Jump to content

Wikipedia:Arbitration/Requests/Case/Conduct of Mister Wiki editors/Evidence

From Wikipedia, the free encyclopedia
Main case page (Talk) — Evidence (Talk) — Workshop (Talk) — Proposed decision (Talk)

Case clerks: Kostas20142 (Talk) & Amortias (Talk) & L235 (Talk) Drafting arbitrators: Euryalus (Talk) & Newyorkbrad (Talk)

Case scope

[edit]

Case scope:
The case will examine:


  1. The alleged breach of WP:PAY by Salvidrim! and Soetermans in reviewing each other's AfC submissions.
  2. Salvidrim!'s use of admin tools to grant additional userrights to the account Salvidrim! (paid).
  3. Any other evidence relevant to conflicts of interest by editors associated with the Mister Wiki business

Notes:

  1. Should new evidence suggest the case scope requires amending or expansion, this amendment will be published in this section and the community invited for input on the associated talk page.

Evidence presented by Jytdog

[edit]
Background
  • The abstract proposal that "admins should not use their bit for pay" has found no consensus twice recently, here in July 2017 and here in October 2015.
  • Discussion in August 2017 about paid editing and NPP/autopatrol, spurred by KDS4444. Consensus against paid editors having these privileges.
  • After the KDS444/OTRS scandal emerged, an RfC was opened at VPP which is getting high level consensus but is vague.
  • That is dealing with stuff abstractly. In every specific instance that I know, people who were entrusted with advanced privileges and did extensive underhanded things for pay had rights stripped and some have been indeffed. There aren't many.
    • Nichalp (crat, oversighter, admin, OTRS), used a sock for paid editing. Not indeffed but all privileges stripped by Arbcom in 2009 here
    • Wifione same deal as Nichalp, except that Wifione was desysopped and indeffed via the arbcom case in Feb 2015. Times change.
    • Malunrenta was a New Page Patroller working with a big UPE sockfarm. Indeffed for socking.
    • KDS444 voluntarily gave up NPP and autopatrol after being asked to, was stripped of OTRS access rights by the OTRS admins off-WP, but after the OTRS stuff emerged here on WP, was indeffed here.

The purpose of laying that out, is to show that while the community cannot agree on a policy about advanced privileges and paid editing in abstract discussions....when admins and people with advanced privileges actually do extensive underhanded things in the course of commercial paid editing, the community does not approve at all, those people lose the trust of the community, and we take away their bits. This is what User:Dennis Brown wrote at the VPP RfC. This is commonsensical. What Salvidrim is doing here, clinging to his bit with wikilawyering, is not what we do.

More background about "conflict of interest" and "prior review" and what they mean for the integrity and quality of WP content.

COI isn't something personal. It is a structural situation. A person has two sets of relationships that put him or her in conflict in some given situation leading to the question - "which interests do I serve here"? COI is managed in knowledge-producing and knowledge-publishing institutions everywhere in the developed world, and in government and companies that are well-run. There is well-deserved scandal for institutions when COI is not managed. This is a structural, common sense thing.

I do not believe the community would ever approve giving advanced privileges to anybody who says they edit for pay, commercially. These scandals have all arisen when people were not upfront about that at the beginning, or started editing for pay (commercially) after they got advanced privileges.

AfC is one of our key processes to manage COI, as well as to ensure that articles created by new editors, who haven't learned the policies/guidelines/norms, are appropriate for WP.

  • As the Introduction at Wikipedia:WikiProject_Articles_for_creation says AfC works as a peer review process in which registered editors can either help create an article submitted by an anonymous editor or decline the article because it is unsuitable for Wikipedia.
  • Further, the COI guideline says, at WP:PAY, that paid editors should:
    • propose changes on talk pages (to which you can call attention by using the {{request edit}} template or by posting a note at the COI noticeboard), so that they can be peer reviewed before being published;
    • put new articles through the articles for creation process instead of creating them directly, so they can be peer reviewed before being published;
  • Further, the COI guideline says at WP:COIRESPONSE that Editors responding to edit requests from paid editors are expected to do so carefully, particularly when commercial interests are involved.

What Salvidrim and Soetermans did on both sides of the AfC (Salvidrim passing Soetermans' submission; Soetermans passing Salvidrim's 2 submissions) was nothing like actual peer review -- they ignored both the letter and the spirit of AfC and the COI guideline, and corrupted that process by jumping each other's articles through the AfC hoop. The corruption is uglier because each one knew that the other had a paying client behind him.

Both Salvidrim and Soetermans also believed it was fine for each of them to directly edit in mainspace for pay. In my view, there is broad consensus in the community -- in its actual practice and what people actually expect of one another (with sharp dissents from both "wings") -- that commercial paid editors should not directly edit in mainspace except for fixing vandalism and making noncontroversial changes but instead should submit content for prior review. I am talking about that part of the COI guideline. Regardless, the whole of WP:COI is currently a guideline - something that per WP:PAG describes "best practices that are supported by consensus. Editors should attempt to follow guidelines, though they are best treated with common sense, and occasional exceptions may apply".

Point 1

Salvidrim (diff) and Soetermans (diff) have said that they did not do the AfC review for pay. I believe them, and this is not really the problem. User:JacobMW could perhaps shed more light on this. But this kind of one-off is impossible to prove for us.

The community has indefinitely blocked people for extensive patterns of editing that very clearly appears to violate PAID and COI. Just for that. The cases are:

  • 2017-01 Kavdiamanju (nondisruptive behavior other than consistent PAID/COI violations)
  • 2017-01 FoCuSandLeArN (nondisruptive behavior other than consistent PAID/COI violations)
  • 2017-04 jennepicfoundation, disclosed and not editing directly, but doing nothing but battering talk page (WP:PAYTALK)
Point 2

See the timeline at COIN and this diff by TonyBallioni, and acknowledgement by Salvidrim here

Point 3

The case request contained links to:

  • a discussion with Salvidrim! at my TP. I realize that I was kind of prosecutorial there but I did not want this to escalate - per DR I wanted to resolve this locally, at a user talk page as a first step. And I was satisfied enough by the end, that Salvidrim understood where he had gone wrong, as he has now expressed that self-insight several times, that even he can be affected by conflict of interest caused by getting paid and he should have allowed the community to manage it.
  • COIN discussion. I had posted at COIN to have the content that was directly edited for pay (or fake AfC reviewed) reviewed by independent editors. This unsurprisingly also led to further discussion, and during that discussion Salvidrim did good things like voluntarily give up bits he had given to his paid alt account.
  • What the COIN discussion came down to, was several editors asking (some demanding) that Salvidrim either a) give up the bit or stop editing for pay and b) if he wants to keep the bit, resign and resubmit to RfA. His reaction was this, which i view as defiant and which in my view, is Salvidrim continuing the same arrogance, judging for himself whether he still has the trust of the community rather than allowing the community to decide. It is not for the person who broke trust, to judge just how bad it is. His self-insight at my talk page wasn't generalized.

The case should be focused on whether Salvidrim has lost the trust of the community and should have his bit removed. The key issues are:
1) MEAT or GANG or collusion or whatever word you want for it, for what Salvidrim did at AfC as part of him getting paid. Namely submitting two articles to AfC that he and Soetermans arranged off-WP for Soetermans to review. Putting the articles through AfC was Salvidrim's plan for getting the tags removed (the paid task), and he cited the corrupt AfC multiple times before its nature was made clear. Shown in the timeline.
2) Point 2.
3) His RfB, cited multiple times in the case requests.
4) His refusal to allow the community to judge whether he still had the trust of the community after all the stuff above, by putting himself back through RfA, and instead wikilawyering to retain his bit, causing all us to spend yet more time. On him. Instead of on building content. Jytdog (talk) 23:37, 1 December 2017 (UTC) (trimmed a lot Jytdog (talk) 04:16, 4 December 2017 (UTC)) (expanded a bit again. Jytdog (talk) 20:53, 4 December 2017 (UTC)}[reply]

  • User:Salvidrim!, When you write things like WP:PAY makes no mention of user-rights or page moving this is like Bill Clinton's "it depends on what 'is' is.. This proceeding is about whether the community can trust you to administer the policies and guidelines according to their spirit; their letter is not hyper-detailed on purpose. You should perhaps ask yourself if this kind of ... argument, increases or further decreases the community's trust in your judgement. Jytdog (talk) 23:31, 4 December 2017 (UTC)[reply]
[edit]

I did a search of RfA for "paid" and have looked through the results over the past couple of days. here is what i found.

  • The only !vote where i saw a concern raised about paid editing was an !oppose vote, here back in 2013; that was a concern about a possibly paid article (raised very gently) and was one among several concerns raised. It failed for other reasons with !votes 21/23/17.
  • this RfA was for someone who was paid to research WP. Passed. (GLAM/WiRish)
  • Wikipedia:Requests for adminship/Sadads GLAM/WiR editor, passed.
  • Wikipedia:Requests for adminship/Zappaz, failed. Mostly for POV editing; claim of possible paid editing by one opposer.
  • Wikipedia:Requests_for_adminship/Kevin_Gorman, GLAM/WiR, did a lot of work on Wiki-PR matter, passed.
  • Wikipedia:Requests_for_adminship/Keithbob, withdrawn with about 50/50 support/oppose, with opposes driven by ...unresolved concerns about a cluster of issues on the border of COI and advocacy, and additionally unresolved sock/meat/GANG whatever, all related to the transcendental meditation movement.
  • Wikipedia:Requests for adminship/Brianhe failed, probably due to concern about overly fierce pursuit of COI/paid editing. The only one like this I found.


It is something frequently brought up in the questions - for example (these are links to whole RfAs - just search for "paid") asking what they think about paid editing and how to manage it, eg. here, here, and here, sometimes asking if the person ever edited for pay here. Saying that you are not opposed to paid editing, is not a death sentence, per this one and this one and this one. Although, in this one a relatively inexperienced editor had a userbox supporting paid editing (Template:Paid editing supporter) (added here) , and the those two things - inexperience + this support for paid editing, gave several "opposes" pause. Mostly failed due to inexperience/youth... I think.

Paid editing was extensively discussed in the questions at one of the most recent RfAs.

That is what i found.

Nobody who said "I edit for pay" in the commercial sense. Jytdog (talk) 03:25, 14 December 2017 (UTC)[reply]

Evidence presented by Salvidrim!

[edit]

Good faith

[edit]

Although I severely underestimate how carefully one must manage a COI and I definitely let it affect my judgement without fully realizing it, I didn't act in bad faith, in an intentionally nefarious manner or in an intentional subversion of policy. As evidence of this I'm submitting two private conversations (which the other party of has authorized disclosure of)

  • My conversation with a WPVG friend and fellow MisterWiki paid editor Soetermans which led to his approval of the two drafts I had been paid to cleanup, which I think shows we both had misgivings about the appropriateness of the reviewing but nevertheless made the mistake of going through with it
  • Submitted to ArbCom by e-mail 20:03, 1 December 2017 (UTC) - My email exchanges with Jacob from MisterWiki, concerning the Dan & Reza articles, the Studio71 rename, as well as two other articles which I've declined working on and explained why I think the requests couldn't be completed within policy; I believe this shows that I wasn't just interested in "taking money" but that I attempted to do significant mentoring with Jacob (inexperienced with Wikipedia) and that I didn't blindly accept mandates for projects I did not honestly, in good faith, believe in (however skewed my "honest judgement" turned out to be)

Additionally, although "admins also doing paid editing on the side" is a subject of unresolved discussions amongst the community (whereas actually using admin tools or other advanced permissions is widely opposed), several editors (and most crucially several current ArbCom candidates) have said that the only way that it could be plausibly performed would be by erecting an impermeable "wall of china" between the paid and admin account; I think my setting up of a second account clearly named "paid" with a disclosure-filled userpage shows that I intended to install a clear division between the roles, and although I underestimated how carefully COI must be managed to preserve this independence, I did make a good-faith, best effort to act in the way that is now being recommended by the community as the only way my situation could have been handled. Ben · Salvidrim!  20:03, 1 December 2017 (UTC)[reply]

Page mover on my alt

[edit]

The granting of the permissions to my alt was commented on at length; the rollback and confirmed perms don't seem to be an issue, and the reviewer perm went unused, so the issue was with granting the page mover right which I then used to rename Studio 71 to Studio71. Opinions amongst people who have made statements were divided, with some claiming that the move was controversial and should have gone to RM, and others yet supporting my initial notion that this was not a controversial move and if it would have within policy to perform it as a volunteer, then performing it as a paid edit wasn't a violation of article titling policy. Nevertheless, once confronted I removed the permissions from my alt myself. Ben · Salvidrim!  20:03, 1 December 2017 (UTC)[reply]

AfC reviews were not "paid", but...

[edit]

My AfC review of Datari Turner (upon request from a friend, Soetermans) was not paid editing on my part, and it predates any contact with Jacob or even any thought of paid editing. However, the AfC review clearly sowed the seed that later led to Jacob and me to get in contact and engage in a paid editing relationship.

Soetermans' review of the Dan Weinstein and Reza Izad articles he claims he was not paid for directly (and I've no reason not to believe); I asked him for the reviews (see above) and I, however, was doing the asking against expectation of payment (the mandate being "article maintenance tag cleanup" and the AfC drafting being implicitly part of that process), even though ultimately that part did not result in payment for the obvious reason that both articles ended up deleted.

Of course, this is a relatively nitpicky detail that some might prefer to paint over with a wide brush of "Soetermans and Salvidrim were both generally paid editors for MisterWiki and the individual mandates or payments or whatever don't change the fact they both had paid COI related to MisterWiki", and there is in hindsight some validity to that argument; attempts at minute compartmentalizing can't defeat the fact we were both operating implicitly to "please" a business partner where maintaing a positive relation was liable to result in further paid work, so although it wasn't defined in scope and amounts, anything we did involving any MisterWiki-related article could be thought of as "against the expectation of future paid work". Ben · Salvidrim!  20:47, 1 December 2017 (UTC)[reply]

Meatpuppetry vs. Collaboration

[edit]

The wording of WP:MEATPUPPETRY explicitly refers to WP:CANVASSING (recruiting like-minded editors to influence discusions) and dispute resolution processes, which don't seem to be involved here. I asked a WPVG friend for help, he accepted, that was a perversion of the purpose of AfC (review by a neutral editor) but I don't think the spirit or wording of WP:MEAT covers this kind of collaboration -- otherwise, it would also cover any instance where someone posts on Discord, WT:VG, IRC or Facebook about wanting help on an article -- for example, several Discord members recently discussed and collaborated on Lupinball and Dust II. Were we meatpuppets? I've posted on AN/I for help from other editors to deal with a discussion being disrupted. Were those who responded meatpuppets? I think the definition of meatpuppetry is intentionally narrow and "editors helping editors" outside of disputes and/or discusion doesn't seem to be what is covered in that policy. I'm not defending the AfC collusion, but I'd caution against trying to make it fit under an apparently-new definition of WP:MEAT. Ben · Salvidrim!  21:46, 1 December 2017 (UTC)[reply]

User-rights, adminship and paid editing

[edit]

Firstly, which user-rights paid editors are eligible to hold does not seem to be covered in any policy, nor has it been the topic of much discusion prior to me granting page mover to my alt. The question "would a neutral admin have granted user-rights to an admin's paid-alt if requested in early November" I think can only be answered with "depends on the admin but I don't think this is an unambiguous no", which is to say that, lacking a policy concerning paid-editors and their user-rights, I don't think it is "obvious no admin would have granted them" at that point in time.

Whether admins can edit for pay has been the subject of discussion for years (this 2015 discussion was amongst the most extensive, albeit inconclusive.) A discussion opened on Nov 1st ([1]) shows that there is pretty much consensus against the use of admin tools (and other advanced permissions like CU/OS, which I'll include under "admin tools" for simplicity) against payment (or implicitly in any situation involving the admin being a paid editor). The discussion was started after KDS444's documented misuse of OTRS to solicit clients (otrswiki:Agent issue (2017)) and continued picking up steam after I started editing for pay. However my granting of the userights to my paid-alt predates any consensus that may have emerged from the discussion (Nov 1st, practically simultaneously with the discussion's very beginning). However, although the VPP discussion linked above is currently pretty definitively against the use of admin tools for paid editing, it's much less definitive about admins engaging in paid editing at all (without using their admin tools), which is a proposal not yet agreed upon and thus found nowhere in policy. In fact, the idea of setting up a paid-alt separated by a "chinese wall" is one supported by multiple editors (including current and future Arbitrators) as the only plausible way admins can engage in paid editing adequately. Support for it is certainly not unanimous (see WT:COI), but these quotes from ArbCom candidates demonstrate that although I was the first to try it, my "admin + paid-alt" solution has singificant support as the only way for paid editing amongst those most interested in policy:

Quotes from future Arbitrators
Opabinia regalis – "All paid edits must be segregated. No mixing paid and volunteer edits under the same account. All volunteers must disclose their paid account if they have one. All paid editors must do all of their work under one account."
Sir Joseph – "It's hard to say if someone can be a paid editor and still be an admin, I think it might be possible if that admin puts up a Chinese wall between the paid work and any other area of Wiki that the admin patrols and enforces policies."
SMcCandlish – "In theory, someone who does paid work regarding a topic could be an admin in good faith, though it would require extra care. (...) I don't think summary desysopping is in order, however; if anyone loses the admin bit it should be for actual cause (or because they became inactive for too long)."
RickinBaltimore – "With regards to admins and using their tools with payment, along with general editing, this is a very tenuous area. Admins, by their nature, need to remain neutral and refrain from any conflict of interest. If an admin should be shown to be using their tools in conjunction with being paid to edit, then this is a violation of their admin roles I feel. However, as of now we do not have a community consensus on this area, but personally an admin should recluse themselves from any possible area of a conflict of interest. I would hope that an admin in this situation would refrain from making any and all edits, up to and including admin tasks, that could show a direct COI."
Callanecc – "Any editor with advanced permissions is expected to use them in accordance with community policies, if they're being paid to use them I'd question whether they would have developed a significant-enough COI to make them INVOLVED in that topic area. If the admin has properly disclosed and they are very careful to avoid using their tools in the area in which they are paid editing then I wouldn't necessarily have a reason to desysop them. Having said that, it's going to depend very significantly on the circumstances."
Mailer diablo – "Any use of administrator tools or/and advanced permissions to aid, abet, counsel or procure paid editing is incompatible with and fundamentally breaches the community's trust in that editor to hold such privileges. But I will not object to an admin accepting paid editing if and only if the admin does not use any admin tools or admin status in doing so (i.e. saying that "I am an admin, I have more credibility, let me write your article for payment!" is *not* acceptable.)"
Worm That Turned – "I cannot see a reason that paid use of administrator tools is acceptable - admin tools are given on "trust" and the community definitely hasn't got to the point where they trust paid editors. Yes, admins can do paid with respect to the project and maintain neutrality - I would include all WMF staff and Wikipedians-in-residence in this category - but in any situation where it is attempted, the individual should go above and beyond to ensure the perception of propriety."
Premeditated Chaos – "I don't think there is a community consensus yet on admin tools and declared paid editing (that discussion is still ongoing in lots of places), but my personal opinion is that it's a terrible idea for admins to edit for pay. The only way I could hypothetically see it working without compromising the community's ability to trust someone would be for them to use a separate account exclusively for paid edits so they could be scrutinized easily, and then to only ever suggest changes on the talk pages of the articles in question using the requested edit system, never to actually make any edits themself."
KrakatoaKatie – "Currently there is no prohibition on admins acting as paid editors. I personally do not believe administrators should do it, and I would not do it myself; whether that's going to be the community consensus or not is a different matter."

In summary, I'll fully accept the blame for underestimating how difficult to actually do this separation is, and I failed by letting COI get the better of my judgement, but the idea that there is consensus that "admins can never engage in any paid editing, in any form" is demonstrably incorrect. The first attempted implementation of the only solution, a completely separate paid-alt, has failed somewhat spectularly, but very few "first attempts" are complete successes and to claim that I violated policy updates that resulted from my actions is... creative. Ben · Salvidrim!  04:25, 2 December 2017 (UTC)[reply]

Evidence presented by TonyBallioni

[edit]

Salvidrim! engaged in meatpuppetry while an SPI clerk

[edit]

As to the statement above that MEAT only applies to canvassing and that this didn't count as canvassing, that doesn't match the text of MEAT which states [s]anctions have been applied to editors of longer standing who have not, in the opinion of Wikipedia's administrative bodies, consistently exercised independent judgment. Additionally, Soetermans statement shows that he was under the impression that this was a quid pro quo situation. Salvidrim! had reviewed his work, so therefore he was "returning the favour". Even if he wasn't paid for this, he can't reasonably be thought to have had independent judgement in this case since he had a common employeer and because Salvidrim! had reviewed similar content for him in the past. Salvidrim! knew this, and asked this specific user for it anyway, rather than any number of other users he might have asked. While this was not a discussion, Salvidrim canvassed a specific editor who he had reason to believe would agree with is judgement and who had a similar conflict of interest, and it prevented an unbiased editor from reviewing the content. As an SPI clerk he should have known that this was in violation of the sockpuppetry policy, and only makes the situation worse. TonyBallioni (talk) 02:37, 2 December 2017 (UTC)[reply]

  • I'm adding Ivanvector's statement on the workshop page, and the SPI case he brought up, to be considered in evidence on this point, because I think it is an excellent summary of how Salvidrim! violated the sock policy, and addresses concerns that we would be expanding the definition of meatpuppetry beyond that which already exists. We block editors for this type of behavior at SPI, and an SPI clerk should have known that we do. TonyBallioni (talk) 19:28, 7 December 2017 (UTC)[reply]

Salvidrim! violated WP:PAY and WP:TITLECHANGES using advanced permissions he granted himself

[edit]

The title change policy makes it clear that any move that should be controversial, and the guidelines for commenting on requested moves state clearly that Please disclose whether you have a vested interest in the article, per WP:AVOIDCOI. This combined with the instructions of WP:PAY to avoid directly editing articles show that any page move where the person executing the move has been paid to do it would be controversial and need community oversight. As GMG points out, this would not even have been an option for any paid editor as the accounts would likely not have been granted the tools by an uninvolved admin. In terms of whether this move was controversial on its own, yes, it was a MOS:TMSTYLE, which can be some of the most controversial move topics. In this case, Salvidrim bypassed community review to move a page to the preferred brand of his employer without regard for the moves process.

I also find it hard to believe that we are debating as to whether a page move counts as editing. Of course it does. If we want to break it down to technicalities here, the fact that moves are recorded as edits in the page history makes it clear that a page move counts as an edit. TonyBallioni (talk) 14:24, 2 December 2017 (UTC)[reply]

Salvidrim! moved articles written by another paid editor to draft to avoid community scrutiny at the request of a client

[edit]
  • Articles in mainspace tagged for COI and notability: [7], [8]
  • Salvidrim! moves tagged articles to draft: [9], [10]
  • Salvidrim!'s notes on the drafts: [11], [12]

The edit summaries and paid declaration by Salvidrim! here make it pretty clear what was going on: the client didn't like having the COI and notability tags on the article, and he was paid to find a way to get them off. He sent them to draft space where they were less likely to be deleted, and then contacted a friend to review them without making any changes and move them back to mainspace without the tags. This was only discovered because of the COIN posting after Jytdog had engaged Salvidrim! about the topic of being a paid editor while an admin. Otherwise these articles would have likely stayed in mainspace with no oversight undetected. Also worth noting is that when he sent these drafts to draftspace, he suppressed the redirect using the extendedmover flag he had granted himself, so there was no one aware of the process other than himself and the eventual AfC reviewer he selected. Both articles were eventually deleted at AfD. TonyBallioni (talk) 15:08, 2 December 2017 (UTC)[reply]

Soetermans joined AfC for the purpose of reviewing Salvidrim!'s drafts

[edit]
  • AfC join request, granting, and removal: [13], [14], [15]
  • Only drafts reviewed by Soetermans:[16], [17]
  • Salvidrim!'s declaration involving these articles: [18]

This is pretty obvious: Soetermans was not an AfC reviewer and did not have access to the AFCH script. He had been contacted by Salvidrim! in order to review the drafts that Salvidrim had moved to draft space to get rid of the maintenance tags, as demonstrated above. Soetermans did not disclose to Primefac that he had been contacted by someone who was paid by the same firm as himself to review the drafts, and then within 9 minutes of being given access to AFCH, Soetermans reviewed the drafts he had been asked to review by Salvidrim!. These were the only drafts he reviewed before Primefac removed access, and it is clear he only joined AfC to review the drafts that Salvidrim! had asked him to review. TonyBallioni (talk) 15:08, 2 December 2017 (UTC)[reply]

Salvidrim! has violated the community's trust

[edit]

If I need diffs on this statement, let me know and I can provide them, but I think it is simpler without them. As Beeblebrox points out below WP:ADMINACCT states Administrators who seriously, or repeatedly, act in a problematic manner or have lost the trust or confidence of the community may be sanctioned or have their administrator rights removed. Adminship is fundamentally based on having the trust of the community, and policy allows for removal based on the loss of trust even if it is not a pattern.

Others have alleged a pattern of behavior here of poor judgement, which I think it is worth the committee to examine, but my reason for filing this case was simple: even if this was a good faith series of mistakes, they were such a severe breach of trust in the most contentious policy issue on en.wiki currently that Salvidrim! can only act as an administrator if he goes through an RfA again. Salvidrim! knew how controversial his actions were, knew that if they were discovered would be handing the community drama, and decided to do take these actions anyway. He thinks he still has the trust of the community and some who have commented think that this isn't a reason for him not to be an admin. That is a fair position, but the only way to test if that is true is through a new RfA. At the COIN thread 9 editors, including several admins and a functionary, called for him to voluntarily resign the bit and stand again at RfA to see if he had the community's trust. During the case request many called for this to be a summary desysoping by motion and for it to be dealt with quickly.

All of this shows that there was a severe breach of trust by Salvidrim! and that the community has a right to decide whether or not he should remain an admin through a new RfA. The Arbitration Committee is the only body that is capable of giving the community that right, and in my mind it has an obligation to the community that elected it and that it represents to do so, and that this obligation can be seen by the wording from the administrators policy quoted above. TonyBallioni (talk) 15:25, 2 December 2017 (UTC)[reply]

Evidence presented by Beeblebrox

[edit]

Pattern of poor judgement?

[edit]

I’ll keep this short and to the point: I like Ben/Salvidrim, and don’t recall ever having any negative interactions with him. However, there comes a point at which we must question if a particualr individual truly has the judgement needed for a position of trust. Salvidrim basically already used his “get out jail free card” when his account was compromised in 2015 due to him using the same, very simple password on another website that was subsequently hacked. This was after a previous wave of admin accounts being compromised, after which all admins were urged to strengthen their passwords,* but this advice was clearly not heeded.

He’s not a terrible admin generally, and in fact has done a lot of great work, but given that this is not the first time he has exhibited extremely poor judgement the committee should consider WP:ADMINACCT, which reads, in part ”Administrators who seriously, or repeatedly, act in a problematic manner or have lost the trust or confidence of the community may be sanctioned or have their administrator rights removed” and goes on to specify that consistent or repeated poor judgement is one of the reasons for invoking it. Is there enough of a pattern of poor judgement to warrant removal of tools? I’ll leave that to the committee to decide, only asking that they not make their decision lightly. Beeblebrox (talk) 01:04, 2 December 2017 (UTC)[reply]

*I remember this happening but cannot locate whatever page it took place on. So, take that for whatever you feel it is worth. Beeblebrox (talk) 21:09, 11 December 2017 (UTC)[reply]

Evidence presented by GreenMeansGo

[edit]

I've mostly already said my piece in the COIN thread and the case request, but I suppose we're expected to say it again. Granting rights to a paid editing account was a misuse of the tools, plain and simple, because they're rights no other paid editing account would have had if it had not been an alt of a sysop. The "trusted to use them anyway" argument falls flat, because no non-sysop with the same rights who showed up to PERM asking for them to do paid editing would have gotten them, should have gotten them.

The wikilawyering over the definition of meat puppetry is silly, but if you want to call it what it is in plain language, colluding off wiki to abuse AfC for the purpose of paid editing is just unethical, and I'm not sure the average person needs a policy to spell that out for them. Arguing something like this at an AfD for one of those very articles is... at best blindness to the fact that they never would have been at that AfD were it not for their COI. Arguing for the removal of maintenance tags based at least partially on a AfC review process that was started in favor of a client, and then circumvented through collusion is... bad.

Whatever your decision is, if you have a conception of community trust that is consistent with retaining a sysop who can't be trusted to do AfC, then it is vastly different from my own. GMGtalk 02:53, 2 December 2017 (UTC)[reply]

Evidence presented by Deryck C.

[edit]

Many admins engage in paid editing and declare them

[edit]

With all due respect to Kudpung and Ivanvector, both en.wp sysops whom I regard highly, I disagree with their statement that "advanced permissions are incompatible with paid contribution". Many other high-profile en.wp sysops have taken up paid editing positions with employers who aren't Wikimedia affiliates; examples I can think of include Charles Matthews, Keilana, Sadads, and Andrew Gray. Some of them have separate accounts for paid edits and wikilink the two user pages; others don't have separate accounts and simply declare paid edits on their main account. The fact that some paid editing is called "Wikimedians in residence" and others simply "paid" should not result in different standards of transparency.

Re @Alanscottwalker:: The above paragraph is not meant to be an exhaustive list of current admins who have been paid for their contributions to Wikipedia. Neither do all paid editing by other admins fall under the "Wikimedians in residence" umbrella. Cyberpower678, for example, gets paid by the Internet Archive to develop a bot to promote the use of the Internet Archive to combat link-rot, and his position is not classified as a "Wikimedian in residence". Deryck C. 15:26, 4 December 2017 (UTC)[reply]
I am really not fond of being dragged in to ArbCom matters. My contract with Internet Archives is strictly to dedicate work towards combating WP:LINKROT in the development of IABot. The terms are quite clear. The software is and will remain open source, the bot will comply with all local bot policies and run under consensus, and the bot will not be exclusive to the Wayback Machine.—CYBERPOWER (Merry Christmas) 16:25, 4 December 2017 (UTC)[reply]

Punishing an editor because of a declared COI will always be counter-productive

[edit]

I'd like to repeat DHeyward's elegant statement here: "We have a COI policy that really only works if we reward disclosure. Whether that's real world identity or a declared "paid" account. Hammering disclosed accounts does not further the interests of reducing COI. Rather, we should encourage disclosure and challenge contributions on good faith presumptions. Disclosure should help us identify where review is needed, not where punitive actions should be levied. Desysopping and banning are ways to ensure that COI is not willfully disclosed. Remuneration is not the measure upon which we should measure COI, rather it is hidden and undisclosed ties that are the problem."

In other words, because Salvidrim! has openly declared his paid editing interests, the yardstick for applying any sanction should be "Is this a punishable misuse of privileges if he wasn't paid?". Any deviation from this yardstick will mean that ArbCom is punishing an editor because of a declared COI and pouring en.wp's COI policy down the drain. Deryck C. 12:12, 4 December 2017 (UTC)[reply]

Evidence by Alanscottwalker

[edit]

Deryck Chan's representation of four WiR as "many" (four) seems odd. At any rate, contrary to the representation WiR does have an explicitly different status under WP:COI (And WiR is explicitly not meant to 'wag the dog'). The detail of disclosure and duty of the WiR job is in fact totally different from PAID [19] and extensively regulated by Wikipedia. Alanscottwalker (talk) 14:50, 4 December 2017 (UTC)[reply]

Deryk, now you are reaching to BOT's? Also extensively regulated, including pre-publication review. Alanscottwalker (talk)

Evidence presented by Ivanvector

[edit]

Salvidrim! and Soetermans properly disclosed their conflict of interest

[edit]

In case it needed to be said. See User:Salvidrim! (paid) and User:Soetermans#Paid editing. Further, both editors have demonstrated exemplary openness and transparency once this process was initiated, which ought to be commended whatever the outcome of this case. Furthermore, there is not presently any policy which forbids editing for pay neither by all users nor by users with any particular set of permissions; however both are strongly discouraged by policy.

Two accounts working together for some purpose may be considered one user

[edit]

Repeating a finding from the 2005 Regarding Ted Kennedy case which is cited in the meatpuppetry section of the sockpuppetry policy: "For the purpose of dispute resolution when there is uncertainty whether a party is one user with sockpuppets or several users with similar editing habits they may be treated as one user with sockpuppets." I believe Salvidrim! and myself are the only sockpuppet investigations clerks who have yet edited this page, and while I think it's clear that Salvidrim! and Soetermans are separate users, I expect Salvidrim! knows that users who tag-team in this way are violating the spirit of the policy if not its letter, and most are at least temporarily blocked.

Salvidrim! and Soetermans collaborated to avoid scrutiny

[edit]

This conversation with Jytdog indicates that Salvidrim! had been asked by Soetermans to expedite an AfC review of Datari Turner, a page which had been previously deleted four times for blatant promotion and whitewashing by the article's subject, so that Soetermans' client would not have to wait for the long AfC queue; it was deleted again after a second AfD which also cited its promotional content.

Later, two articles, Dan Weinstein (business executive) and Reza Izad, were created by an editor paid by Studio71 (Weinstein and Izad's company and a Mister Wiki client) and who had not disclosed this relationship. Both were patrolled and tagged with {{COI}} and {{notability}} by JJMC89 in good faith. Salvidrim!, who by this time had been engaged by Mister Wiki, indicated in the above-diffed conversation that his client asked for the tags to be removed, at which point Salvidrim! moved both pages to draft space, removed the tags, and flagged the drafts for AfC review. Three days later Soetermans, who Salvidrim! knew had also been engaged by Mister Wiki, offered to review the drafts, and subsequently moved them back to article space having made no changes to the article prose at all. This is a comparison of Reza Izad before JJMC89's patrol versus after Salvidrim! and Soetermans' edits. Both articles were subsequently deleted at AfD: Dan Weinstein and Reza Izad, in both cases with Salvidrim! being the only editor arguing against deletion (Soetermans did not participate).

It stands to reason that a truly neutral AfC reviewer would not have promoted the articles, as in all three cases consensus overwhelmingly supported their deletion. At best, the two editors both repeatedly exercised extremely poor judgement with regard to promotional content and their own conflicts of interest (with Mister Wiki and with each other). At worst, the two acted as meatpuppets to evade scrutiny and effect a result which has been shown to be against consensus.

Replies

[edit]

This section may be better off on a talk page, I'm unsure of the right process here.

@Deryck Chan: I think the Committee has indicated that the issue of who should be permitted to edit for pay is outside the scope of this case, but I'll briefly reply. I think you're referring to this comment (accidentally duplicated), and I stand by it. On one hand there are paid Wikipedians (WiR, outreach, and the like) who work in various ways to promote the project, build and maintain useful tools, conduct outreach events and engage resources to make Wikipedia better for everyone. Then on the other hand there are paid Wikipedians who are engaged by third parties to write content that advances the goals of the third party. There is a bright line and a wide gap between those two types of paid editing; the second works directly against WP:NPOV, one of our core content policies. We all know this is the case: nobody can edit neutrally from a conflict of interest as this case amply demonstrates. That's why we have paid disclosure requirements and a conflict of interest policy as checks and balances against conflicted editing. While I don't personally think we should stop editors from accepting pay for editing (or even that we could if we wanted to) I disagree entirely that Wikipedia should reward disclosure: the "reward" is the paid editor's salary and not having their edits deleted and their client demanding their money back. And if the "price" of being permitted to edit on behalf of a third party is [temporarily] putting down the tools you can use to circumvent the checks and balances, then so be it. Observe that all of the paid editors in the first group have already disclosed, with no "reward" forthcoming from the community. Okay, this wasn't so brief. Ivanvector (Talk/Edits) 19:59, 4 December 2017 (UTC)[reply]

Evidence presented by Soetermans

[edit]

Unfamiliarity with PAID; an error in judgement

[edit]

I've said what I wanted to say at COIN and on my talk page. To me, my mistakes boil down to two things.

1
I started doing some paid editing improvidently, all in good faith. I thought the subjects were notable, and created the articles, but I did so without reading the guidelines properly. For instance, while expanding the article on Arne & Carlos, I did not add the paid editing disclaimer, which is required. This is odd, as Template: Paid is talking about "have been paid", which I wasn't (and I've waived any fees I would have received). Same goes for my talk page, I haven't been paid for Arne & Carlos. I've talked about this at COIN, I think that WP:COIEDIT is awkwardly phrased, as it "strongly discourages" paid editors to edit articles themselves, getting other people to do the work for them, while getting paid all the same, which still leads to WP:BOGOF, which is about "undisclosed" paid editing. But maybe that's a discussion for another time and place.
2
The quid pro quo case with Ben (Salvidrim!) concerning AfC is what this whole thing set in motion. On October 20th, I asked Ben off-Wiki to okay the AfC on Datari Turner; this was before his involvement with Mister Wiki and was my first paid gig. It was through me Ben got involved with paid editing. On November 14th, Ben asked me to do the same for him, giving the green light on two article at AfC. This was clearly not okay and a breach of trust. Let me again say that we weren't trying to "game the system" or anything. I honestly believed that the two articles must've been notable. There was a screenshot, but I saw it can't be used as evidence.

I've been honest about my paid editing, and I haven't done any paid editing since (and it's safe to say that my relationship with Mister Wiki has gone sour). I actually got re-energized, looking up sources and creating articles, so for instance, the article I created on Dutch brewery vandeStreek on November 7th was not a paid gig, nor is the one about another brewery, which is still in my draft space. soetermans. ↑↑↓↓←→←→ B A TALK 10:25, 6 December 2017 (UTC)[reply]

Evidence presented by Softlavender

[edit]

Salvidrim's COI activity. needs to be analyzed together with his RfB, because they form a continuum of evidence of poor judgment, and disregard/dislike of Wikipedia policies. They form a continuum of loss of community trust, and during the RfB a desysop/ArbCom case were requested/suggested.

Salvidrim has lost and/or violated the community's trust

[edit]
  • In July 2017, Salvidrim nominated himself for RfB, stating as his main rationale: "as a WP:BN regular, it's where stuff with the best interesting-to-stupid-drama ratio happens" [20].
  • Many editors raised Salvidrim's stated strong dislike of privacy rules -- namely as he stated during his ArbCom run: "I've long been an anti-privacy advocate ... and as such, I've never been particularly opposed to "sleuthing" and collecting publicly available data in other to identify who's-who. ... I find that connecting the real-life identity of a Wikipedia editor with their account is a much less abhorrent and repulsive thing to do than many other people in the community. I'm not proposing anything specific w/r/t Wikipedia but I wish the general pulse of Internet users tended more towards transparency than privacy." (full context available here) [21] -- and his love of wiki-sleuthing off-wiki as very concerning; here are some but not all of those: [22], [23], [24], and at least one editor felt it incompatible with his being an OTRS volunteer [25].
  • GorillaWarfare posted two sets of evidence of disturbing off-wiki behavior under the Salvidrim screenname on Reddit: The first, from 10 months prior, both related instances of disturbingly abusive behavior as a child, and also commented on his own mental health ("I tick most boxes on most versions of said [psychopathy] checklists. I also fit neatly in the common definitions of both the 'narcissism' and 'ISTJ' boxes. This reads like a biography to me: https://en.wikipedia.org/wiki/Psychopathy_in_the_workplace :p") [26]. The second bragged about multiple recent instances of public urination and about recently both flashing himself (indecent exposure) to a taxi driver and peeing in front of the driver (subsequently deleted from Reddit) [27].
  • The material being discussed was so disturbing that Ks0stm NOINDEXed the RfB: [42].
  • During the RfB, Salvidrim publically called out GorillaWarfare on Twitter (deleted within the hour): [43], [44]; text of tweet: [45].
  • Several !voters called for or suggested a desysop because of GorillaWarfare's evidence and/or his Twitter call-out of GorillaWarfare: [46], [47], [48], [49], and an ArbCom case to desysop him was suggested at that time: [50].
  • Salvidrim withdrew the RfB 20 hours after opening it with a tally of 28/41/3: [51].

Salvidrim has been evasive during his paid/COI editing and during the investigation of it

[edit]
  • Salvadrim endlessly evaded answering direct questions in the 19/20 November discussion on Jytdog's talkpage (needs to be read in full: [52]). One other continued evasion is "contact with MisterWiki was established" -- never saying who contacted whom (did he contact Mister Wiki, or did Mister Wiki contact him?), why, or precisely when.
  • The Facebook conversation Salvidrim supplied in his defense [53] is truncated; it starts midstream and the beginning of the coversation is not shown, nor is the top of the window or the date, etc. So there is missing information that would be pertinent to the case.
  • In terms of the timeline of the various COI and meatpuppetry activities, I will submit the COIN thread which covers a great deal of ground and needs to be read in its entirety: [54].

Closing remarks

[edit]

Boing! said Zebedee said in the COIN thread: "Paid editors reviewing and accepting each other's AFC submissions shows a serious lapse of judgment that I think is incompatible with the community trust required to remain an admin - and I think the only honorable response here is to resign the admin bit and re-run for RFA." More than that, given the serious evidence and disturbing patterns brought to light in the July RfB, combined with the recent multifaceted COI evasions, errors of judgment, and policy violations, we have far exceeded the loss of community trust necessary for retaining adminship.

-- Softlavender (talk) 19:30, 14 December 2017 (UTC)[reply]

Evidence presented by {your user name}

[edit]

before using the last evidence template, please make a copy for the next person

{Write your assertion here}

[edit]

Place argument and diffs which support your assertion; for example, your first assertion might be "So-and-so engages in edit warring", which should be the title of this section. Here you would show specific edits to specific articles which show So-and-so engaging in edit warring.