (talk) 12:39, 18 October 2022 (UTC)[reply]
Comments from Rhododendrites (amend WP:MASSCREATE)

Adding "also this applies to manual editing" gets at the main reason efforts to point people to those policies don't work, but there are still unresolved issues: apart from the bot policy not seeming like the right place to talk about manual editing (it should have a pointer elsewhere IMO), it doesn't sufficiently define what sort of manual editing it applies to. There's the number "25 or 50" but the language around it is weak, and it never defines "25 or 50 what" (i.e. over what span of time, with what similar characteristics, etc.). Point 2 is, I think, taken for granted throughout this RfC -- the areas where we're having trouble agreeing is under what conditions it should apply. — Rhododendrites talk \\ 13:05, 11 October 2022 (UTC)[reply]

Comments from Espresso Addict (amend WP:MASSCREATE)

The bot noticeboard isn't the right place to point people at. Espresso Addict (talk) 23:35, 11 October 2022 (UTC)[reply]

FOARP: If we can agree any kind of usable definition for mass creation, a suitable venue for pre-approving (I'd be more in favour of relevant wikiprojects than a noticeboard), and a suitable place for the advisory message to be placed (not relating to bots), then I'm not strongly opposed.
What would you envisage doing with ongoing series of articles? Eg Peter I. Vardy has been creating lists of English listed buildings for over a decade now, which seem generally well received; would he have to ask for permission to continue? ETA: I'm worried that valuable contributors doing the spadework of building the encyclopedia are going to feel unappreciated, while editors who create poor-quality stubs and don't tend them are unlikely to read the guidelines and seek advance permission. Espresso Addict (talk) 08:46, 12 October 2022 (UTC)[reply]

BilledMammal: I would love it if someone were to be able to codify the difference between "prolific creation and mass-creation". Espresso Addict (talk) 03:57, 13 October 2022 (UTC)[reply]

BilledMammal: "more than ten highly similar articles" -- that's not really definitive. I plan to create >10 articles on listed buildings based on broadly overlapping sources, which will all probably turn out to be fairly similar, over the coming months; I anticipate it being 100% uncontroversial. The problem is the risk of the good-faith content creator being put off and walking away from the project -- "the encyclopedia that anyone can edit as long as they ask permission first" isn't the slogan that I signed up for. I suggest wikiprojects because they have relevant expertise. Village pump is a bit of a free for all; the average commenter is likely to know a great deal less than I do about the available sourcing for English listed buildings. If there's a problem with the sports wikiprojects being accepting of articles that the broader community rejects, that seems to come down to conduct issues on the part of the wikiproject members knowingly going against the community consensus. Espresso Addict (talk) 04:35, 14 October 2022 (UTC)[reply]

TheCatalyst31 "what happens to a proposer if their proposal for a mass creation fails? Are they de facto topic-banned from creating new articles in a topic area..." Excellent question. Espresso Addict (talk) 04:06, 18 October 2022 (UTC)[reply]

Comments from ActivelyDisinterested (amend WP:MASSCREATE)

I'd agree the BOT board isn't the right place for discussion, all mass creations should be discussed at the new board (as I see it that would be the primary concern of the new board). -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 15:29, 12 October 2022 (UTC)[reply]

Comments from Lurking Shadow (amend WP:MASSCREATE)

The bot noticeboard is the right place for these disputes if the other noticeboard doesn't form. If you are creating articles on scale then there is some routine in the process. Bot-like editing can be addressed in a similar fashion to bot editing, only easier because you can ask the "bot" what it does directly.— Preceding unsigned comment added by Lurking shadow (talk • contribs) 16:17, 12 October 2022 (UTC)[reply]

@Steven Walling:The page WP:BOLD also includes WP:CAREFUL, and mass-creation is one of these actions that is difficult to fix if it includes some articles that shouldn't get deleted. Mass-creation of acceptable articles is not a problem. Mass-creation of only speedily deletable articles can also be easily reversed. But mass-creation of articles with some articles that are deletable and some that are not... that's the real problem.Lurking shadow (talk) 04:00, 13 October 2022 (UTC)[reply]

Comments from TheCatalyst31 (amend WP:MASSCREATE)

Would this proposal only apply to repetitive, bot-like mass creation (e.g. cookie-cutter stubs with a database source), or would it affect all large-scale article creation? There are editors who write many distinct articles on similar topics in a way that a bot couldn't possibly replicate, and it doesn't make sense to make them ask for approval on a bot noticeboard. TheCatalyst31 Reaction•Creation 23:27, 12 October 2022 (UTC)[reply]

I brought this up in my !vote above, but I think it's worth asking separately: what happens to a proposer if their proposal for a mass creation fails? Are they de facto topic-banned from creating new articles in a topic area, or are they still allowed to create them as long as they aren't sufficiently similar to their past work? TheCatalyst31 Reaction•Creation 01:48, 18 October 2022 (UTC)[reply]
@FOARP: While there's an obvious difference if we're talking about 1-2 articles vs. 10,000, WP:MASSCREATE says 25-50 articles can be a mass creation, and with those numbers creating 2 articles would get you 4-8% of the way to your goal. (Hopefully Q3B would give us a better definition down the road, but that's what it says for now.) While I think most supporters of this policy are concerned with getting mass creators to slow down and put more time and effort into their articles, I want to make sure this won't be gamed by people who want to block creation of certain categories of articles. TheCatalyst31 Reaction•Creation 03:49, 19 October 2022 (UTC)[reply]
Comments from JoelleJay (amend WP:MASSCREATE)

I think the existing guidance at WP:NASTRO on mass creation may be a good framework to base this proposal off of. JoelleJay (talk) 00:27, 13 October 2022 (UTC)[reply]

Comments from BilledMammal (amend WP:MASSCREATE)

I would suggest WP:VPR as the correct location to request approval for mass creation; it has been the usual place for such discussions in the past.

@Espresso Addict and TheCatalyst31: It shouldn't affect editors like Peter I. Vardy, as there is a difference between prolific creation and mass-creation; their edits fall into the former category. BilledMammal (talk) 03:47, 13 October 2022 (UTC)[reply]

@Espresso Addict: Prolific is creating a large number of articles; mass-creation is creating a large number of highly similar articles. Peter I. Vardy's articles aren't highly similar, so they haven't engaged in mass-creation. BilledMammal (talk) 04:03, 13 October 2022 (UTC)[reply]
@Espresso Addict: Could you support Editors creating more than ten highly similar articles are required to obtain prior consensus to do so at the village pump? I would oppose using WikiProjects, as they are often not representative of the broader community; most sports WikiProjects would approve the mass creation of sports biographies, while the broader community would reject such mass creations. BilledMammal (talk) 04:15, 14 October 2022 (UTC)[reply]
@Espresso Addict: "Fairly similar" is a far lower standard than "highly similar". As for your articles on listed buildings, if they are going to be like your most recent creation, 31 Madingley Road, it isn't going to be possible for them to be "highly similar". BilledMammal (talk) 04:49, 14 October 2022 (UTC)[reply]
@Blue Square Thing: WP:MASSCREATE is a policy, not an essay. BilledMammal (talk) 06:21, 13 October 2022 (UTC)[reply]

@WhatamIdoing: You don't demand that the editor time-travel to get permission before creating the articles, you merely require that they get consensus before creating more. BilledMammal (talk) 02:12, 14 October 2022 (UTC)[reply]

Comments from Blue Square Thing (amend WP:MASSCREATE)

@BilledMammal: thank you - got too confused between different pages; is it meatbot that's part of an essay then? Blue Square Thing (talk) 06:31, 13 October 2022 (UTC)[reply]

Comments from WhatamIdoing (amend WP:MASSCREATE)

User:BilledMammal, if you want an "interrupt and discuss" option, then you need to re-write your proposal. You could say something like "If another editor claims you are improperly creating too many articles, then we'll ask you to pause for a while, during which we'll have a consensus-driven discussion about whether your actions are created".

Also, in your definition here (despite opposing having any definition at all above), you need to specify whether that's "more than ten highly similar articles per day" or "more than ten highly similar articles per lifetime" or something in between. Some editors have already been here for 20 years, and I cannot imagine the community agreeing that creating one article every two years is "mass creation". Also, if you look at the normal and correct way to handle some subjects, "highly similar" is apparently a goal. If you're following Wikipedia:WikiProject Albums/Album article style advice, then one album stub is going to look pretty much like any other album stub.

User:David Fuchs, I do not believe that the WP:AWB folks will agree with your assertion that doing "any large-scale series of edits" should be discussed in advance. WhatamIdoing (talk) 15:33, 14 October 2022 (UTC)[reply]

FOARP, the problem with MASSCREATE is that it is open to interpretation. Specifically, this text:
Any large-scale automated or semi-automated content page creation task must be approved at Wikipedia:Bots/Requests for approval. [...] While no specific definition of "large-scale" was decided, a suggestion of "anything more than 25 or 50" was not opposed.
Leaving aside the widely ignored "semi-automated" requirement, the existing statement has been interpreted in two different, incompatible ways:
  1. If you plan to create "anything more than 25 or 50" articles per day, your task must be approved in advance.
  2. If you plan to create "anything more than 25 or 50" articles per lifetime, your task must be approved in advance.
Under the first interpretation, #100wikidays is 'legal', even if you are obviously creating 100 articles in 100 days from an alphabetized list and writing down the same three facts about each subject. Under the second interpretation, many of the editors involved in #100wikidays have violated the bot policy. MASSCREATE has most of a definition. It does not have a definition that is clear enough that editors can see when it applies, and when it doesn't. We need one that is so clear, so devoid of opportunities for differing judgments, that editors cannot claim MASSCREATE applies when they desperately want it to – and vice versa. WhatamIdoing (talk) 22:50, 30 October 2022 (UTC)[reply]
Comments from Steven Walling (amend WP:MASSCREATE)

Reply moved from #Comments from Lurking Shadow (amend WP:MASSCREATE) –MJL ‐Talk‐☖ 20:16, 1 November 2022 (UTC)[reply]

@Lurking shadow: Either nuking select articles or writing an adminbot is there as you say, and if they aren't nuke candidates, then fixing stuff over time is fine, since perfection isn't required and Wikipedia is a work in progress. What's unfixable is discouraging people from ever creating missing articles in the first place, because we added rules where you have to ask permission before creating more than 10 articles on a subject or whatever. Steven Walling • talk 02:51, 15 October 2022 (UTC)[reply]
Comments from Rlendog (amend WP:MASSCREATE)

User:FOARP I am not sure what you are referring to when you say "Particular opposing "until there is a definition" is something I don't understand - MASSCREATE includes a definition, whatever you think about it, it is already policy." MASSCREATE says "While no specific definition of "large-scale" was decided, a suggestion of "anything more than 25 or 50" was not opposed." (emphasis added) Before we can apply any restrictions to manual mass creation we need to have a consensus definition of what mass creation means. Rlendog (talk) 23:25, 29 October 2022 (UTC)[reply]

Comments from Editor X (amend WP:MASSCREATE)

Please open your own section with username in the heading. Please limit comments within a section to 300 words.

Question 18: Expanding the powers of WP:BRFA

Proposed: The powers of WP:BRFA is hereby expanded to cover mass actions that could reasonably construed to meet WP:MEATBOT concerns. Specifically, permissions for mass actions in general, such as (but not limited to) mass creations and mass edits, should be requested there before being allowed to proceed. The Community remains empowered to define terms such as "mass creation" or "mass edits" should they wish to do so.

Support (expansion of BRFA powers)

  1. As proposer. I have been involved in a mass edit event. Since such a proposal would force mass actions to go through BRFA first, this can ensure consensus for a potential mass action. Thanks, NotReallySoroka (talk) 02:37, 30 October 2022 (UTC)[reply]

Oppose (expansion of BRFA powers)

  1. The bot approval process is not the place to deal with actions which are manual, even if they do meet MEATBOT definitions. I'm not sure that having an open ended "any mass edit" is awfully helpful - does this mean that any time that someone goes through and deals with dodgy categorisations that they'd need to ask permission first? It seems to me to be more likely that a series of mass edits on, say, UK place articles (such as the addition of historic counties - very much a divisive issue fwiw), would be best discussed at the appropriate wikiprojects where people with an interest in the content might see it, rather than at a notice board where people interested only in process are likely to see it. Blue Square Thing (talk) 08:57, 30 October 2022 (UTC)[reply]
  2. As I said above this could have unforeseen consequences. Plus, the BAG is chosen for their technical abilities and this would be a massive expansion of their duties as well as skills required. --Rschen7754 23:45, 30 October 2022 (UTC)[reply]
  3. Bots are bots, people are people. People should not need to ask permission to work hard at Wikipedia. They should be encouraged to do so, not given a wall of bureaucracy. --Jayron32 11:46, 31 October 2022 (UTC)[reply]
  4. Just no, per all above. Espresso Addict (talk) 22:45, 31 October 2022 (UTC)[reply]
  5. The bot policy has a section about the bot approvals group. The members of the group are experienced in writing and running bots and have programming experience. This isn't relevant to article creation at scale by humans. ~ ToBeFree (talk) 08:14, 1 November 2022 (UTC)[reply]
  6. What they said above. · · · Peter Southwood (talk): 17:50, 1 November 2022 (UTC)[reply]
  7. BRFA is not the place to have this. BAG members are selected for their skill in making bots, programming and checking for automated bot issues. They are not selected to authorise mass edits by people, which IMO is a different ball game. Dreamy Jazz talk to me | my contributions 11:44, 3 November 2022 (UTC)[reply]

Comments (expansion of BRFA powers)

  • I kind of would like to know what the BRFA regulars think about this. Are they interested in this work, or would it feel like we're foisting difficult problems off onto them? WhatamIdoing (talk) 22:54, 30 October 2022 (UTC)[reply]
    Can't speak for all BAG members, but I doubt there'd be interest to do this. BRFA only has a few active BAG members processing requests at a given time. I think very few BAG members would wish to assess whether consensus exists for a mass editing job by a person, especially as some of these will be contentious I imagine. The result would probably just be severe delays, if the requests were processed at all.
    In general (regarding the various BRFA/bot proposals on this page): It's convenient to see WP:MASSCREATE and WP:MEATBOT applied to article creation at scale, and while I do think those policies apply, I think it's ineffective to tinker with the bot policy or bot processes to try solve this problem. Firsly, because content editors don't read the bot policy or care about it. Secondly, because quoting the bot policy has typically lacked muster in these kinds of WP:ANI cases, even though the text is already clear. A clear guideline needs to be made and put into a content policy if it's to have any hope of success, and there needs to be quick action at WP:ANI, including effective action against the violating content. Without appropriate enforcement measures, the policy will be toothless. The bot policy has enforcement measures for bots (i.e. block the violating bot), which gives it standing in the context of bots, but these block rationales often aren't applied to human editors engaging in similar behaviours. ProcrastinatingReader (talk) 02:06, 3 November 2022 (UTC)[reply]

Further proposals

Please bring up further proposals on talk, with a ping to the nominators.

Discussion

Comments from ActivelyDisinterested

As per my comment on question 5, it could be useful to have some limitation with WP:BLPs. Personally I wonder whether allowing it at all is a good idea. But I'm unsure how to word any proposal correctly. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 19:00, 4 October 2022 (UTC)[reply]

Comments from JoelleJay

I brought this up in the earlier discussions, but given the difficulty in defining "mass creation", why don't we just have a limit on how many undersourced new articles a single person can have in their contributions (from X date)? If they go over that limit, they get escalating warnings up to a p-block from page creation, which can only be removed if they adequately source N other non-compliant articles. So it would be fine to create 100 microstubs if they included GNG sourcing from the start. And it wouldn't matter if other people expanded their articles after creation: the whole point is to discourage generating standalones for which other editors end up needing to find notability-confirming sources. 22:19, 6 October 2022 (UTC)

I don't know anything about technical feasibility in tracking the state of article sourcing at creation, but if it's doable I could add it to the proposals. "Undersourced" could be defined by AfC norms or hashed out more fully later. I intended for the counter to start only after extensive global notification across some period of time, so it wouldn't apply to old articles. Adding sufficient sourcing to other editors' articles would also be an option for offsetting one's own unexpandable-microstub footprint, it just would be harder to keep track of it. 21:01, 7 October 2022 (UTC)
Also agreed with FOARP that GEOLAND is huge mass-creation fodder and should be covered. 21:22, 7 October 2022 (UTC)

Comments from Espresso Addict

@JoelleJay: This is an interesting suggestion; I'd like further details on how it might work in practice. Is this technically feasible? What about articles which the contributor creates with sources but someone else removes the sources (this is reasonably common, someone "improves" the article by writing over it, losing all sourcing). Would you get points for sourcing other people's work? Anything that persuaded editors to work on the un(der)sourced articles heap would be positive. How is undersourced defined? My understanding of the way this RfC works is that if this is to be a proposal, it needs to be formally started in the next few days. ETA: Might want to include a date threshold; I think there might need to be a "statute of limitations" for very old contributions before referencing standards were anything like today's standards and when online sources were much more sparse. Espresso Addict (talk) 00:04, 7 October 2022 (UTC)[reply]

@Aquillion: Well, I tried to put forward proposals but both of them have/are failing. I don't personally feel this discussion format is at all helpful in bringing people together to find a solution to the undoubted problem. I wish I'd been able to participate in the brainstorming discussions but they happened when I was sufficiently injured to be unable to type. Espresso Addict (talk) 02:07, 8 October 2022 (UTC)[reply]

@FOARP: I, and perhaps others, would be a great deal happier with proposals that relate to populated settlements under GEOLAND, rather the entire guideline, which also covers heritage-listed buildings, mountains, and similar. Espresso Addict (talk) 01:12, 10 October 2022 (UTC)[reply]

@FOARP: I had a look at Zykov Glacier and there is a little online about it beyond the existing text. After the expedition that photographed it there seem to have been at least two published independent studies of the rock there, which are reviewed in a handful of papers. A reader encountering Zykov Glacier in this literature might want to know where it was exactly. Someone who actually understood what the papers are talking about might be able to improve the article further. Espresso Addict (talk) 23:39, 10 October 2022 (UTC)[reply]

A lot of this is beginning to seem more like a conduct issue to me. Is it not the case that administrators are already empowered to tell someone that what they're doing is disruptive because they are flooding NPP with articles based on an inadequate/unreliable source, and that they should slow down/stop and form consensus to continue at an appropriate venue? Espresso Addict (talk) 04:20, 14 October 2022 (UTC)[reply]

Comments from Dlthewave

@JoelleJay: This seems like a great idea - Besides removing the open-ended "someone will improve it in the future" assumption, it will limit the problem of articles that can't be improved because coverage doesn't exist, which are currently a massive time sink. Even if we have to tweak the wording, this seems like something that the community might support as it doesn't encroach on article creation quite as much as the GNG-sourcing-from-the-start requirement. Would love to see this added to the RfC. –dlthewave ☎ 02:09, 7 October 2022 (UTC)[reply]

Comments from FOARP

I am very concerned that people seem to not appreciate the extent to which mass-creation in the GEOLAND area is a really big problem, at least as big as that which was in NSPORTS. Exempting and excluding GEOLAND from a lot of the proposals above (why?) means there will essentially be no fix for the problems we are addressing. FOARP (talk) 08:40, 7 October 2022 (UTC)[reply]

@Valereee: - My GEOLAND-specific fix is that any fix for mass-creation also applies to GEOLAND articles. There is nothing special about GEOLAND articles that requires a carve-out for them. Going through a half-understood database and translating every line of it into its own article is as much a problem there as it is everywhere else. This whole thing started with Lugnuts mass-creating Turkish “village” articles that were likely neighbourhoods or individual farms. FOARP (talk) 14:33, 9 October 2022 (UTC)[reply]

@Valereee: - Having looked at this, I think this is better stated as a simple oppose !vote for proposals that (inexplicably) contain carve-outs for subject-areas known to be heavily affected by mass-creation. Beyond saying that Agro-Industry Complex and Harry Oppenheim were both undesirable mass-creations, what more is to be said? Instead I will just propose something based on an already existing policy that already applies to all subject-areas anyway. FOARP (talk) 10:32, 11 October 2022 (UTC)[reply]

Espresso Addict - I cannot see how the rash of articles about geographical features in Antarctica mass-created in 2006 literally by word-for-word copying of single-sentence GNIS database entries (e.g., this one) were much better than the Iranian villages articles mass-created at the same time. By any definition these articles lack significant coverage and are never likely to be expanded because they are about features where no-one lives and which are almost never visited except fleetingly. The are straight NGEO-fails given that the only information about them is name and location. The one thing I will grant them is at least they are not simply the same single-sentence template repeated over and over, but this is because the people who created the GNIS data did not do that. FOARP (talk) 07:37, 10 October 2022 (UTC)[reply]

Espresso Addict - Passing mentions in a long list of places that samples were collected in is not anything that will add content to an article. The same thing was true for minor planets (i.e., their observations are regularly recorded and many are regularly mentioned fleetingly in long lists) but we sensibly weeded out all the ones without SIGCOV years ago. FOARP (talk) 08:00, 11 October 2022 (UTC)[reply]

@Valereee, 127(point)0(point)0(point)1, and 127(point)0(point)0(point)1: - Engagement is lower than you might expect partly because the title is very anodyne and doesn't really say what it's about. This subject does cause major drama so I expect if we had a title that was closer to what it was actually about (what to do about mass creation of articles?) more people would be clicking through from CENT. FOARP (talk) 14:34, 11 October 2022 (UTC)[reply]

@AKAF - You are right that WP:MASSCREATE already exists - sharpening it up a bit is the subject of question 17. FOARP (talk) 10:47, 12 October 2022 (UTC)[reply]

Comments from Aquillion

I'm noticing at least a few people seem to be opposing virtually all proposals. Is this because they do not believe there is a problem, or because they believe that none of these are the correct solution to the problem? If it's the latter, do they have proposals of their own? --Aquillion (talk) 18:23, 7 October 2022 (UTC)[reply]

Comments from XOR'easter

This whole discussion seems to have come unmoored from any specifics, in a way that creates an illusory unity of problems. Are we worried about copyright violations? Inaccurate information? The inclusion of accurate information in chunks that might be too small? I don't think "article creation at scale" is one thing, and so I don't think there can be a single rulebook for it, or a single remedy for it going wrong. XOR'easter (talk) 19:08, 7 October 2022 (UTC)[reply]

Hut 8.5, yes, I think that many people are just used to Wikipedia having lots of tiny articles about obscure things. I'm a bit surprised that there hasn't been a proposal along the lines of, "Before you create a lot of stubs, get community input on whether the source you are using is reliable." XOR'easter (talk) 14:03, 9 October 2022 (UTC)[reply]

Comments from Blue Square Thing

@Aquillion: I seem to have found myself opposing almost all of the proposals. I didn't expect this to be the case - I expected to support some. But the arguments put forward by other people have led me to oppose much more frequently than I expected to. I'm not completely convinced that there is that much of a problem here than can't be dealt with in other ways; if there are other solutions I'm afraid that I'm not clever enough to see them just now. Blue Square Thing (talk) 08:51, 8 October 2022 (UTC)[reply]

Comments from Valereee

Espresso Addict, we'll be workshopping the AfD RfC soon at WT:ADAS (and are discussing the timeline for that now; we are considering starting workshopping as soon as things significantly slow here). If none of the proposed solutions here gains consensus, that RfC will by necessity still need to cover any mass creation issues that affect AfD at scale. We ran this first to see if the community could find solutions to this one major problem, because if it could be solved at least partially here, that RfC could build on those solutions.

XOR'easter, we decided to do this RfC first not because creation at scale was the only thing affecting dysfunction at AfD but because there was a strong feeling that it was a major contributor, and if we could find solutions here before the RfC about deletions at scale, it might actually make that RfC less of a train wreck. Valereee (talk) 14:43, 8 October 2022 (UTC)[reply]

FOARP, can you propose a GEOLAND-specific fix? Valereee (talk) 14:48, 8 October 2022 (UTC)[reply]

@FOARP, can you create a proposal for that? Valereee (talk) 14:40, 9 October 2022 (UTC)[reply]

Ingratis, I'm not going to prevent something being proposed that is related to mass creation, if someone can come up with a proposal. The RfC has been announced in multiple places, including WT:N. Valereee (talk) 12:04, 10 October 2022 (UTC)[reply]

@127(point)0(point)0(point)1, the RfC has been announced at the articles for deletion talk page, the Arbitration Noticeboard, the administrators' noticeboard, the Bot policy talk page, Village pump (policy), Wikipedia talk:Notability and Centralized discussion. Some of those noticeboards have thousands of watchers. Valereee (talk) 14:17, 11 October 2022 (UTC)[reply]

@127(point)0(point)0(point)1, we can't actually force people to be interested or understand. @FOARP, you really think it's the name? Don't answer here, let's take it to talk. Valereee (talk) 14:51, 11 October 2022 (UTC)[reply]

Comments from 127

I'm rather surprised by the participation numbers here. Problems of this scale and discussions of reworking fundamental processes and the highest number of editors in any one counting stack response is... 13 at time of writing. There are more well populated ANI threads. I know there's more time to run, but at the rate there are new voices coming in, I don't think we're at a place we can realistically proclaim a consensus for or against. This RfC needs more visibility. --WhoIs 127.0.0.1 ping/loopback 20:56, 8 October 2022 (UTC)[reply]


@User:Valereee I appreciate that it's been announced in all those places, and don't wan't this to be framed as some sort of lack of effort. But that still doesn't change the fact that we don't seem to have the participation we need for the kind of changes we're proposing here. Either those watching eyes aren't coming and aren't interested or they don't understand the scale of the RfC. -WhoIs 127.0.0.1 ping/loopback 14:36, 11 October 2022 (UTC)[reply]

Comments from Hut 8.5

I get the impression that this is an issue a small number of editors care about a lot but where the wider community doesn't care. That explains the low particpation and the fact that none of the proposals is doing very well. The largest problem I've seen in this area is with people creating lots of articles on populated places that don't exist by citing sources that aren't reliable or don't really indicate that something is a populated place. The proposals here either wouldn't address this or are far too heavyweight. Hut 8.5 10:50, 9 October 2022 (UTC)[reply]

Comments from Editor AKAF

This RFC seems a lot of faff to codify the current community position:

  • Creation of "high quality" articles (whatever that means) is a net positive of which pretty much everyone approves.
  • Creation of articles in a "bot-like manner", including manually populating templates from a database, is something to which editors appear indifferent or negative.
  • A small percentage of articles created in a "bot-like manner" have been expanded to be "high quality" articles, and everybody likes these articles.

For bot-like edits, there is an existing route to approval (WP:BRFA), which could be enforced. AKAF (talk) 07:25, 10 October 2022 (UTC)[reply]

Comments from Ingratis

@Valereee I'm very concerned at the apparent danger that this RfC may be used as backdoor leverage on the issues surrounding GEOLAND. As you will be aware, there is a long and rather vicious history of many attempts to change it which have met with much strong objection. Because of the complexity of the issues they need to continue to be discussed fully at the SNG itself, not tacked onto this RfC, the context of which will certainly slant the discussion, so it seems to me inappropriate for any GEOLAND reform proposal to be put forward here. Ingratis (talk) 11:28, 10 October 2022 (UTC)[reply]

@Valereee I understand that of course the focus here is on mass creation but there are many more issues involved in GEOLAND than that. Any proposal to draw its teeth as an SNG by removing its present criteria and making it subject to GNG (as advocated by for example FOARP) must surely require a separate RfC, which would doubtless attract a lot of participation, since there are many editors who have very firm views on whether GEOLAND should go the way of NSPORTS. Ingratis (talk) 08:19, 12 October 2022 (UTC)[reply]

Comments from Rhododendrites

I wish everyone responding to various proposals here with "that will just be wikilawyered" or "this will just be gamed" will consider whether the potential for wikilawyering/gaming would be greater if we have some of those basic rules/guidelines in place than the current situation. It seems like so many things are failing for two big reasons: this, and because there's a divide as to whether this is all about quality or all about quantity. On the latter, I don't know how you can say which creation projects the quality requirements would apply to without getting into quantity ("all new articles must have X sourcing/claims/whatever" is outside the scope of this RfC after all, so to what will it apply?). — Rhododendrites talk \\ 12:16, 10 October 2022 (UTC)[reply]

Comments from LessHeard vanU

Anything that makes the job of admins, and Arbcom, easier in determining whether someone is not editing to the benefit of the encyclopedia is to be commended. The truth behind the fact that not a lot of the editorship cares about the issues around mass article creation is that they are not required to improve or even consider low quality content (and why should they, we need every good editor to be happy!). Those who care need a base from which they can address the issues and present them for necessary action. I support the purpose of this rfc. LessHeard vanU (talk) 16:26, 13 October 2022 (UTC)[reply]

I am impressed by the reasoning of @Mangoe: below, especially their last point - where there is a singular source used in a batch of mass creation, that source needs investigation. This RfC might address how a contributor assessing a batch of edits might review the sources and how they may communicate their concerns to either a board or initially the original contributor. Certainly, mass creation of articles from a dubious source may be quickly deleted even if there may be better sources simply because a poor source on notability devalues the encyclopedia. Better to get rid en-masse and then rescue those which can be attributed to a better clain of notability at some future date. An editor who returns to a poor source, or recreates the same articles from it, after such articles are removed can be introduced to the WP:CLUESTICK. LessHeard vanU (talk) 16:06, 15 October 2022 (UTC)[reply]
@@WhatamIdoing: Eh? We are talking mass creation of articles - in this proposition, a quantity from one poor source - so your example of one article having the source clicked perhaps 3 times a year is multiplied by the number of articles; dozens, scores or even hundreds of articles. 100 -/+ clicks to a link that is little more than a list or an index provides no indication of the notability of an article. Anyone with an interest in the individual subject might think that it deserves a better source so they can confirm the article content - they may indeed find and link that better source, but they may have done that in creating an article of a subject they are interested in. Mass creation from a bare list does not provide evidence of notability, even when the subject may be notable. LessHeard vanU (talk) 09:33, 31 October 2022 (UTC)[reply]
@WhatamIdoing: I am going to make one final comment, as I think we are not talking about the same issue, and you may respond as you please. I am talking mass creation of articles, using a poor source, which creates far more work for patrollers, admins, and editors, than any benefit to the project; anyone looking for the WhatamIdoing Fine Dining Establishment may find it as it is created from a list called "Food Venues Using Names of Wikipedians" hosted on some obscure site on the internet, but may be better served by a source that is taken from a review publication of some repute. If it has an entry on the latter, then notability is easily determined. If the article does not yet exist, then someone can create it - hopefully using the good source - as a single one time subject. This improves the encyclopedia, as it takes less effort to review and add to than initially sorting the wheat from the chaff. This what I intend by my two ending sentences of my 15th October post. LessHeard vanU (talk) 16:09, 31 October 2022 (UTC)[reply]

Comments from Ovinus

My worry from the preparatory RfC—that we don't have enough examples and data—grows stronger. Of the controversial mass creations/creators I can recall, we have 4 mentions of Carlossuarez, 7 of Lugnuts, and... that's it, according to Cmd-F. I feel I cannot make an informed decision without understanding how these proposed guidelines or solutions would have applied were they in existence five years ago. Ovinus (talk) 01:16, 15 October 2022 (UTC)[reply]

Comments from Mangoe

First, @Hut 8.5:: I'm not terribly concerned that the "greater community" is ignorant of the issue, if indeed they are. They do not set the standards: comparable works, and the writing of academic works set the standards. And besides, the thing is so huge at this point that it is unsurprising that any given person is unaware of some aspect of the problem. I'm only aware of the issues because I follow AfD and have become active in geographic deletions; I am somewhat aware WRT to Olympic competitors, but beyond that I just don't have the time to follow further.

Moving on to the issues:

  • Personally, I'm coming to the opinion that creating stubs at all is bad. I've made a very few myself, but I don't think I will ever do so again, especially if I'm not going to come back later and flesh the article out. There's just too much of a distance between the satisfaction of creating an article and that of expanding someone else's article.
  • The second big issue is that mass creation from a source requires a supreme level of confidence in that source. In practice what has happened is that it takes time to realize that a source is not as good as people assume. GNIS is the poster child for this: one would think that a database from the official geographers of the country would be as reliable as possible, but in practice it took a lot of examination to expose its fallibility. It's unlikely that a review way back when would have exposed this before large numbers of stubs were created.
  • The other substantial problem, as @FOARP: has pointed out, is that the subject guidelines are quite problematic. Again, I'm going to pick on GEOLAND, which I find, in its history, was not the result of an extensive discussion among many parties, but which was written, in pieces, by individuals, with no prior agreement. The understanding that they trump GNG, which is widely held, has meant many, many fights over notability. But the subtler issue is that these guidelines lead article creators to misclassify article subjects so as to fall under the aegis of a guideline— not necessarily on purpose, but because it is the path of least resistance. Therefore we have spent years trying to get rid of articles on nonexistent towns, er "unincorporated communities", and we had the recent cleanup in which a bunch of Iranian "towns" were imagined out of what turned out to ad hoc descriptions of census tracts.

So I'm not keen on mass creation. But if it has to happen, the articles and the process need to be held to very high standards. And the solution for mass errors has to be mass clean up, not the laborious process we're being put through now. Mangoe (talk) 03:56, 15 October 2022 (UTC)[reply]

@Ovinus: Carlos Suarez wasn't the only person who made a huge run of stub place articles; he was simply more egregious about it. From what I can tell all town/"unincorporated community" articles were stubbed out from GNIS by a variety of persons, and so far I've only found one state (Oregon) where there was a concerted effort to expand the stubs and clear out dubious entries. Mangoe (talk) 02:20, 17 October 2022 (UTC)[reply]

Comments from Scope creep

I saw two news articles today that are in conflict. They were Why Mastering Language Is So Difficult For AI, where it states automated tooling with AI is currently a form of plagiarism. The second article is [3] where a student is using it to create her homework. While I think we are on the cusp, its very early days. scope_creepTalk 21:36, 17 October 2022 (UTC)[reply]

Moved reply from Atsme to own section. –MJL ‐Talk‐☖ 20:09, 1 November 2022 (UTC)[reply]
I hadn't seen that but it was really interesting. It was kind of thing I was thought was going to happen in the future but has now been formalised in the present. It seems to be a first generation attempt to go beyond the static model found on Wikipedia and is likely the first of many. His criteria for inclusion makes it similar to a search engine, but gives you an idea what is possible now. Unfortunately the software we have on here is too decript to even to begin use any of that. I suspect increasingly we are going to be seen as monks beavering away in the monastery while the Gutenberg revolution happens elsewhere, but its early days. scope_creepTalk 08:51, 31 October 2022 (UTC)[reply]
I question what it will actually take us as group, to actually get off this software. The software environment is negating our capability over the long term of creating better and higher quality articles and more of them. It may be questionable saying better, but if we were on a platform that enabled us to quikly have deeper insight into certain sources using automation, I'm sure it would be used to create better articles within policy and consensus. These kinds of questions need to be asked. When is it going to happen. Are we still going to be on this software in 2030 or 2040. There seems be no plan by the WMF with its head in the sand and that is reflected in this software which unbelievably decrepit. At the moment, any kind of automation is going to be done off-wikipedia, outwith our control. So at some point we are have to say, when are going to move off this platform and how are we going to do it. scope_creepTalk 09:37, 31 October 2022 (UTC)[reply]

Comments from User: Stuartyeates

Much of the discussion is about mass creation of articles in a scripted fashion from databases with a few facts about a very large number of items. Wikimedia already has infrastructure to support databases with a few facts about a very large number of items, called wikidata. Maybe what we need is a solution involving people putting data into wikidata and that being displayed via a scripted template in wikipedia which isn't actually created as a fully expanded article until human comes along to end it? Stuartyeates (talk) 00:24, 21 October 2022 (UTC)[reply]

Comments from WhatamIdoing

LessHeard vanU said: "a poor source on notability devalues the encyclopedia."

How? Like, literally, how does this happen? What's the mechanism you're hypothesizing here?

Imagine that I create an article on something you've never heard of, like Oculodentodigital dysplasia. Hardly anybody will ever see the article, though I imagine that its existence pleases the families of the ~100 patients in the world that are affected by this exceedingly rare disease. The article gets 3 or 4 page views per day. We know that nobody clicks on the little blue clicky numbers in 99.7% of page views, so if it's an average article, then just four people per year will click on a ref tag, and there's no guarantee that any of them will click on the links (if any) in the citation.

So: I've made a very low traffic article that nobody checks the sources on. How could the contents of the source nobody's reading, recorded in a ref tag that practically nobody is clicking on, on a page almost nobody's reading in the first place, affect anybody's view of anything? WhatamIdoing (talk) 23:10, 30 October 2022 (UTC)[reply]

LessHeard vanU, readers don't care whether the subject of a Wikipedia article is notable. Remember that almost everyone gets here from a web search engine. They're getting here because they're voluntarily, intentionally looking for information (usually a very tiny bit of information) about that subject. It is extremely unlikely that they will search for information about WhatamIdoing's Gas Station, click on a link to read a Wikipedia article about WhatamIdoing's Gas Station, and then think, "I'm offended! How dare Wikipedia have an article about the thing I was looking for! Also, how dare they have hardly any information in it and how dare they not cite sources that prove it complies with Wikipedia's policies and guidelines, about which I know absolutely nothing, including whether any such things exist!"
Have you heard the joke about the restaurant? "The food is terrible, and the portions are much too small."
Wikipedia's reputation is never in any danger when we provide readers with accurate information about the subject they are seeking. I've read a lot of research about Wikipedia's readers over the last dozen years, and I don't recall seeing even a single comment from a non-editor that says anything about there being too few sources in an article. I've seen some in which researchers have to point out to readers what those little blue clicky numbers are for, but never even one complaint about too few sources in an article. A list of sources to prove notability is just not part of the readers' goals. WhatamIdoing (talk) 14:44, 31 October 2022 (UTC)[reply]
LessHeard vanU, I think you're right: We are probably talking about different things. When you said that articles would "devalue" the encyclopedia, I assumed you were thinking about reputation. It now sounds like you meant something closer to "creating articles makes work for some editors (=the ones who have volunteered to do that work)". WhatamIdoing (talk) 18:50, 31 October 2022 (UTC)[reply]

Comments from Atsme

Reply moved from #Comments from Scope creep –MJL ‐Talk‐☖ 20:09, 1 November 2022 (UTC)[reply]

Are you aware of Botipedia? Atsme 💬 📧 13:15, 21 October 2022 (UTC)[reply]

Comments from 69.203.140.37

Reply moved from #Comments from User: Stuartyeates –MJL ‐Talk‐☖ 20:09, 1 November 2022 (UTC)[reply]

You are right about the obvious fit of Wikidata to projects such as mass creation. But that just transfers the problem to a more opaque platform when it comes to policy. Garbage-in/garbage-out applies with a vengeance at Wikidata. Dubious "references" and self-dealing (in the form of circular references) is a real issue. 69.203.140.37 (talk) 13:23, 22 October 2022 (UTC)[reply]


Comments from Levivich

ONLY POST IN
YOUR OWN SECTION
CLERKS WILL SUFFER
NO EXCEPTION
Burma-shave
Levivich (talk) 23:21, 1 November 2022 (UTC)[reply]

Comments from Editor X

Open additional comments sections below. Please limit comments within a section to 300 words.

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Arbitration_Committee/Requests_for_comment/Article_creation_at_scale&oldid=1123025533"