Wikipedia talk:Disambiguation/Archive 38

This is an archive of past discussions on Wikipedia:Disambiguation. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 35

Archive 36

Archive 37

→

No blue and no red link

Hello. In a disambiguation can we write for example a name without a blue link (there is no article about the place because it is not notable) and without a red link (the place is not notable); If the answer is no (as Ι assume) which point of the page Wikipedia:Disambiguation show that; Xaris333 (talk) 22:04, 21 January 2013 (UTC)

I think the answer is yes on the dab policy page in case someone may wish to create an article.--Canoe1967 (talk) 22:07, 21 January 2013 (UTC)

Sorry, i was not specific. The place in not notable for example. Xaris333 (talk) 22:10, 21 January 2013 (UTC)

Well, the first sentence includes "more than one topic covered by Wikipedia articles". DexDor (talk) 22:08, 21 January 2013 (UTC)

Wikipedia:Manual_of_Style_(disambiguation_pages)#Specific_entry_types, found it.--Canoe1967 (talk) 22:13, 21 January 2013 (UTC)

Weigh in

Can I get people to weigh in at Talk:Chad (slang)? That's the first post in 3 1/2 years, so I don't forsee lots of people watching the page. CTF83! 11:23, 23 January 2013 (UTC)

RFC

There appears to be a difference of opinion on whether (a) placement of a page in a category must be supported by reliable sources, and (b) whether disambiguation pages may have explicit categories other than disambiguation categories, such as Category:Fooish surnames. The community should form a consensus on how to handle this. Carlossuarez46 (talk) 21:58, 30 December 2012 (UTC)

Article content needs reliable sources; disambiguation pages are not articles. WP:DCAT already describes the community consensus. When lists of surname-holders or lists of given-name-holders are included on a disambiguation page, it is acceptable to put the more specific Category:Fooish surnames on the disambiguation page, or to split the list of name-holders to its own article with the more specific category. Editors are welcome to split any surname-lists + surname categories from any dab pages to new surname articles, but there's no "surname article must be created" mandate nor any BLP violation (since having the category "Fooish surnames" on the page "Bar (disambiguation)" does not mean that every holder of the surname Bar is Fooish, only that some might be). Keeping short surname lists on dabs (with their categories, if any) is one of the compromises we reached through consensus. -- JHunterJ (talk) 23:02, 30 December 2012 (UTC)

The manual of style shows that dab parameters should be used. More specific disambiguation categories (such as counties mentioned above) are tolerated but adding substantive categories like Fooish surnames is just wrong without reliable sources. "Fooish surnames" is not a subcategory of Category:Surname disambiguation pages nor ought it be. I - and most readers - would expect anything categorized as Fooish surnames to have some encyclopedic content about the surname; not just a list of people and things, some of which may or may not even have the supposedly Fooish surname. As for your argument that not everyone in the dab page needs to match the category - there is no encyclopedic value to adding the cat; the user expects some content. As for your "no[] BLP violation" - then you'd agree to adding Category:LGBT people to any dab page with at least one LGBT person's article in it, since adding Category:Fooish surname is ok with one Fooish person on the page? Illogical. Carlossuarez46 (talk) 00:26, 31 December 2012 (UTC)

No. -- JHunterJ (talk) 02:18, 31 December 2012 (UTC)

Please provide a link showing where this was discussed and the consensus you claim exists reached. Carlossuarez46 (talk) 22:11, 2 January 2013 (UTC)

WP:DCAT, so probably its Talk archives. Please provide a link showing where WP:DCAT does not actually represent the community consensus, and where the consensus you claim exists was reached instead. -- JHunterJ (talk) 00:30, 3 January 2013 (UTC)

I agree with JHJ. In general, categories (other than disambig cats) should not appear on disambig pages. However, surname and given name categories are a necessary exception, because of the many cases in which proper "name articles" have not yet been split off from the disambiguation pages. Where a distinct name article does exist, that article and not the disambiguation page should be categorized in any appropriate name-related categories. --R'n'B (call me Russ) 22:42, 2 January 2013 (UTC)

Is there an assumption that the surname is notable? Carlossuarez46 (talk) 22:48, 2 January 2013 (UTC)

Not by us. But see Wikipedia:WikiProject Anthroponymy/Home backup#Background reading. -- JHunterJ (talk) 00:30, 3 January 2013 (UTC)

Articles about names, including lists of people with a name, are not disambiguation pages. Any such lists on dab pages should be split to separate pages. MOS:DABNAME sets out the difference. WP Anthroponymy is not particularly active but such splits are ongoing. It's good to see from the contents of Category:Surname disambiguation pages that the splits have been practically completed for A to Q. Once R to Z are done, there should not be a need for that category. – Fayenatic London 21:26, 4 January 2013 (UTC)

The contents of the ill-considered new Category:Surname disambiguation pages are dwindling I believe because the WP engine is catching up to the revision of the ill-considered changes to Template:Disambiguation/cat that switched Category:Surnames to the new Category:Surname disambiguation pages in late December. -- JHunterJ (talk) 21:55, 4 January 2013 (UTC)

Well, in reading the above I am thinking about a general rule of thumb regarding categories:

For an article (we're obviously talking specifically about articles and dabs of article topics here I presume?) - For an article to be placed in a category there should be a reference in the article establishing that the article topic meets the criteria for inclusion in that particular category.

As for what is "allowed" to be placed on a dab page, the criteria for inclusion is anything which helps clear up ambiguity.

The purpose of categories is to help navigation. The purpose of a dab page is to help navigation. So restricting anything sounds an awful lot like restricting navigation rather than helping it. WP:CREEP comes immediately to mind. Don't forget that an MoS (Manual of style) page is merely a guideline. Content trumps style every time. And helping readers navigate to pages obviously trumps some hierarchy of rules and regs of some prescriptive style. - jc37 17:43, 26 January 2013 (UTC)

Malformed RFC. Please do not start RFCs about two separate questions. There is no relationship at all between a) whether DAB pages can be topically categorized (obviously they can't, because it's impossible to predict what will be added to them over time, and they almost invariably end up covering more than one topic), vs. b) whether addition of a category requires sourcing (as with everything on WP, yes it does, but this is not enforced unless someone disputes it). — SMcCandlish Talk⇒ ɖ^⊝כ^⊙þ Contrib. 19:26, 26 January 2013 (UTC)
Comment I think the problem is that right now the disambiguator key places the surname part of the disambiguation into the general surnames category, not the surnames disambiguation pages category.John Pack Lambert (talk) 03:30, 27 January 2013 (UTC)

John Hendry Park

This re-directs to Trout Lake (Vancouver) but it should be the other way around. The donator of the park insisted on the name so shouldn't we follow her wishes as well? --Canoe1967 (talk) 00:41, 28 January 2013 (UTC)

This is not really an issue for disambiguation. You can ask to move (i.e., rename) Trout Lake (Vancouver) to John Hendry Park following instructions at WP:requested moves. older ≠ wiser 01:41, 28 January 2013 (UTC)

The "Primary Topic"

The primary topic should, in terms of both usage and of long term significance, be the topic any reader is much more likely to be seeking than the combination of all other topics identified for disambiguation.

I propose to insert the above definition of "primary topic". The words are taken primarily from Template:Primary Topic. Currently, there is no reasonable definition on the project page. (conseqeuently, people imagine their own definitions).

I inserted it here, but have been reverted for little more than "no consensus". Are there any substantive objections? --SmokeyJoe (talk) 00:31, 10 January 2013 (UTC)

It is much better than the present mess at WP:PRIMARYTOPIC, but will probably get pushback from those who always like to be able to pick a primarytopic even for very ambiguous terms. I'd support it, or something along that line if better suggestions come along. And that template Template:Primary Topic should be deleted; it's just Born2cycle trying to hide away a definition where he can control it; he has been advised many time to stop that harmful and confusing practice. Dicklyon (talk) 00:50, 10 January 2013 (UTC)

Pushback is good. It is an excellent way to find boundaries. I don't agree that everything Born2cycle does is evil. --SmokeyJoe (talk) 00:55, 10 January 2013 (UTC)

I support it as well (although the current text is not a mess). It probably will get pushback from those who always like to be able to ignore a primary topic even when the readership would be better served with a primary topic for the title, and who always like to be able to overqualify titles from ambiguous terms. -- JHunterJ (talk) 13:20, 10 January 2013 (UTC)

JHunterJ's edit, to

The primary topic is the topic with long-term significance or the topic the readership is more likely to be seeking than the combination of all other topics for the ambiguous title, or both.

was productive and should be reinstated. Noetica should stop reverting progress without substantive objection. --SmokeyJoe (talk) 12:51, 10 January 2013 (UTC)

I don't think the last words "or both" are needed. --SmokeyJoe (talk) 12:53, 10 January 2013 (UTC)
I'd happy to exclude ", or both". Any editor should feel free to reinstate that progress despite Noetica's reversion. That text simply restates the existing Primary Topic criteria that's still in the guidelines. The template's text is wrong; the primary topic does not need to be "both" long-term significance and usage. It also does not need to be "much" more likely than the combination of all others; only more likely than the combination of all others and much more likely than any of them. I was unaware of the template's existence until now. -- JHunterJ (talk) 13:13, 10 January 2013 (UTC)
And now that I've looked, I'm not sure I see the problem with the template. It doesn't hide anything away. It simply says whatever this guideline says. Changes here will be reflected in its output. -- JHunterJ (talk) 13:23, 10 January 2013 (UTC)

I have reverted edits made in the last thirteen hours or so (see diff). In the section "Is there a primary topic?" Together the changes affect the conditions for determining that there is a primary topic. We shifted from this:

Although a word, name or phrase may refer to more than one topic, it is sometimes the case that one of these topics is the primary topic. This is the topic to which the term should lead, serving as the title of (or a redirect to) the relevant article. If there is no primary topic, the term should be the title of a disambiguation page (or should redirect to a disambiguation page on which more than one term is disambiguated). The primary topic might be a broad-concept article, as mentioned above.

To this (change affecting the definition is underlined):

Although a word, name or phrase may refer to more than one topic, it is sometimes the case that one of these topics is the primary topic. The primary topic is the topic with long-term significance or the topic the readership is more likely to be seeking than the combination of all other topics for the ambiguous title, or both.
If there is a primary topic, the ambiguous term leads the reader to it. If the article is not titled with the ambiguous term, then the ambiguous title redirects to the primary topic. The primary topic might be a broad-concept article, as mentioned above.
If there is no primary topic, the term is the title of a disambiguation page or a redirect to the appropriate disambiguation page (if it covers multiple similar ambiguous titles).

Now, the rest of the page is unaffected. It contradicts the addition that I highlight above. For example:

There is no single criterion for defining a primary topic. However, there are two major aspects that are commonly discussed in connection with primary topics: ...
...
There are no absolute rules for determining whether a primary topic exists and what it is; decisions are made by discussion among editors, often as a result of a requested move.

And more. Such a contradiction is no good for anyone; and undiscussed moving of the goalposts for primary topics is bound to be controversial. Personally, I am against making it any easier to discover a primary topic, and I would like to see the matter discussed here with wide participation to find where consensus lies.

I have removed "the" from the section heading here, so that it does not prejudge in favour of there being primary topics by default. No accuracy is lost by that omission.

(Editors, please remember that we are in various time zones around the world. Don't expect quick responses!)

Noetica^Tea? 12:59, 10 January 2013 (UTC)

Hi Noetica,

If you feel the need for highlighting the before and after texts, would you consider doing it in side-by-side cells of a table (or similar).

I agree with you about contradictions. I don't see the contradiction, just lack of clarity.

I am also against making it easy to assert that a topic is "Primary". I think the way to do this is increase the obyectivity of the measurement of primacy. Currently, it is pretty weak, and allows any editor to assert that their favourite topic is Primary without risking a contradiction to the the guideline they bluelink.

Personally, I think a Primary Topic should be in respect to *both* usage and long term significance. Where the two contradict, we should assume that we have two groups if readers, readers interested in current, popular usage, and readers interested in more serious, academic or historic long term significance. We should not ignore either group, and if the topic is ambiguious, it should be disambiguated.

When JHunterJ changes the "and" to "or", I think it is still better than the previous versions, but that we need to find something in the middle. If one meets ~~the~~one definition, but not the other, it may be completely fine. The problem with a straight "or" is that it can in theory define two different Primary Topics for the one topic. This is a point of rare frequency, and I think it best to define a reasonable definition as a starting point.

A benchmark I have in mind is "Big". I think it absurd that this is considered to be the primary topic, from a historical perspective, across all of the English speaking world. I think it is a"Primary Topic" for Tom Hanks fans, but not for a global audience. Recent RM and Move Review discussions revealed multiple editors could claim opposite conclusions of the interpretation of WP:PIMARYTOPIC, thus revealing the ineffectiveness of the guideline.

I put back the original section heading because it is the target of incoming links. I don't think it is a problem expecting people to understand that for a given topic there is not necessarily a "Primary" topic. If you do, the guideline should say so more plainly. --SmokeyJoe (talk) 04:35, 11 January 2013 (UTC)

I agree that Big is a Big problem. Even if we agree that it's the primarytopic (due to no other article vying for that title in WP), it makes little sense to use such an ambiguous topic name as an article title, when Big (film) is so much more precise and recognizable. But the disambig guidelines are written in such a way as to encourage naming a primary topic, and having named one, using it as title. This should be fixed. Dicklyon (talk) 04:43, 11 January 2013 (UTC)

Yes. Primary Topic should not be defined in terms of existing article titles. Usage and long term significance goes beyond the standard contents of an encyclopedia. We should also think of downstream and external usage. Can the link en.wikipedia.org/wiki/Big be considered a Primary? I think Primary should be considered the antonym of ambiguous, though softer than unambiguous. --SmokeyJoe (talk) 05:56, 11 January 2013 (UTC)

OTOH, Primary Topic should not be defined in terms of downstream usage (if downstream means future; if it means something else, let me know). Future Wikipedia can handle future usage. Primary Topic is already defined in terms of external usage, of topics that have Wikipedia coverage, which is how it should be. "Primary topic" only has meaning in cases of ambiguity; if a term is not ambiguous, then it has only one topic, whether you call that position first, last, or middle. Dicklyon's disagreement with WP:PRECISION (putting qualifiers on titles that don't need them) should be addressed there. The Primary Topic guidelines do not encourage using the term as the title; they rely on the article naming guidelines for that, and if the article is named something else, the disambiguation guidelines simply say to make the ambiguous term with that topic as primary should redirect to the title given by the naming guidelines. -- JHunterJ (talk) 11:58, 11 January 2013 (UTC)

Dont start negatively

Noetica, this [revert] was a bad edit. The asserting of a negative fact in introducing something is intellectually weak. The asserted fact is also quite dubious. The whole section is a dogs breakfast, and is demonstrobably ineffective a guidance. Suggesting an important emphasis on what? And do you even appreciated the substance of the paragraph, or are you too busy elsewhere to allow editing here? Please consider the applciability of Wikipedia:Status quo stonewalling, please read Wikipedia:Don't revert due solely to "no consensus", and please tell me what part of WP:Consensus I am not following. --SmokeyJoe (talk) 04:48, 11 January 2013 (UTC)

Briefly, Smokey: Yes, I am flat-out busy with a hundred things. It doesn't help anyone, if edits are made to a central guideline (deferred to in policy at WP:TITLE) while the issues are being discussed here, and while precise applications of the guidelines are attempted at more than one current RM. There is no rush. Nothing is lost by orderly, methodical treatment here on the talkpage. This is what ArbCom has stressed for editing of guidelines recently; and it's especially important when we have strong advocates for one position or another posting both here and at affected RM discussions!

I appreciate your careful reply above. I'm being careful also; and we share concerns about these guidelines and how they have been applied, or certainly misapplied and misread in many cases.

Let others have their say, and that takes time; we all want to see the range of opinions, right? I will come back here when I have a break in my other activities. Meanwhile, keep it on the talkpage? Please?

O, and a word to the wise: linking Born2cycle's polemical custom-made essay is really not going to help keep the peace around here. Mmm-hm?

Noetica^Tea? 05:11, 11 January 2013 (UTC)

Hi Noetica. Discussion at WT:TITLE and ongoing RMs are endless. We shouldn't wait for these things to improve an obviously improvable guideline. I am well aware of the impract of WP:PRIMARYTOPIC on RMs. I actually think the lack of clarity, of definition, or objectiveness, and even of correctness, of WP:PRIMARYTOPIC is a root cause for the non-productiveness of many RM discussions.

If I have time and interest in doing this today, it is not for you to tell me to wait for something else first. Sorry. I admit to being frustrated by your BRD reversions. I respect your reversion, try to discuss (very few seem truly interested), and try again with a smaller edit. You seem prepared to revert any edit not pre-approved. I strongly object to that, it drives the life out of the project, it wraps the editing process in red tape. The most enjoyable and lively editing comes from improving on others' edits. You should only revert as a last resort. Reverts disencourage lurkers from contributing. The best wasy to encourage others to contribute is to make a small reasonable edit and wait to see what other's make of it. If there is no new discussion, it is time for a new edit.

I knew Born2cycle was an author, but don't know about it being "custom made". For you? I don't recall seeing direct interaction between you two. I try to keep up with the essays, and I liked it. It does reflect an editing philosophy obviously different to yours, but I didn't see it as offensive. --SmokeyJoe (talk) 06:20, 11 January 2013 (UTC)

It is misleading to characterize Template:Primary Topic as "Born2cycle's polemical custom-made essay". The template merely transcludes text of this guideline from the section "Is there a primary topic?" contained within the <onlyinclude> </onlyinclude> tags. As JHunterJ mentioned in the previous section, this template will reflect whatever changes are made to that text. However, that said, it certainly seems to introduce a rather curious circularity and duplication to try to insert the text from that template elsewhere into this same guideline. If there is an actual disagreement between text in the various parts of the guideline, that should be addressed, but with a little more thoughtfulness than circular duplication. older ≠ wiser 12:00, 11 January 2013 (UTC)

"custom-made essay" II think Noetica was referring to the stonewalling essay. --SmokeyJoe (talk) 12:10, 11 January 2013 (UTC)

Yes, that's right. The polemical one, not one of those for which he puts transclusion markup in Wikipedia policies and guidelines for his own purposes. Ask Born2Cycle what prompted him to write the stonewalling one. And while you're there, ask him a related question: what happened to him at ArbCom, what happened to PMAnderson (the only other editor picked out for mention), and what became of his accusations against me in the same case. (Heh. I only mention it because it will probably get relevant here, sadly.)

No time today. I'll be back in a day or so. Busy with other pages and with real life. Be consensual while I'm gone, y'all.

Noetica^Tea? 12:28, 11 January 2013 (UTC)

Ah yes, sorry for misreading. While there are aspects I agree with, the stonewalling essay is clearly a polemic. older ≠ wiser 12:57, 11 January 2013 (UTC)

There is still not a single substantive statement in support of "there is no single criterion for defining a primary topic".

What does it mean?

There is not a single criterion. It is unpredictable.
There are multiple criteria, always, never just one (why not just list the criteria).
Implicitly asserted: "primary topic" is something to be defined (but by who).

I think it is meaningless or misleading, however read. --SmokeyJoe (talk) 10:52, 12 January 2013 (UTC)

[Outdent from all] SmokeyJoe argues for reinstatement of this version: "The primary topic is the topic with long-term significance or the topic the readership is more likely to be seeking than the combination of all other topics for the ambiguous title, or both." SmokeyJoe labels this "productive", and accuses Noetica of "reverting progress without substantive objections". I find Noetica's objections substantive and have many of my own, and I do not see SmokeyJoe's preferred wording as any form of progress.
It is productive, but only of significant problems. Basically, it just doesn't parse when you break it down. As Noetica pointed out elsewhere, leading in with "The" implies strongly that there is always a primary topic, which is wrong. The wording "is the topic with long-term significance" suffers a similar logic breakdown – it expects, usually incorrectly, that there is only one such topic. However, per WP:N, any topic with an article here has to be something with long-term significance as evidenced by non-trivial coverage in multiple, independent reliable sources. So, that whole "primary topic" criterion just collapsed. Next, the "topic the readership is more likely to be seeking" clause falls apart, too. WP does not have a monolithic, homogenous user base; different subsets of the readership are seeking different things. All this talk of likelihood (even more "than the combination of all other topics"!) is all statistical language, but obviously written by someone who is not a statistician, since these propositions are not actually testable. They are nothing but a recipe for endless argument. Whether "or both" is needed at the end (it's not) is basically a moot point, like whether a corpse needs its back rubbed.
— SMcCandlish Talk⇒ ɖ^⊝כ^⊙þ Contrib. 19:45, 26 January 2013 (UTC)

SMcCandlish, you misunderstand me at least slightly, and I find your post unable to understand despite significant effort. I think you have mixed multiple lines of thought together. JHunterJ's edit was productive in that it addressed a troubled text, even if you don't completely agree. Noetica's edits were completely unproductive as they were reverts without substantive rationale, which serve to derail progress on improving the text. I am left completely confused as to what you and Noetica think WP:PRIMARY should say. --SmokeyJoe (talk) 02:08, 29 January 2013 (UTC)

I don't see the implication boogeyman that you and apparently Noetica see, so I'd see no change in meaning if we accommodated it with "The primary topic, if any,..." or "The primary topic, if there is one,..." I realize Noetica mistakenly believes that I am on the lookout for primary topics where none exist, but I obviously don't feel that there's always a primary topic (see the edit history of WP:MALPLACED). The logic breakdown, if any, or if there is one, of the topic with long-term significance could be similarly addressed; how would you like accommodate Apple leading to the fruit if the usage was overwhelmingly for Apple, Inc. (that's the type of case the long-term significance bit was added for)? The usage argument does not fall apart just because you disagree with it. Wikipedia's user base is not homogeneous, no, but it does exist and it can be measured (and tested) in aggregate, not with perfect accuracy but with useful accuracy, as any statistician will tell you. It is possibly to serve all of the subsets of readership and serve the readership as a whole best by having a primary topic sometimes and having no primary topic other times. -- JHunterJ (talk) 21:50, 26 January 2013 (UTC)
A much better statement would be "a primary topic is the main way the term is used. If in English speach when the term is used it primarily refers to the thing under discussion, it is the primary topic." The issue should be use of the term, which is actually measurable.John Pack Lambert (talk) 03:43, 27 January 2013 (UTC)
That certainly does sound like a better way to look at it, and would take care of the problems at Big and Brand New. Dicklyon (talk) 03:48, 27 January 2013 (UTC)
So far, though, we have lined up with WP:NOTDICT and used something more akin to "If when the term is used as an English-language encyclopedia article title it primarily refers to the thing under discussion, it [the thing under discussion] is the primary topic." I'd have little objection to a change to leaving articles on commonly used nouns as the base name regardless of other ambiguous topics and their popularity (Apple), but extending it to adjectives (and adjectival phrase) like Big and Brand New is problematic: there is no encyclopedia article on the in-English-speech-primarily-refers-to concept "big" at Big (disambiguation). -- JHunterJ (talk) 12:12, 27 January 2013 (UTC)
I have a thought about that. For adjectives like Anterior, we have section redirects to Anatomical terms of location, which is basically a glossary of such terms. Why not have a similar glossary of Terms relating to dimension (big, large, long, tall, wide, short, narrow, heavy, thin, tiny, etc.). Most of those are currently disambiguation pages. The ones that are not, like Big (a movie) and Large (a surname page), are not necessarily the primary uses. At least then we would have an in-project target for such terms when they get linked, and a possible primary meaning target if the dispute comes up. bd2412 T 15:14, 27 January 2013 (UTC)
That sounds like progress to me, if terms relating to dimension or whichever can be created and survive. -- JHunterJ (talk) 17:44, 27 January 2013 (UTC)

Red link, Verifiability and Notability

How can we be sure that a red link in a Disambiguation is verifiable and notable? Xaris333 (talk) 20:24, 27 January 2013 (UTC)

See the section immediately previous to this one. -- JHunterJ (talk) 11:57, 28 January 2013 (UTC)

Proposal: state the purpose of primary topic at WP:PRIMARYTOPIC

Due to the discussion at Talk:The_Wizard_of_Oz_(1939_film)#Requested_move, I've realized that there is a lack of understanding and appreciation for the purpose of recognizing and establishing primary topics on WP.

Accordingly, I propose inserting the following lead sentence to WP:PRIMARYTOPIC, to clearly state this purpose:

The purpose of recognizing primary topics is to best serve those readers using the Wikipedia Go search function when searching with a term associated with one particular topic much more than any other, by taking them directly to the article about that topic, rather than to a dab page.

Any objections? --Born2cycle (talk) 17:50, 10 January 2013 (UTC)

Yes. While I agree with the sentiment, I think the wording is terrible - I had to read it a couple of times before I understood it. Better, but not perfect, would be something like:

Many search terms can refer more than one topic, in such cases the primary topic is the one that a significant majority of people expect to find when searching for or linking to that title. For example, most people searching for "Winston Churchill" will be looking for the article about the British Prime Minister rather than his grandson or any of the various schools and ships named after him. So, to enable these people to find the information they are looking for as quickly and as easily as possible, we place the article about the primary topic at the plain title (in this case Winston Churchill) and the articles about the other topics at disambiguated titles (e.g. Winston Churchill (1940-2010), MS Winston Churchill, etc) accessible through a disambiguation page at Winston Churchill (disambiguation) which is prominently linked from the top of the primary topic article.

This is almost certainly too long, but with some trimming would I think achieve your goal. Thryduulf (talk) 18:36, 10 January 2013 (UTC)

This seems to miss B2C's point, which is that it's about the "Go" button on the search box. I think he means specifically to admit that searching and linking provide little or no reason to have a primary topic. Am I right? Dicklyon (talk) 18:50, 10 January 2013 (UTC)

Well, yeah, once links are established, primary topics don't matter for linking to work. In fact, titles don't matter for linking to work, as long as the link links to the intended article. But Thryduulf is right that a very important reason for primary topics is that when a given term X has a primary topic, editors creating a link [[X]] expect it to link (perhaps through a redirect) to the article about X's primary topic. My proposed wording misses that. I also agree the wording needs improvement in general.

But, yes, people searching with web search engines like Google, whose algorithms pay no attention to the content of URLs or web page headings (which is technically all that a WP article title is), are essentially unaffected by our choices for titles. That is, if we moved Winston Churchill to 123ABC, users Googling for "Winston Churchill" would find it exactly as well.

It's those searching (and linking) within WP, particularly with GO, who benefit when the search term with a primary topic is also the title of, or redirect to, that topic's article.

One way or another I think we need to find a clear and succinct way to explain this. --Born2cycle (talk) 19:14, 10 January 2013 (UTC)

The "Go" button was removed in the default Vector skin years ago and many editors probably don't know what "Go" refers to. And we don't need a long text for the purpose when the guideline is more important. WP:PRIMARYTOPIC currently starts:

"Although a word, name or phrase may refer to more than one topic, it is sometimes the case that one of these topics is the primary topic. This is the topic to which the term should lead, serving as the title of (or a redirect to) the relevant article."

I suggest adding one short sentence after that:

"This ensures that readers using the search box are often taken directly to the article they are looking for."

Possibly add ", and wikilinks are less likely to need disambiguation."

PrimeHunter (talk) 19:48, 10 January 2013 (UTC)

Ha! Didn't know about the missing GO button! Thanks! But I see the default action when you type in a string and press Enter is to "GO" (rather than "Search").

Anyway, I support PrimeHunter's wording/suggestion. --Born2cycle (talk) 20:07, 10 January 2013 (UTC)

I didn't know about the GO button not existing either (I find monobook a significantly more userfriendly skin for the way I work), but yes I mostly support PrimeHunter's wording, but think that the second sentence could be slightly improved as:

"This ensures that readers using several common search methods are usually taken directly to the article they are looking for."

I say several methods as it's the search box, the go button, direct URL entry (I and others do do this), and firefox's search from the URL bar (I don't know if other browsers have this function or not) all work this way. "Usually" I think seems more significant than just "often". Thryduulf (talk) 21:36, 10 January 2013 (UTC)

Okay, but how about "certain" instead of "several common"?

"This ensures that readers using certain search methods are usually taken directly to the article they are looking for."

See new section below. --Born2cycle (talk) 19:40, 11 January 2013 (UTC)

Oppose in favour of the above proposal to add a better definition of "Primary Topic". The two are too connected to have two discussions at the same time, and I think definition is more important to come first. --SmokeyJoe (talk) 04:11, 11 January 2013 (UTC)
I missed that discussion above. But having read it now, it seems to me that discussions like that would be more productive once we had clarity on the purpose of primary topic. In particular, I remain unclear on what the whole "historical importance" aspect of primary topic is about. That is, what purpose of primary topic is served by giving consideration to historical importance? --Born2cycle (talk) 19:55, 11 January 2013 (UTC)

Oppose – yes, it may be a fair representation of the purpose of primarytopic. But this purpose is often misguided, in my opinion, by putting emphasis only on the one advantage of getting many readers directly to the article they might be seeking, at the expense of all the other readers who are taken to a wrong article where they have to do extra work to find the disambig page. It's as if web search usually defaulted to the "I feel lucky" result instead of giving you options. Not a great experience, even if the majority of the time the lucky result is what you want. So, emphasizing this purpose is a step in a bad direction, in my opinion. Dicklyon (talk) 04:50, 11 January 2013 (UTC)
That tradeoff is well recognized and precisely why primary topic is only supposed to apply when a topic is "much more likely than any other topic, and more likely than all the other topics combined" to be the one being sought. --Born2cycle (talk) 19:40, 11 January 2013 (UTC)

Oppose. Born2cycle's claimed motivation: he's "realized that there is a lack of understanding and appreciation for the purpose of recognizing and establishing primary topics on WP". Hmmm. So when one's own take on a guideline is questioned in a hotly contested RM that is not going as one likes, with hard argument one is at a loss to answer, one denigrates the opposition as "confused", and rushes off to change the guideline in midstream? Sorry: not the Wikipedian way as I learned it. We agree that the guideline needs sorting out, right? Given the extensive use it is gets at RMs, we need a big RFC. And a fair, properly constructed one (a rarity these days, but the only way to avoid trips to ArbCom I think). Please though: not now. Too much else going on. Noetica^Tea? 12:38, 11 January 2013 (UTC)
No, we do not agree the guideline needs "sorting out". It would benefit from a bit of clarification about purpose; that's all. --Born2cycle (talk) 19:40, 11 January 2013 (UTC)

PrimeHunter/Thryduulf/B2C wording

Okay, what about adding the following clarification to WP:PRIMARYTOPIC as proposed by PrimeHunter and slightly modified by Thryduulf and myself above? The proposed new wording is highlighted.

Although a word, name or phrase may refer to more than one topic, it is sometimes the case that one of these topics is the primary topic. This is the topic to which the term should lead, serving as the title of (or a redirect to) the relevant article. This ensures that readers using certain search methods are usually taken directly to the article they are looking for. If there is no primary topic, ...

--Born2cycle (talk) 19:40, 11 January 2013 (UTC)

Support (fairly obviously). Thryduulf (talk) 11:01, 12 January 2013 (UTC)

Oppose, equally obviously. This ensures nothing of the sort. The dominant search methods supply prompts, and those prompts need adequate precision to direct readers reliably to the articles they want. For example, a Wikipedia internal search on "the wizard of oz" is given reliable prompts that are updated as one types, with The Wizard of Oz (1939 film) coming at the top because it is the article of most interest as measured by pageviews. Without that precision in the article's title, that prompt would not appear. The Wizard of Oz (1939 film) first appears as a prompt when one has typed only this far: "the w"! Noetica^Tea? 23:37, 26 January 2013 (UTC)

Comment - while I'm not sure the proposed addition adequately explains the purpose of the primary topic, I'm also not sure I follow what your opposing view is either, Noetica - could you rephrase your concern? You seem to be arguing a somewhat different point than is being discussed, unless I misunderstand. Are you saying that the search method requires a user to have an adequately precise ability to guess and correctly spell the desired article's title (or that of one of its redirects), without which, the search cannot ensure that it will return the desired results? If so, I think that stating that here would conflate and confuse the statement's purpose. AdventurousSquirrel (talk) 00:31, 27 January 2013 (UTC)

Well AS, let the proposal state what is meant by "certain search methods". That is very vague, and does not necessarily refer to the main ways readers find articles. I gave one particular kind of scenario, involving use of the WP search engine. Similar considerations apply with Google searching. If the proposer has neither of these in mind, let him state what he does mean, and how the intended addition is true and useful.

You might understand if you actually try typing "the w" into the search box at top right of this screen, and seeing the prompts. Then add letters progressively: "the wi", "the wiza", and so on. That's how searching actually works.

Noetica^Tea? 01:13, 27 January 2013 (UTC)

How about this?

Still looking for consensus on how to clarify the purpose of primary topic... How about adding the statement highlighted here?

Although a word, name or phrase may refer to more than one topic, it is sometimes the case that one of these topics is the primary topic. This is the topic to which the term should lead, serving as the title of (or a redirect to) the relevant article. This ensures use of this term for searching or linking leads directly to the article that is most often associated with the term. If there is no primary topic, ...

--Born2cycle (talk) 20:52, 29 January 2013 (UTC)

Discussion about deletion of dab pages

The "Deletion" section was changed earlier today to read:

There are some cases where a disambiguation page can be speedy deleted.

Any page ending with (disambiguation) that only lists two articles is unnecessary, and can be deleted - provide hatnotes on the two articles instead.
Any disambiguation page that lists only one or zero articles can be speedy deleted.
Note that for both of the above, articles means an article that currently exists. A red link does not count for the above purposes. Editor's should write the articles first and only create the disambiguation page afterward.

Any article meeting the above criteria can be nominated for speedy deletion by placing {{Db-disambig}} on the top of the page.

If a disambiguation page does not meet the above critera, it may be listed at Articles for deletion to discuss whether the disambiguation page should be deleted.

I think some of this is plain wrong and some needs further discussion, so I've reverted the change for discussion here - WP:BRD.

Points of concern:

Any page ending with (disambiguation) that only lists two articles is unnecessary, and can be deleted - provide hatnotes on the two articles instead.

No, hatnote is only needed on the page at the basename.

Could benefit from expanding to "two articles, one of which is at the basename, ...", to clarify. (Though I can't work out a situation where there would correctly be only two articles, neither at the base name).

Note that for both of the above, articles means an article that currently exists. A red link does not count for the above purposes. Editors should write the articles first and only create the disambiguation page afterward.

This is ambiguous as to whether an article with a redlink, which has a bluelink to an article linking to the topic, is "counted". I argue that such an entry in a dab page is useful and important and justifies the existence of the dab page.

I'm thinking scenarios like this: Working on a page I make a link to John xyz, a 16th century Member of Parliament. I check it and find that there's a page for a different person of that name, say a rock musician. On checking the incoming links to that page I find there are some intended for a third person of that name, a 19th century poet. I spend some time on disentangling the three individuals, turning incorrect bluelinks into piped redlinks, but am not inclined to create the pages for the parliamentarian or the poet today. We need a dab page which includes the redlinked entries John xyz (poet) and John xyz (MP), each with a bluelink to a page which mentions them: both for the sake of readers who have searched for John xyz and found the rock star, and (for the readers' sake indirectly) so that future editors will know that these are the established redlinks and won't create John xyz (writer) or John xyz (parliamentarian) instead (or will sort out the links, if they choose to do so).

The statement "Editors should write the articles first and only create the disambiguation page afterward" disparages the activities of editors who spend time working on disambiguation in this way and should not be included here.

A third category of entry needs to be considered: an unlinked entry in black type, with a blue link in the entry, for a topic mentioned in an article, such as the Maggie Andersons at WP:DABMENTION. I think that these, too, are legitimate entries in a dab page and "count" to justify its existence.

There is nothing to be gained by deleting dab pages which have been soundly constructed and which offer a blue link for each entry. Any revision of the "Deletion" section should reflect this. PamD 09:12, 25 January 2013 (UTC)

[1] was entirely out of line. The editor needs to have a look at WP:DABRL which already exists. --Joy [shallot] (talk) 11:02, 25 January 2013 (UTC)

I share PamD's reservations here. Dab articles are a help to navigation, and where there are several possible targets for a search, as with people with similar names who might be contemporaries, the most helpful thing is to list them so that the user can quickly decide whether a particular linked article is the one they want. If purists keep insisting that such lists are forbidden unless every one is populated with an article, we either remove a very real help for users or make mandatory the creation of stub articles. In my view there are too many stubs in WP which are inaccurate and sometimes positively misleading. I would much rather discourage editors from creating articles unless they were familiar with the topic and preferably had access to at least one RS, heretical as that view may be. Much better to clean up inappropriate dabs pages on a case-by-case basis than elaborate policies that create new problems. --AJHingston (talk) 12:01, 25 January 2013 (UTC)

As to "Though I can't work out a situation where there would correctly be only two articles, neither at the base name", that can happen with some projects' pages where the project's naming convention specifies the use of qualifiers where none are needed (contrary to WP:LOCALCONSENSUS, but that's a separate issue). In those cases, the disambiguation page at "XXX (disambiguation)" for articles "XXX (qualifier1)" and "XXX (qualifier2)" should simply be moved to the base name, unless incoming wikilinks or other criteria indicate a primary of qualifier1 or qualifier2. In the latter case, the article should be moved to the base name and a hatnote added (and the dab deleted)—if the move is reverted, the redirect will still have been created. Agreed on the reservations; I don't mind expanding the description of when {{db-disambig}} applies and doesn't apply, but those groups should shift with a redefinition. -- JHunterJ (talk) 12:16, 25 January 2013 (UTC)

I agree with PamD and have to thank him for attracting attention to yet another attempt to tamper Wikipedia guidelines. The change by Ego White Tray is unacceptable. Disambiguation pages with exactly two entries may be fairly useful – there were several discussions, see e.g. Wikipedia talk: Disambiguation/Archive 36 #Mandatory redirects for primary topics. Incnis Mrsi (talk) 08:02, 26 January 2013 (UTC)

Some clarifications:

When I state that redlinks don't count, that in no way whatsoever means that a redlink somehow voids a bluelink as is kind of implied. Oue style guidelines for disambiguation pages require including a relevant bluelink if the ambiguous name has no article, so this bluelink would count and the presence of the redlink would not void that. Similarly, using a name in black text with a blue link to an article with a different name is a valid bluelink and the fact that the ambiguous name itself isn't a link doesn't void the link already there. In the case of the ambiguous name, its disambiguation page should include a link to an existing article (John XYZ, member of the 135th Parliament), and this blue link is not voided because there's a red link on the same line. I didn't think my text in any way implied that black text or redlinks voided a blue link, but of course revising the text is a reasonable idea.
I never intended to suggest that we need an article about every listed topic on a disambiguation page, just that if there isn't, we should supply a relevant link.
I rescind my idea on disambigs with two topics being deleted. You're right and I was mistaken.
We should have a section on speedy deleting disambiguation pages on this page. I added this text when I placed a link to the deletion section to this page on the speedy deletion page so it could have a relevant link and put in what I wrongly thought was consensus.

In short, no harm intended, and I'm willing to draft up an alternative text that addresses these concerns. I'll post it here, not to the guideline page. Ego White Tray (talk) 13:46, 29 January 2013 (UTC)

Take two

There are some cases where a disambiguation page can be speedy deleted:

Any disambiguation page that lists two or fewer Wikipedia articles and whose title ends in "(disambiguation)" (i.e., there is a primary topic); or
Any disambiguation page that lists only one or zero articles.

Note that for both of the above, articles means an article that currently exists. A red link does not count for the above purposes. There is, however, no requirement that the title of the linked article has the same name as the disambiguation page. Any article meeting the above criteria can be nominated for speedy deletion by placing {{Db-disambig}} on the top of the page.

As in all cases, editors are encourage to find good alternatives to deletion. For example, if a disambiguation page lists only one article, does not end in "(disambiguation)" and is a valid alternative name for the article linked, it should be changed to a redirect. If a redlink on a page is a valid topic for a future article, editors are encouraged to add a link to the most closely related topic rather than deleting.

If a disambiguation page does not meet the above critera, it may be listed at Articles for deletion to discuss whether the disambiguation page should be deleted.

For now I didn't specifically state that a bluelink doesn't nullify an adjacent redlink. I don't think this needs to be stated, and I wasn't able to word it clearly, and it also has some element of instruction creep.

Thoughts? Ego White Tray (talk) 23:32, 29 January 2013 (UTC)

Disambiguation pages list Wikipedia topics (which is why red links and unlinked entries are allowed, with blue links to the Wikipedia coverage of the topics on other articles); the bullets should say "topics" instead of "articles" (with or without the mention of appropriate blue link). I don't think "A red link does not count for the above purposes" needs to be stated, but if it does, it needs to be clarified to indicate which red links are acceptable (those in use on Wikipedia articles, with a link to Wikipedia coverage of the topic in the description on the dab page) and which are not (all others). -- JHunterJ (talk) 01:51, 30 January 2013 (UTC)

We are discussing three separate issues really, the first two of which are already covered in the guidelines.

Every entry in a dab page should include one blue link, whether this is for an article on the topic being disambiguated or for an article which mentions it. This is already stated, and has consensus. (The topic being disambiguated may be the blue link, or a red link, or plain text per WP:DABMENTION). Having established this, we do not need to discuss red links in dab page entries, we can just discuss dab page entries.
A disambiguation page with a title including "(disambiguation)" and which has only two entries should be speedily deleted, with the appropriate link provided by a hatnote to the article which is at the base name. While I agree with this, it is not univerally agreed - there was a whole spate of discussion a while back because some editors felt that a hatnote pointing to a rock group or similar was an unwelcome distraction at the top of an article on a "serious" topic. Those editors would want to make a link to a dab page comprising only two entries, for the primary topic and for the rock group. They created the {{only-two-dabs}} for this purpose. The discussion seems to have been in June 2012, judging from date of creation of that template. .... Ah, found Wikipedia_talk:Disambiguation/Archive_36#The_.22santorum_rule.22 which seems to be at least part of the discussion. It was a hatnote pointing to a neologism which was felt to be disparaging to a BLP, and the current text of WP:TWODABS was drawn up at that point.
A disambiguation page which has no entries or one entry with blue links should be speedily deleted. On the face of it this would always be a candidate for speedy deletion, but on consideration there are some dab pages with useful "See also" sections, and I can imagine a dab page, at the base name, with one entry linked to an article and a group of several "See also" links to related dab pages, or to {{in title}} and/or{{lookfrom}} links for entries which are not appropriate dab page entries but could well help the reader find what they want.

This whole thing may be a case of "If it ain't broke, don't fix it". What's the problem we're trying to answer by changing the guidlines? Ego, please show examples of pages which aren't eligible for speedy deletion at present but which you believe should be. (Or, if they've all been deleted already, describe the sort of page you have in mind). PamD 09:46, 30 January 2013 (UTC)

Red link

How can we be sure that a red link in a disambiguation is notable? How can we prove it? Xaris333 (talk) 20:06, 25 January 2013 (UTC)

If the entry is Hi, I'm a red link, click through the link, then on "What links here" in the toolbox, then select "(article)" in the Namespace dropdown. If the red link is used in the article space, then it's include-able on the dab; if not, it isn't. I don't think notability enters into it. -- JHunterJ (talk) 21:06, 25 January 2013 (UTC)

A caveat, if a link seems particularly curious (for whatever reason) and there are only a small handful of links it may be worth examining the context of the redlink in the article. If there are no references for the link or if the articles containing a link do not provide any substantive content on the topic other than a passing mention, it might be preferable to simply delink the term and remove it from the disambiguation page. This may especially be the case if the editor that added the link to the disambiguation page has also added the same redlink on other articles, which can be an indication of COI editing or self-promotion. older ≠ wiser 22:09, 25 January 2013 (UTC)

But on the other hand there's WP:DABMENTION, which only requires that the topic be "mentioned" in the target bluelinked article. PamD 13:51, 26 January 2013 (UTC)

Sure, but that doesn't mean editors can't make a determination about whether the mention is meaningful. The value of disambiguation pages as navigational tools decreases in proportion to the extent they become a compendious index of any use of the words in any context however trivial. older ≠ wiser 14:29, 26 January 2013 (UTC)

It would be better for the guidance to not allow redlinks on dab pages at all as (1) if there is an article discussing the subject then the link can easily be turned blue by using a redirect and (2) it's not entirely consistent with other guidance such as WP:DABABBREV. Allowing redlinks on dab pages may have been appropriate in the early days of EnWP when there were many notable subjects without articles, but now it just makes it harder to maintain dab pages. DexDor (talk) 05:52, 30 January 2013 (UTC)

Please see points made above in Wikipedia_talk:Disambiguation#Discussion_about_deletion_of_dab_pages, especially section starting "I'm thinking scenarios...". PamD 09:48, 30 January 2013 (UTC)

What is the scope of "ambiguous"? Only ACTUAL uses on WP? Or also POTENTIAL uses on WP?

Despite the clarity apparently conveyed by the opening sentence of this page:

Disambiguation in Wikipedia is the process of resolving the conflicts that arise when a single term is ambiguous—when it refers to more than one topic covered by Wikipedia articles.

(my emphasis)

The question about the scope of ambiguity arises from time to time (for example, in this active RM proposal discussion).

Is the scope ACTUAL uses on WP? Or does it also include POTENTIAL uses on WP?

NOTE: By "potential use" I mean not yet known to actually exist as a topic sufficiently notable to be covered in WP today --Born2cycle (talk) 18:11, 30 January 2013 (UTC)

Is either position clearly supported by policy and practice? --Born2cycle (talk) 17:25, 30 January 2013 (UTC)

That depends on what you mean by "potential". If there's an average guy named Joe Smith who has a talent for songwriting but has never published any, it is possible that he could someday cut an album, rocket up the charts, and become so well known that Joe Smith would be notable. At the moment, he is, of course, "potentially" notable, but supporting an article would require gazing into the crystal ball. On the other hand, if Joe Smith has already put out a decent album or two, he is notable, and should have an article. If no article has been written, then his potential is not for becoming notable, but for his article being written. I am of the mind that if something clearly merits an article, then it should be listed on the disambig page, so that the redlink will alert readers for the need for such an article. bd2412 T 17:32, 30 January 2013 (UTC)

No, if something clearly merits an article, then it should be added to the encyclopedia (mentioned in another article or stubbed onto its own article). After that occurs, then it should be listed on the disambig page (with a red link if the mention uses a red link, with a blue link if stubbed, and with no link if the mention doesn't use a red link. -- JHunterJ (talk) 17:36, 30 January 2013 (UTC)

By "potential use" I don't mean a known use that warrants an article but doesn't have one yet (as noted above, a stub, a least, should be created for such a use)... that would be a borderline potential/actual use on the verge of being actualized. By "potential use" I mean a use that may be imagined to come into existence or become notable in the future, but is not yet known to actually exist as a topic sufficiently notable to be covered in WP today. --Born2cycle (talk) 18:09, 30 January 2013 (UTC)

Right, not those either. Another William Shakespeare might one day arise and write plays, but we won't move William Shakespeare to William Shakespeare (playwright born 1564) in anticipation of him. -- JHunterJ (talk) 18:12, 30 January 2013 (UTC)

I don't think we do use things that are not yet known to exist/be notable, generally, and it seems quite clear that it isn't per policy/guideline. Is anyone arguing that it is per policy? In the National_Pension_Scheme example that isn't quite what's going on, though; there are several other National Pension Schemes that likely warrant articles right now: Korea, Zambia, etc. Why someone didn't write those articles before proposing the RM is unclear to me. ErikHaugen (talk | contribs) 18:49, 30 January 2013 (UTC)

Actual uses on WP. Potential uses on some future instance of WP can be covered by the potential dab on that same future instance of WP. -- JHunterJ (talk) 17:36, 30 January 2013 (UTC)
Awhile ago I tried to suggest that ambiguity/primarytopic be determined based on how ambiguous the term is globally, ie not just on Wikipedia or "potential uses on Wikipedia". For example, the primary topic (in plain english, not wikipediaspeak) for "big" is obviously not the film, so we would move Big (disambiguation) -> Big. I think that feels right. If there are 1000 cemetaries called "sky lawn cemetary" and only one is barely notable enough for inclusion per GNG, it isn't really the primary topic (in plain english), so we would entitle the article Skylawn Cemetary, Springfield and maybe redirect Skylawn Cemetary to that. This feels right to me, too, because I think leaving the article at Skylawn Cemetary would suggest that this is the Skylawn Cemetary that is far and away the most significant (look what is at Paris), and in this case that isn't true. This proposal didn't get much traction. ErikHaugen (talk | contribs) 18:02, 30 January 2013 (UTC)
Disambiguation pages are Wikipedia navigation pages. We use list articles to list things as they exist globally. The primary dictionary topic of "Big" is not the film, but Wikipedia is not a dictionary. Better natural titles for things are in use now on Wikipedia, so if Skylawn Cemetery, Springfield is a better title than Skylawn Cemetery, we use it regardless of the existence of the latter, but we don't use Skylawn Cemetery (Springfield) when such a qualifier is not needed. -- JHunterJ (talk) 18:10, 30 January 2013 (UTC)
I'm not sure if it wasn't clear, but yes, JHunterJ, I realize what our guidelines say. To reiterate, I'm suggesting a change. ErikHaugen (talk | contribs) 18:23, 30 January 2013 (UTC)
Thanks for the clarification. My question is about current policy and practice, not about what related changes each of us might support. --Born2cycle (talk) 18:38, 30 January 2013 (

ErikHaugen, I was trying to explain why that change would not be an improvement. -- JHunterJ (talk) 18:41, 30 January 2013 (UTC)
Did you explain why? I honestly don't see any kind of an argument there, only statements about how we do things now. ErikHaugen (talk | contribs) 18:49, 30 January 2013 (UTC)
I'm afraid my verbose explanations don't carry the right tone, but: Disambiguation pages are Wikipedia navigation pages, so it would not be an improvement to Wikipedia navigation if we instead used them for listing global ambiguities -- we would then need another type of page for navigation. We use list articles to list things as they exist globally, so we should use list articles if we need encyclopedic information on how ambiguous a term is globally. The primary dictionary topic of "Big" is not the film, but Wikipedia is not a dictionary; navigationally, the film is the primary Wikipedia topic for the Wikipedia readers looking for an encyclopedia articled titled "Big"; if there's a better title for the article about the film, such as The film Big, we could move it there, but that does not appear to be an improvement, or if there were encyclopedic coverage of the topic (bigness) that would be primary, we could redirect Big there and move the film to a qualified title; but of the topics currently covered by Wikipedia, the film is the primary topic for "Big". Better natural titles for things are in use now on Wikipedia, so if Skylawn Cemetery, Springfield is a better title than Skylawn Cemetery, we use it regardless of the existence of the latter, but we don't use Skylawn Cemetery (Springfield) when such a qualifier is not needed, so the example of Skylawn Cemetery does not support a suggestion to change WP:PRECISION or WP:PRIMARYTOPIC. -- JHunterJ (talk) 19:36, 30 January 2013 (UTC)
Again, you're still just explaining, albeit in more detail, how we do things now. I don't see any argument for why it is better than what I am proposing. eg we would then need another type of page for navigation—but why? And to be clear it might be best to avoid discussing whether Skylawn Cemetery, Springfield would be used anyway over Skylawn Cemetery even if it was way more notable than any others; that is an important point but a distraction from the issue here. Trying to be more clear: I'm suggesting we use perhaps Skylawn Cemetery (Springfield) (say that is the only reasonable way to disambiguate, no natural option) even though there's no article at Skylawn Cemetery—but only do this sort of thing when the topic we do have an article about is barely notable enough for inclusion and there are many more topics of the same name that aren't quite notable enough for inclusion (no sig cov in RS). We can stop here if you like, though, since as I said my proposal got no traction before and is unlikely to do any better now, despite the apparent support for it at Talk:Big#Support_RM_2. ErikHaugen (talk | contribs) 20:57, 30 January 2013 (UTC)
I've ceased to be amazed by the prevalence of the ability to simultaneously believe in two or more conflicting ideas without experiencing cognitive dissonance. Most people don't seem bothered by it at all. But, then, most people don't seem to be very good at building things that don't collapse either. I believe the reasons these two observations about people exist are the same. Rigorous mental coherency is usually not required for humans to survive and breed. I suggest that is why the same people can reject your idea in a general proposal but accept it in a specific instance (Kauffner and I reject it in both contexts; sorry, but at least we're consistent. I don't know, but I'm guessing Kauffner is quite capable of building things that don't collapse...) --Born2cycle (talk) 21:44, 30 January 2013 (UTC)

This question is also central at Talk:Big#Requested move 2, where the ambiguity with a dictionary word, which has no actual use on WP, is at issue. --Born2cycle (talk) 20:39, 30 January 2013 (UTC)
It's not quite what's going on there, either. Nobody is arguing that we ought to disambiguate because something called "Big" might someday come into being that is going to be as notable as the film. The discussion there is more relevant to my little off-topic tangential proposal here that I hijacked your section with (sorry about that!) than it is to your question. ErikHaugen (talk | contribs) 21:12, 30 January 2013 (UTC)
My question here is not limited to topics that "might someday come into being". It can include topics that already exist, but are not covered on WP (usually for insufficient notability, but the actual reason for not being on WP is irrelevant). The dictionary topic "big" is not yet known to actually exist as a topic sufficiently notable to be covered in WP today. By the way, there is at least one person there arguing that the adjective definition of "big" is appropriate to be a WP topic. In theory, this opinion might gain consensus support and some day the article will be created.
As long as there is no article for a given conceivable topic, it is not an actual use on WP but a potential use. Either the scope of ambiguity with WP is existing uses on WP, or it's something broader that includes potential uses of one kind or another. --Born2cycle (talk) 21:29, 30 January 2013 (UTC)

Still, I don't think many/most of the supporters of the Big RM are thinking about it like that: "not yet known to actually exist as a topic sufficiently notable to be covered in WP today", like the adjective big might someday be notable enough so we ought to disambiguate now just in case. But yeah if you mean "not yet known to actually exist and be appropriate as a topic sufficiently notable to be covered in WP today" then I think I misunderstood you, pardon me. In any case, though, I think our guidelines are pretty clear that we don't disambiguate just because the need to disambiguate might crop up some day. I hardly ever see people make this argument, though. HaugenErik (talk) 22:42, 30 January 2013 (UTC)
Not necessarily so. As I noted in an earlier discussion on this page, for frequently linked ambiguous adjectives like Anterior, we have section redirects to Anatomical terms of location, which is basically a glossary of such terms. If we had a similar glossary of Terms relating to dimension, that would be the natural target of most such adjectives. bd2412 T 00:08, 31 January 2013 (UTC)

I agree with ErikHaugen on this. "Big" is ambiguous, and redefining things to make it not ambiguous, as B2C did in the edit to policy page while this discussion was going on, which I reverted, is not a good idea, and not an acceptable tactic. Dicklyon (talk) 23:00, 30 January 2013 (UTC)
- Can you expound on that, please? Do you believe "Big" is "ambiguous" in the plain English sense of "ambiguous", in the WP:D sense of "ambiguous" ("refers to more than one topic covered by Wikipedia articles"), or both? If you think "big" is "ambiguous" in the WP:D sense of "ambiguous", which topics (please specify), if any, covered by Wikipedia articles, do you believe are sufficiently likely to be sought by someone searching with "big" for the film to not be the primary topic? --Born2cycle (talk) 23:36, 30 January 2013 (UTC)
There is one exception that I don't think has been addressed in this discussion. There a more then a few cases where the primary topic has been endorsed as the dictionary definition even if we don't have an article. Vegaswikian (talk) 00:17, 31 January 2013 (UTC)
- Specific example? Regardless, don't you think that directly contradicts the WP:D meaning of "ambiguous"? --Born2cycle (talk) 00:25, 31 January 2013 (UTC)

Treaty of Paris

As you can see if you click on that title above, there are over 24 different articles on that List with most of their titles being 'Treaty of Paris' followed by the year. I had previously come across some Article Feedback at Treaty of Paris (1898) indicating that a reader had some confusion about all these different governmental agreements with similar names. This reader seemed to have either somehow ended up at the wrong "Treaty of Paris" article or thought they had caught a Wikipedia mistake. Keeping this AF in mind, I then went through and added hatnote linkage for the 'Treaty of Paris/List' to the individual articles that have the words "Treaty of Paris" in their title or that are understood as being a 'Treaty of Paris'. (Another editor then re-crafted my linkage into standardized forms.)
There is now an ongoing discussion about these edits at Talk:Treaty of Paris (1815) Dab hatnote. I'm just trying to help our readers and looking for some additional guidance here. If one reader has given AF at Special:ArticleFeedbackv5/Treaty_of_Paris_(1898) saying "The Treaty of Paris has to do with the American Revolution" then, obviously, they either ended up at the wrong article or they think there is only one 'Treaty of Paris' or something along those lines...why not guide them and the readers who are coming after them to additional knowledge, why not tell them "Hey, look!...there are over 20 other treaties which are all individually known as being a 'Treaty of Paris'?" Thanks, Shearonink (talk) 19:21, 6 February 2013 (UTC)

I think there was logic behind the existing rule stated at WP:DABLINKS

There is no need to add disambiguation links to a page whose name already clearly distinguishes itself from the generic term. For example, Solaris (1972 film) is clearly about one specific movie and not about any of the many other meanings of "Solaris". It is very unlikely that someone arriving there from within Wikipedia would have been looking for any other "Solaris", so it is unnecessary to add a link pointing to the Solaris disambiguation page. [italics added]

However, I think the Internet has changed and most people are not searching in the Wikipedia search box but are coming from a Google search page. If I'm Google searching for the Treaty of Paris but I don't know much else about it, Google provides me with several links to Wikipedia, but once I get there, without a hatnote, I wouldn't have any guidance on how to find the right article. I think the hatnotes are a good thing, but I acknowledge that they are against policy. I think there should be discussion about changing that policy. Thanks, SchreiberBike (talk) 21:07, 6 February 2013 (UTC)

We should generally hatlink a page to a DAB page that links it. There are so many ways someone can find themselves on the wrong page. --SmokeyJoe (talk) 21:16, 6 February 2013 (UTC)
100% agree. The guideline attempting to reduce clutter is well meaning, but problematic. Hatnotes, just as redirects, are cheap; not having one when it's needed hurts a lot more than including it for less clear cases. As the MediaWiki doesn't provide a standard navigational way to find all articles sharing the same title minus the parentheses, using hatnotes from an article with disambiguated title to the DAB page is OK for that purpose. Diego (talk) 22:18, 6 February 2013 (UTC)
Oppose, this is overlinking and cluttering pages. The Solaris quote above explains it well. Hatnote are not cheap, they are very expensive, they take the attention of the reader away. Pedro Gonzalez-Irusta (talk) 22:26, 6 February 2013 (UTC)
Comment. The 'Solaris' example (& that exact quote actually) has been mentioned before, in this discussion and also at the hatnote discussion on Talk:Treaty of Paris (1815). As I also said at the '(1815)' discussion, I am puzzled about the 'Solaris' example. Because if you go to the Solaris (1972 film) article, you will find a hatnote, pointing the way to the Solaris disambiguation page... Apparently, practical usage might have already gotten ahead of policy. In the meantime, until there is some sort of editorial consensus about this issue, I would ask that the 'Treaty of Paris' hatnotes remain in place. Shearonink (talk) 23:06, 6 February 2013 (UTC)
- I undid the linking by the IP Special:Contributions/117.5.50.140 . Pedro Gonzalez-Irusta (talk) 23:19, 6 February 2013 (UTC)

Yes, you did...you went right ahead and altered content while it is being discussed by the editorial community. I asked above that the present links be kept in place so we would all be able to see what was actually being considered. Those hatnotes have been in place since 2010, I see no harm in leaving them in place until all of us reach a consensus, whatever it might be. Shearonink (talk) 23:35, 6 February 2013 (UTC)

I'll repeat here what I said at Wikipedia_talk:Hatnote/Archive_3 on this issue: WP:NAMB could use a second look in this regard. It says (emphasis mine): "Here, the hatnote is inappropriate because a reader who is following links within Wikipedia or using Wikipedia's own search engine would not have ended up at [the unambiguous article]." But it isn't clear why we would only want to help out readers who arrived at the article in those ways, as opposed to, say, from Google. For example, a google search for "Orion" returns one Wikipedia article in the first page (for me, at least): Orion (constellation). In contravention of this guideline, this article does have a hatnote to Orion (disambiguation), which is extraordinarily helpful for those arriving at Wikipedia on this page.--Trystan (talk) 23:48, 6 February 2013 (UTC)

As I've already indicated at Talk:Treaty of Paris (1815)#Dab hatnote, the significant criteria is whether an article title clearly distinguishes itself. While it an article title may be unique, it that does not necessary mean it is clearly distinguished from other similar uses. There is nearly zero harm from including hatnotes (apart from those who claim it is clutter, which seems like nothing other than I just don't like it, and there is at the very least a marginal benefit. older ≠ wiser 00:50, 7 February 2013 (UTC)

@SchreiberBike you can not have been using Wikipedia (or is it Google?) for very long. For many years there was no reliable search engine built into Wikiepdia so it was necessary to use external search engines such as Google (Google has been widely used since before Wikipedia was on the net). Also a minor issue but it can cause problems in understanding of what you support doing: this is a guideline and not a policy, so your support for change is to guidance not to policy. -- PBS (talk) 09:54, 7 February 2013 (UTC)

@older ≠ wiser: In the case of the treaty of Paris, the year in the dab and the content of each one clearly distinguish them from each other. -- PBS (talk) 09:54, 7 February 2013 (UTC)

The reason why the individual Treaty of Paris currently have hat notes is because Shearonink added them (all?) between 02:22, 26 January 2013 and 02:44, 26 January 2013. @Shearonink you wrote above "Those hatnotes have been in place since 2010, I see no harm in leaving them in place until all of us reach a consensus, whatever it might be." (your emphasis) Then by that logic you ought not to object to the the removal of your recent addition to the Treaty of Paris articles until a consensus emerges. -- PBS (talk) 09:54, 7 February 2013 (UTC)

As the proposed change to this guideline will, if it goes through, affect tens of thousands, if not hundreds of thousands, of articles, I do not think that any change should be made without a widely advertised RfC which shows a clear consensus for the change. -- PBS (talk) 09:57, 7 February 2013 (UTC)

I suggest that anyone new commenting here has a look at Treaty of Paris. It is clear that unless the user knows and correctly types the date of the treaty they want, or if they hit a link without noticing the date (which might not be displayed), the chances of getting to the wrong article are high. That being so, what is most helpful to the user? --AJHingston (talk) 10:46, 7 February 2013 (UTC)

I don't get this. I can see how a person searching for the 1783 Treaty of Paris could (very easily) end up on the Treaty of Paris disambiguation page. But how, exactly, would they accidentally get to the Treaty of Paris (1815) article? Nonetheless, my inclination has always been that while this kind of hatnote is not "necessary", as the guideline says, they often do no harm and can be left alone, unless they contribute to excessive clutter at the top of the article. --R'n'B (call me Russ) 11:49, 7 February 2013 (UTC)

For instance you could have heard of "a" treaty of Paris in which France was involved as a party to the conflict, and you don't know what year it happened; from those data, you could end at almost any of the articles, either by clicking randomly from the DAB page or by selecting one article from a search engine. I think links to the DAB page are particularly useful when the disambiguation terms belong to the same class; usually this means they would belong in a set index. I.e. you wouldn't need to include the links to DAB if your three articles are Service (music) and Service (tennis) and Service (motor vehicle), but you need them when your terms are Battle (year 1) vs Battle (year 2), or Film (year 1) vs Film (year 2), or Location name (country A) vs Location name (country B). Diego (talk) 12:53, 7 February 2013 (UTC)

@Russ (and PBS), why presume that a reader knows what year a treaty was signed in? Many of the treaties are from approximately similar periods. How would a reader who might know a treaty that was signed during the Napoleonic period but not exactly which year distinguish between Treaty of Paris (1814) and Treaty of Paris (1815) or for that matter Treaty of Paris (1810). Similarly there is perpetual confusion between Treaty of Paris (1763) and Treaty of Paris (1783) and the latter with Peace of Paris (1783). The ability to create unique titles by appending the year is not necessarily sufficient to help readers find what they want. older ≠ wiser 13:21, 7 February 2013 (UTC)

How am I presuming that the reader knows what year a particular treaty was signed in? To the contrary, I presume that a reader who doesn't know this information will type "Treaty of Paris" in the search box, or on Google, or whatever, and end up on the disambiguation page. There, if they have a little patience, they can read through the descriptions and find that the treaty that ended the American Revolutionary War was the one signed in 1783. Again, no one has explained how the reader (who presumably doesn't know the year) is going to end up on the wrong article without going through the disambiguation page first. --R'n'B (call me Russ) 15:58, 7 February 2013 (UTC)

I presume that a reader who doesn't know this information will type "Treaty of Paris" in the search box, or on Google, or whatever, and end up on the disambiguation page That's a really strong prediction to make, and it's where your argument go awry. Neither a Google search nor a link from an external website is likely to arrive to a disambiguation page; if you try your own suggestion you'll find that the first four Google hits are for the 1783, 1951, 1898 and 1763 articles, with the DAB nowhere in sight. Heck, not even the internal Wikipedia auto-completion search shows the base name unless you type the full title, and then the DAB appears only at third place. Disambiguation pages are awfully difficult to find unless you know how to edit a URL to remove the disambiguation parentheses; you can't expect common users to do that. Thus the need to provide explicit navigation. Diego (talk) 16:13, 7 February 2013 (UTC)

Wow, I didn't realize the dab page was so well-hidden. Thanks for explaining this. Of course, "Treaty of Paris" may be a relatively extreme case because of the large number of similarly-titled articles, but it can't be unique. At least in this situation, I'd favor allowing hatnotes (of course, as already noted, the guideline as currently written just says hatnotes are unnecessary, not that they are necessarily to be removed). --R'n'B (call me Russ) 18:40, 7 February 2013 (UTC)

@PBS, As the proposed change to this guideline will, if it goes through, affect tens of thousands, if not hundreds of thousands, of articles, this seems a bit of an exaggeration. First, I'm not even sure a change to the policy is needed, except perhaps to the example used. The criteria of clearly distinguishes itself from the generic term is rather subjective (unless you want to take a hard-line view that a unique title is a unique title and disregard any consideration for readers). older ≠ wiser 13:27, 7 February 2013 (UTC)

Support the addition of dab hatnotes where the title fails to clearly distinguish itself as per older ≠ wiser. I do not see this as a change. The issue is does a mere date, or other datum, in a title provide enough context to make a rational decision without knowing the alternative? My answer is "Seldom". --Bejnar (talk) 11:08, 8 February 2013 (UTC)

Remove primary topic for proper names

Proper names are best treated as to never be primary topics. Why discuss whether village A (15 people) in country X is more important than village B (10 people) in country Y. And if village A one day is growing to 30 and village B to 60 then who changes all links? This is nonsense work, and outside incoming links cannot be changed by WP editors anyway. It leads to many bad links. If a title is ambiguous and not an English language word, then having a disambiguation at that title removes thousands of errors from Wikipedia like this, which lead the reader to Putumayo Department, before my intervention. Thank you. Pedro Gonzalez-Irusta (talk) 22:33, 6 February 2013 (UTC)

Absolutely not. Proper names which have only one use on WP should definitely be the titles of, or redirects to, the article about that use. See the previous section. So I disagree with your opening statement out of the gate. In addition, many proper names, even though ambiguous, really do have a primary topic. Examples: Nixon/Nixon (disambiguation), San Francisco/San Francisco (disambiguation), Hitler/Hitler (disambiguation), The War of the Worlds/The War of the Worlds (disambiguation), etc. etc. That's just a few off the top of my head, but there are countless examples of proper names that have primary topics, and are properly treated as such. --Born2cycle (talk) 22:47, 6 February 2013 (UTC)

"Proper names which have only one use on WP should definitely be the titles of, or redirects to, the article about that use." - Why? Did you ever consider any of:

article naming stability?
internal link correctness?
outside incoming link correctness?

Pedro Gonzalez-Irusta (talk) 22:50, 6 February 2013 (UTC)

I don't understand what you're talking about. How can it be that a proper name with only one use on WP is not the title or redirect to the article about that use (treated as primary topic)? What do you want it to be instead? The title of a dab page with only one entry? I ask you for a single example so I can understand what you mean. --Born2cycle (talk) 23:01, 6 February 2013 (UTC)

I am talking about current and future ambiguous titles that are proper names and have only one of the current or future meanings covered in Wikipedia. Pedro Gonzalez-Irusta (talk) 23:07, 6 February 2013 (UTC)

I'm baffled as to why you are refusing, reluctant or unable to provide a single concrete example of what you're talking about. If it's so difficult to come up with an example, why are we even talking about this?

That said, I ask again: what do you propose we do with the proper name in such a situation? Here are the choices:

Make the proper name the title of the article about the only topic on WP to which that name refers (treat the topic as primary for that proper name)
Make the proper name a redirect to the article about the only topic on WP to which that name refers (treat the topic as primary for that proper name)
Make the proper name the title of a dab page listing the only topic on WP to which that name refers (generally considered absurd)
Make the proper name a redirect to a dab page listing the only topic on WP to which that name refers (also generally considered absurd)
Something else???

Now, please be clear about what you're advocating. 1, 2, 3, 4 or 5, and, if 5, what exactly? --Born2cycle (talk) 00:05, 7 February 2013 (UTC)

I would rather say it is absurd that Nixon redirects to a person, when there a places named so. I would choose, of course "3". That clearly shows that currently there is only one topic by that name covered in WP. And put a note: "If you know other topics by that name, please list them here". This could result in a massive growth of WP. Pedro Gonzalez-Irusta (talk) 01:29, 7 February 2013 (UTC)

So you think Obama (disambiguation) should be moved to Obama? Obama currently is a redirect to Barack Obama. --Born2cycle (talk) 01:53, 7 February 2013 (UTC)

Yes, several topics are named "Obama". Having a page about one person at that title, or using it as a redirect to one person, is one of the most obvious cases of recentism. Pedro Gonzalez-Irusta (talk) 01:59, 7 February 2013 (UTC)

You may look at La Paila (currently red) and http://www.geonames.org/search.html?q=la+paila listing several "La Paila" to see that if not the general primary topic is the first to make it into an article, then there will be at least one move if a topic that is more "primary" is added later. And the definition of what is PT based on the article set will change. Pedro Gonzalez-Irusta (talk) 02:52, 7 February 2013 (UTC)

Some side questions. Which approach would reduce wasted discussions on the various talk pages? Of course the other question would be in how many cases would not having a primary topic hurt the readers or the encyclopedia? Vegaswikian (talk) 02:40, 7 February 2013 (UTC)

Number "3" would massively reduce the number of discussions about which topic can be under a plain name. Pedro Gonzalez-Irusta (talk) 02:52, 7 February 2013 (UTC)

Sets involved

names of places (geonames says it contains 8 million geographic features, Wikipedia:Size comparison#Geography)
names of persons (Thomson-Gale's Biography Resource Center contains over 335,000 essays, Wikipedia:Size comparison#People)
area codes for dialing, e.g. 020 (disambiguation)

Pedro Gonzalez-Irusta (talk) 03:26, 7 February 2013 (UTC)

I don't know much about phone codes, but the first two are clearly not appropriate for automatic disambiguating. Obama links to the president, but not because of recentism, but because no one else named Obama has held as high a government position as Barack Obama Jr does. Similarly, there is a reason London points to the city in England, as pretty much everyone knows that to find the city in Canada, you need to say London, Ontario. Ego White Tray (talk) 13:17, 7 February 2013 (UTC)

What would be a stable link? Take George Washington, for example. Suppose we move that to George Washington (president); but then suppose that at some point in the future, another person whose name is "George Washington" gets elected President, then we still have to move the article again. So much for stability. bd2412 T 15:20, 7 February 2013 (UTC)

There are known knowns. Pedro Gonzalez-Irusta has referred to cases where there are two (or more) villages with the same name; I have referred earlier to cases of two historic buildings. In each case only one has an article. We are probably both thinking of actual examples. The question we pose is how can the community best help users (and guide editors)? --AJHingston (talk) 15:46, 7 February 2013 (UTC)

For WP:TWODABS situations, there is one link of the two that the reader will be looking for at least half the time. If that title is a disambiguation page, then the reader must figure out which meaning to go on to. In other words, they would type or click the name and get taken to a page from which they need to make another click to get where they really want to go. Where there are only two possible meanings, having the more likely search target at the undisambiguated title means that at least half the time, the reader landing on that page does not need to click any further, because they have found exactly what they are looking for. In that situation, the reader who is looking for the other meaning will find it right there in the hatnote, so they are not inconvenienced any more than they would be if they had been taken to a disambiguation page. This works just as well, in fact, where there are three meanings with one being the most likely search target, if the other two can neatly fit in a hatnote. bd2412 T 18:00, 7 February 2013 (UTC)

That is good logic. But mixed up into this very entwined debate is the idea that if there is only one article on a topic in WP the title should not be disambiguated, because no account should be taken of anything outside WP, and we should not anticipate articles even on notable topics. For the purposes of illustration, suppose that there are several historic buildings called Honeysuckle Cottage. All meet notability criteria. An editor who is aware of this creates an article on the one that he or she knows about. Should the title be in a form that distinguishes the building from others, should the disambiguation be in the lead sentence, or should we leave the user to work out for themselves that the building cannot be the one that they are looking for bearing in mind that may not be readily apparent (and there are many topics where this willl be more difficult than with places). And if disambiguation should be in the title (which is probably the easiest for users) do we want to require the creation of stub articles for the others, bearing in mind the disadvantages of that approach? --AJHingston (talk) 18:34, 7 February 2013 (UTC)

And at least half of the time people see a HATNOTE that they are not interested in at all. Hatnotes clutter pages and are ugly. Beside some string match, there is mostly no other relation to the article. Pedro Gonzalez-Irusta (talk) 20:14, 7 February 2013 (UTC)

I think that is the point. People get there because the string matches, and in order to pick between otherwise unrelated articles, they need to know what the options are. Otherwise it is just a search and reject operation. --Bejnar (talk) 11:18, 8 February 2013 (UTC)

There will be hatnotes either way; either to direct visitors to the non-primary topic, or to the disambiguation page. bd2412 T 22:16, 8 February 2013 (UTC)

There does not need to be a hatnote for the disambiguation page, if one follows the Solaris example. Pedro Gonzalez-Irusta (talk) 03:21, 9 February 2013 (UTC)

Obama recentism and other presidents

Look here and here to see that "Obama" in 2004 used to be about a place in Japan. BTW, I have never heard about this 180 years dead John Adams. How many people are actually looking for any of the other listed at John Adams (disambiguation)? Pedro Gonzalez-Irusta (talk) 20:11, 7 February 2013 (UTC)

Incoming link stability

Please consider incoming link stability in the disambiguation policy. "Primary topic"-moves are against this. Pedro Gonzalez-Irusta (talk) 22:41, 6 February 2013 (UTC)

Examples for wrong incoming links

Putumayo

Several links pointed to "Putumayo" but were not meant to lead to the Putumayo Department. The Putumayo region is much older and larger than the recent department. It is recentism to have Putumayo point to Putumayo Department. Pedro Gonzalez-Irusta (talk) 01:56, 7 February 2013 (UTC)

That doesn't mean we create a disambiguation page with only one link on it. That would be ridiculous. The better answer might be to create an article about these earlier Putumayos and then change Putumayo to a disambiguation page once there is more than one topic. Another solution is to add a good history section explaining these earlier Putumayos, when they existed, and what area they covered.

And I don't consider it a case of recentism. China, Greece, Rome and Egypt have thousands of years of history, but users typing in the name of the country expect to see the modern-day country in the article, not a historic one. Most of our readers understand to type in Ancient China, Ancient Greece, Ancient Rome or Ancient Egypt for the historical places. Modern day places with a name are usually (but not always) primary over historical places. Ego White Tray (talk) 04:28, 7 February 2013 (UTC)

Putumayo (disambiguation) already exists. Way back in 2004, the disambiguation page was moved to have Putumayo redirect first to Putumayo, Colombia. It was RFDed with intent to move Putumayo, Colombia to the shorter name, but was kept as a redirect. The target later changed to Putumayo Department and to Department of Putumayo and back again. In 2010 it was changed to redirect to the disambiguation page for seven days, but that was reverted with rationale of there being indications of a primary target. Then the only other change until very recently was to change target to Putumayo department. It may well be that there is no longer a primary topic for the term, but that should be determined through discussion at Talk:Putumayo (disambiguation), preferably through listing a requested move. older ≠ wiser 06:20, 7 February 2013 (UTC)

I am not interested in moving. I just want to point Putumayo to the disambiguation page. There were several false incoming links. Does anyone care about reducing false links? Pedro Gonzalez-Irusta (talk) 20:01, 7 February 2013 (UTC)

Yes, you are interested in moving it, because if you just change the redirect, you'll create a WP:MALPLACED disambiguation page, which would then be undone. You should also assume good faith. -- JHunterJ (talk) 20:11, 7 February 2013 (UTC)

No. Pedro Gonzalez-Irusta (talk) 20:22, 7 February 2013 (UTC)

My mistake. I assumed you wanted to work with the community. -- JHunterJ (talk) 20:27, 7 February 2013 (UTC)

That I like to work with others is shown at your talk. Not being interested in moving does not mean I am not interested in working with the "community". I could be interested in removing MALPLACED and in changing the redirect. That is a prove your claim was wrong. Pedro Gonzalez-Irusta (talk) 22:18, 7 February 2013 (UTC)

Glad to hear it; the response "No" to the suggestion to follow guidelines was evidence that you didn't want to follow guidelines. If your plan is to form new consensus to change WP:MALPLACED and then change the redirect, that's fine (although unlikely to succeed). -- JHunterJ (talk) 12:18, 8 February 2013 (UTC)

I have to agree with Pedro on this one; MALPLACED is not a project guideline nor policy - at most it's a Wikiproject guideline, that is not binding to editors outside the project. PRIMARYTOPIC accepts either placing the disambiguation page at the base name or as a redirect to the DAB. MALPLACED says "The acceptance that malplaced dab pages should be corrected is outlined at Wikipedia:Disambiguation", but I don't see evidence for it at this guideline; there's no reason outlined for why this "correction" is a good idea. Diego (talk) 15:02, 8 February 2013 (UTC)

This page: "If there is no primary topic, the term should be the title of a disambiguation page (or should redirect to a disambiguation page on which more than one term is disambiguated)." -- JHunterJ (talk) 16:31, 8 February 2013 (UTC)

Exactly what I said - redirecting the title to the disambiguation page is accepted by the sentence you quote, so the change that Pedro proposed (redirecting Putumayo to Putumayo (disambiguation)) would not be against the guideline and there's no reason why it had to be reverted. Either that or moving the DAB to the base name would be equally valid, and neither WP:D nor WP:MALPLACED states any reason to prefer one upon the other. Diego (talk) 16:47, 8 February 2013 (UTC)

You are misreading that sentence. If there is no primary topic for "Putumayo", "Putumayo" and not "Putumayo (disambiguation)" should be the title of the disambiguation page (or should redirect to the disambiguation page on which another term is disambiguated)." Since we're not redirecting to a disambiguation page for another term that also covers the term "Putumayo", "Putumayo" and not "Putumayo (disambiguation)" should be the title of the disambiguation page if there is no primary topic for "Putumayo". Exactly what I said, and exactly the message of WP:MALPLACED as well. -- JHunterJ (talk) 18:52, 8 February 2013 (UTC)

I agree with JHunterJ interpretation of the guidance. So, what is the easiest and best procedure to follow to get that result? Is it a discussion at Talk:Putumayo (disambiguation), initiated by listing a requested move to Putumayo? Or is it a discussion at Talk:Putumayo department, initiated by requesting that it is not an appropriate primary topic for the term, since that is where more readers/editors are likely to see it? Or do you request the move at Talk:Putumayo (disambiguation) and notice the request at Talk:Putumayo department? Or is the discussion better here where engaged editors are more likely to be familiar with the guidelines? Or do you just notice the discussion here to get expert opinion? Would it be appropriate in some guideline or other to layout a best practices for something like this? There is a related issue with Narayanpur, which, IMHO, should have no primary. --Bejnar (talk) 22:05, 8 February 2013 (UTC)

I think the approach of a requested move at Talk:Putumayo (disambiguation) to move that page to Putumayo would be the correct one. But a pointer to the discussion should be added at Talk:Putumayo department. -- JHunterJ (talk) 22:58, 8 February 2013 (UTC)

Is there any reason at all why such DAB pages should be moved to the base name when they're located at "Name (disambiguation)"? Without a reason, MALPLACED looks like instruction creep. I don't mean that there isn't a reason - there might be one-, only that the guidelines don't mention any. Diego (talk) 23:26, 8 February 2013 (UTC)

There's an archive of the past talk discussions. Rather than assuming anything you disagree with is unreasoned, you might check it. http://en.wikipedia.org/wiki/Special:Search?search=malplaced+OR+title+all+disambiguation+pages+with+%28disambiguation%29&prefix=Wikipedia+talk%3ADisambiguation%2F&fulltext=Search+archives&fulltext=Search -- JHunterJ (talk) 23:31, 8 February 2013 (UTC)

You'll agree that search results make bad guidelines. If there is a reason why a disambiguation page wihtout (disambiguation) should be preferred, it should be listed in the guideline to guide people reading it, not buried under a talk archive that may or may not show consensus for it. Diego (talk) 17:05, 12 February 2013 (UTC)

You'll agree that the guidelines are in the guidelines even in the absence of the search results, and the reasons could be found in the search results. If you have a suggestion for how to add an articulation of that reason to the guidelines, please propose it. But guidelines that happen to omit their reasons are not any less guidelines. -- JHunterJ (talk) 17:29, 12 February 2013 (UTC)

Note: a discussion touching on this has come up at Talk:Ford#Redirect to disambiguation page. older ≠ wiser 17:59, 12 February 2013 (UTC)

@JHunterJ: First I'd want to know what is the purported reason that should be included in the guideline. If you're referring to this one, it didn't reflect consensus as no consensus was formed, and no reason was given to prefer one form over the other - only to avoid having both at the same time (with which I agree). Diego (talk) 18:15, 12 February 2013 (UTC)

To the best of my recollection, the consensus has been that "X" redirecting to "X (anything)" is pointless, which puts disambiguation pages in line with articles (WP:PRECISION). If you'd like to retain my research services to go read the archives for you, we can discuss that off-Wiki. -- JHunterJ (talk) 18:23, 12 February 2013 (UTC)

Just a few other places this has been discussed to some extent.

The topic has come up many many times on this and other pages though. older ≠ wiser 18:29, 12 February 2013 (UTC)

Thanks to JHunterJ for pointing me to the fact that a move request may be the fastest way for a change. Thanks also to Bejnar for listing the options where to start. I now added a pointer from the current kind of primary article Talk:Putumayo department#Primary topic contested. Pedro Gonzalez-Irusta (talk) 01:30, 9 February 2013 (UTC)

Examples for Primary Topic randomness

Sometimes a town sometimes the larger region around sometimes the country sometimes dab

For Tacna the city gets the place of the plain name. But doesn't have the Tacna Region more inhabitants? Chihuahua is a disambiguation page, neither the city nor the state get the primary title. Mexico it is the country, not the state, not the city. Veracruz it is the state, not the city inside, and despite several other meanings at Veracruz (disambiguation). Let's assume a country get's precedence, oh no, there is Georgia, which is a dab, despite the country and the state. Pedro Gonzalez-Irusta (talk) 00:50, 8 February 2013 (UTC)

This not really a problem, and consistency is not desirable in this situation. Common use varies as to whether a city, region or country is more commonly understood for each name; other times, there are uses that are not geographic competing for the same word (Chihuahua may very well be a dog, so either the state or the city would be problematic). The best solution is to let local consensus decide what is best for each case, guided by the common criteria mentioned at PRIMARYTOPIC (frequency of use, long term significance, derivation of several terms from the original) as well as any other that may be relevant. Diego (talk) 09:30, 8 February 2013 (UTC)

Why is it best to let random group of people to make inconsistent decisions? Pedro Gonzalez-Irusta (talk) 01:23, 9 February 2013 (UTC)

Why should be good to force the same outcome for different situations? Diego (talk) 14:56, 12 February 2013 (UTC)

Exactly. The decisions are made from consistent reasons; applying those reasons consistently in different situations can (and should) yield different conclusions. This is not "inconsistent decisions"; this is "absence of foolish consistency". -- JHunterJ (talk) 14:59, 12 February 2013 (UTC)

With respect to Georgia, note that the state of Georgia is not within the country of Georgia, and in fact, the state was for a brief period effectively a country of its own (in the time between the American Revolution and the ratification of the U.S. Constitution), while the country was, for a time, effectively a state of the Soviet Union. I can't think of a comparable circumstance. bd2412 T 22:44, 8 February 2013 (UTC)

And some years back Mexico was not a country. But point taken, that for Georgia the state and country are not in the same area. Pedro Gonzalez-Irusta (talk) 01:23, 9 February 2013 (UTC)

cc	Name	primary	dab	main region/state	main city	other
US	Mississippi	state	Mississippi (disambiguation)	Mississippi	-	Mississippi River
MX	Chihuahua	dab	Chihuahua	Chihuahua (state)	Chihuahua, Chihuahua	Chihuahua (dog), Note: it is a breed not one dog that is named Chihuahua
MX	Veracruz	state	Veracruz (disambiguation)	Veracruz	Veracruz, Veracruz
MX	Yucatán	state	Yucatán (disambiguation)	Yucatán Peninsula (larger than the state)	-
MX	Puebla	state	Puebla (disambiguation)	Puebla	Puebla, Puebla	Note: two hatnotes
PE	Tacna	city	Tacna (disambiguation)	Tacna Region	Tacna	Tacna District
PE	Lima	city	Lima (disambiguation)	Lima Region	Lima	Note: If the region would be labelled "state", would then the state be primary?
VE	Miranda	dab	Miranda	Miranda (state)
BR	Rio de Janeiro	city	Rio de Janeiro (disambiguation)	Rio de Janeiro (state)	Rio de Janeiro
BR	São Paulo	city	São Paulo (disambiguation)	São Paulo (state)	São Paulo
BR	Santa Catarina	dab	Santa Catarina	Santa Catarina (state)
CL	Santiago	city	Santiago (disambiguation)	-	Santiago	many other localities and also a personal name
AR/ES	Córdoba	dab	Córdoba	Córdoba Province, Argentina	Córdoba, Argentina	several other places
IN	Narayanpur	town	Narayanpur (disambiguation)	Narayanpur district	Narayanpur	several other localities

Pedro Gonzalez-Irusta (talk) 02:39, 9 February 2013 (UTC)

Sometimes dab sometimes person sometimes name sometimes other things

When creating this list, the page that appeared after typing Ford was a surprise.

Name	primary	dab	name	Note
Washington	dab	Washington	Washington (name)
Adams	dab	Adams	Adams (surname)
Jefferson	dab	Jefferson	Jefferson (surname) (only listing), Jefferson (given name) (only listing)
Madison	dab	Madison	Madison (name)
Monroe	dab	Monroe	Monroe (surname)
Jackson	dab	Jackson	Jackson (name)
Harrison	dab	Harrison	Harrison (name) (about given and surname)
Tyler	name	Tyler (disambiguation)	Tyler
Polk	dab	Polk	none
Taylor	dab	Taylor	Taylor (surname), Taylor (given name), List of people with surname Taylor
Fillmore	dab	Fillmore	none
Pierce	dab	Pierce	Pierce (surname)
Buchanan	dab	Buchanan	Buchanan (surname)
Lincoln	dab	Lincoln	Lincoln (surname)
Johnson	name	Johnson (disambiguation)	Johnson
Grant	dab	Grant	Grant (name)
Hayes	dab	Hayes	Hayes (surname)
Garfield	dab	Garfield (disambiguation)	none	Hatnote link to Garfield (character)
Arthur	name	Arthur (disambiguation)	Arthur, Arthur (surname)	Note: primary page has image of a statue
Cleveland	city	Cleveland (disambiguation)	not found on the dab page
Roosevelt	dab	Roosevelt	Roosevelt (surname) (flower image with caption "The native European rose."), list is at List of people with surname Roosevelt
Wilson	dab	Wilson	Wilson (surname), List of people with surname Wilson, List of people with given name Wilson, Note: no given name article linked
Kennedy	dab	Kennedy	Kennedy (surname), Kennedy (given name)
Nixon	redirect to president Richard Nixon	Nixon (disambiguation)	Nixon (surname)
Ford	redirect to Ford Motor Company	Ford (disambiguation)	Ford (surname)
Carter	dab	Carter	Carter (name)
Bush	dab	Bush	Bush (surname)
Clinton	name (family name)	Clinton (disambiguation)	Clinton
Obama	redirect to president Barack Obama	Obama (disambiguation)	Obama (surname)	Look here and here to see that "Obama" in 2004 used to be about a place in Japan.

Pedro Gonzalez-Irusta (talk) 02:39, 9 February 2013 (UTC)

Note about hatnotes

The question about rather to have hatnotes on pages with disambiguators in the title was also discussed at Wikipedia Talk:Hatnotes. After seeing no support for such a ban in the long discussion, I deleted the section telling people not to include such hatnotes. No one has object in the over 12 hours since I did this, and I don't expect anyone to. I'll check this page to remove any reference from that guideline here as well. Ego White Tray (talk) 04:02, 8 February 2013 (UTC)

I completely agree. In particular, where there are a large number of people are places sharing a name, even the disambiguators can leave some room for confusion. bd2412 T 22:47, 8 February 2013 (UTC)

But more tweaking of Wikipedia:Disambiguation is needed - too late at night to come up with suggestions myself, but (a) the scenarios under "Hatnotes" don't include "On a secondary topic page for a term that has a disambiguation page". (And the case of "two other topics" isn't covered either). and (b) the newly-minted first bullet point under "Usage" needs a bit of fixing. Perhaps the wording should be drafted here on the talk page. See you tomorrow, unless the Real World has more to offer. PamD 00:06, 9 February 2013 (UTC)

I disagree. If there is a dab link, this is sufficient. Hatnotes are linking topics not related with each other beside a string match. Each link on top of an article that links to some offtopic page is distracting readers. Pedro Gonzalez-Irusta (talk) 02:43, 9 February 2013 (UTC)

The dab hatnote is only off-topic when the "hit" is not a "false drop". Sometimes the article reached is off-topic and the article linked in the dab hatnote is where the reader wished to go. This is true even if the article title has a disambiguator (such as a date) in the title. In such a case, not only is the hatnote not distracting, it is positively encouraging. As I understand it, the concept of primary topic is that a majority of searchers will be happy with the default because that is the topic for which they were looking, and the purpose of dab hatnotes is to help the minority who weren't. Whether the dab hatnote points to the disambiguation page, or to an alternate topic doesn't matter in terms of the hatnote's existence. Question: Where would you put the dab link if not in a hatnote? --Bejnar (talk) 05:18, 9 February 2013 (UTC)

This was about situation where there is already a disambiguation link in the hatnote. So, here I did not argue for the disambiguation link to be removed. But off topic links pointing directly to other topics should go away. Offtopic stuff should be reduced to a minimum. Because it is off topic and distracting. Pedro Gonzalez-Irusta (talk) 06:25, 9 February 2013 (UTC)

Where are these off-topic links? --Bejnar (talk) 07:10, 9 February 2013 (UTC)

On the top of the articles. Before the actual content. E.g. Veracruz, Veracruz has

This article is about the city. For the state, see Veracruz.
For other uses, see Veracruz (disambiguation).

"Veracruz, Veracruz" is not even a name for the state. Pedro Gonzalez-Irusta (talk) 17:09, 9 February 2013 (UTC)

Maybe not, but search engines do weird stuff sometimes. A person who types Veracruz in a search engine might be sent to Veracruz, Veracruz, and since the state is one of the most likely alternate destinations, the hatnote isn't crazy. It's not about correctness - it's about whether someone was likely sent to the wrong article. In this case, I think this is likely. Ego White Tray (talk) 07:12, 12 February 2013 (UTC)

RFC: Shall we clarify the meaning of "ambiguous" consistently on this page?

The first line of this page defines "ambiguous" as follows:

"when [a single term] refers to more than one topic covered by Wikipedia articles."

Note the definition of ambiguous is clearly stated in terms of whether there is "more than one topic covered by Wikipedia articles". No other definition for "ambiguous" is given on the page.

It seems to me that it is unreasonable to assume that another meaning of the word is intended anywhere else on the page, especially without an explicit redefinition. Yet some people seem to choose to ignore the specific meaning clearly provided at the top of the page, and assume a broader definition, particularly when interpreting WP:PRIMARYTOPIC.

To remedy this, I added some clarification to that section, essentially re-iterating the definition above. But this was reverted with the claim that my edit was "redefining things" ("Let's get consensus first before redefining things so narrowly").

It seems to me that the only redefining going on here is by those who seek to interpret ambiguity differently at WP:PRIMARYTOPIC than how it is clearly defined at the top of the page.

So, I'm seeking support for this minor clarification (really redundancy) to make sure "ambiguity" is interpreted throughout this page consistent with how it is defined in the first line of the page.

For clarity, I'm seeking to add the following highlighted words to the text of WP:PRIMARYTOPIC:

Although a word, name or phrase may refer to more than one topic covered in Wikipedia, it is sometimes the case that one of these topics is the primary topic. This is the topic to which the term should lead, serving as the title of (or a redirect to) the relevant article. If there is no primary topic, the term should be the title of a disambiguation page (or should redirect to a disambiguation page on which more than one term is disambiguated). The primary topic might be a broad-concept article, as mentioned above.

There are two major aspects that are commonly discussed in connection with primary topics:

A topic on Wikipedia is primary for a term, with respect to usage, if it is highly likely—much more likely than any other topic on Wikipedia, and more likely than all the other topics for that term on Wikipedia combined—to be the topic sought when a reader searches for that term.

A topic on Wikipedia is primary for a term, with respect to long-term significance, if it has substantially greater enduring notability and educational value than any other topic on Wikipedia associated with that term.

Please indicate your support or opposition, reasoning, and any comments or suggestions, below. Thank you. --Born2cycle (talk) 00:06, 31 January 2013 (UTC)

Comment. I've listed this RfC at Wikipedia:Centralized discussions following WP:PROPOSAL#Good practice for proposals (wide-ranging effects should normally be listed at WP:CENTRAL) and Wikipedia:PUBLICIZE (consider being bold), as I consider that the proposed change can affect every move discussion involving a disambiguation page from now on. A change to several parts of the policy at once - with an emphasis on WP:PRIMARYTOPIC, as the proponent suggests - needs to be informed by a community-wide consensus. It's important to disclose that the RfC and this very publicizing note are made by editors involved in several open WP:Requested move discussions involving PRIMARYTOPIC. At least Big, The Wizard of Oz, Doctor Zhivago and Brand New are discussions that I'm aware of currently open or which have been cited as relevant examples there by several of the editors commenting below, but there may be more. Diego (talk) 21:01, 31 January 2013 (UTC)

Support. Disambiguation is about navigation of the information on Wikipedia, to help users find the information on Wikipedia that they're seeking, if it exists on Wikipedia. Disambiguation is not about listing or accounting for every meaning of a term. The proposed change appears to clarify that point. Theoldsparkle (talk) 17:40, 31 January 2013 (UTC)

Oppose There are two problems here which are largely about a narrow cultural perspective in an international encyclopedia with much wider aspirations. Firstly there are still a surprising number of gaps in Wikipedia, and too many articles treating only one specific example or context (usually US) of a term. Ignoring everything not yet in WP is not necessarily desirable. Secondly, statistics may only reflect the behaviour of one sub-set of potential users. If a word is used as the the title of or character in a television programme very popular with US teenagers it may generate a very large number of page views on that topic, but it really can be very confusing for everyone else looking for other and more established uses of the term if they are taken to an article about that programme. Arguments take place over these things because people do not agree; trying to make rules to stop them just generates more debate. Inconsistency may just be the result of going for the best answer. --AJHingston (talk) 18:32, 31 January 2013 (UTC)
- What, exactly, is not desirable about ignoring everything not yet in WP in the context of deciding titles? In particular, what stated/recognized goal/purpose of titling in WP is compromised by ignoring uses not yet in WP? Please note, we're not talking about recognized topics that should have articles but just don't yet - stubs, at least, should be created any time those are encountered.
  Yes, statistics might favor one subset of users - the subset otherwise known as most users. Isn't that good? In any case where a term has multiple uses, we are going to confuse or inconvenience someone, no matter what we do. We use statistics to choose titles so that we minimize the number of people that are confused or inconvenienced. That's the main idea behind primary topic. I don't understand why you think some minority should be favored over any majority in this calculus. Because someone believes the majority in some case is "US teenagers" (the relative importance of which you seem to discount) and the minority is people "looking for more established" uses (the relative importance of which you seem to inflate, puzzlingly)? That smacks of elitism, no?
  The point of good rules is to eliminate or at least reduce areas in which there is something to argue about. Tightening up consistent use of definitions of terms on a policy or guideline page is certainly part of that.
  Finally, you've not addressed at all the central question here. You did not acknowledge the definition of "ambiguous" specified in the first line of WP:D, nor did you talk about why or why not you favor a consistent application of that definition throughout the page, changing the definition at the top of the page, or what. --Born2cycle (talk) 18:52, 31 January 2013 (UTC)

The fact that you cannot see a problem is why I am afraid this debate will not get us anywhere. We do have to be prepared to stand back and think outside our own culture and interests in order to understand whether Wikipedia is as comprehensible, usable and accessible to everyone. That means wherever in the world they are coming from, not just the confines of Wikipedia, or the US. There are different ambiguities, and different ways of addressing them. --AJHingston (talk) 01:53, 1 February 2013 (UTC)

I agree with thinking outside of our culture to make sure WP meets everyone's requirements. That said, my questions stand. Unanswered. --Born2cycle (talk) 02:16, 1 February 2013 (UTC)

Support, as this is not actually any change in disambiguation guidelines. The additions are all repetitions of the statement in the lead, so an RFC is not particularly needed, IMO. Articles exists at titles, and sometimes redirects exist from alternate titles or to articles where the topic is mentioned but not the subject of the article. Disambiguation pages exist at all to allow a single title to become navigable to multiple topics: the list of topics whose Wikipedia articles could have had the title and also the list of topics that are only mentioned on other articles and could therefore have been redirects to those articles. Other wishful interpretations of what Disambiguation Pages could have been (sources of information themselves, instead of navigational tools to address the limitation keeping multiple things from having the exact same title) are served not by disambiguation page but by set index articles or other list articles. -- JHunterJ (talk) 19:06, 31 January 2013 (UTC)
- For the record, the reason I started this discussion and RFC is because my attempt to insert this clarification (not change) was reverted. Since then it has been reverted again, but the discussion expounding on the reasoning can't hurt, if nothing else to demonstrate that any disagreement with this is really with what WP:D has long (if not always) said, not with this change per se. If people want to broaden the meaning of "ambiguous" with respect to title decisions, that's a huge change in practice and consensus (not to mention documented guideline), that needs to be separately proposed, evaluated and discussed (frankly, I believe a broader definition of "ambiguous" would not improve anything for anyone, and would just create more grist for disagreement). --Born2cycle (talk) 19:16, 31 January 2013 (UTC)

Support as a reasonable clarification of existing guidelines and practice. --BDD (talk) 20:56, 31 January 2013 (UTC)

Oppose as in not-a-chance-in-hell. There's no consensus that the meaning of "ambiguous" is the one mentioned in the proposal; there's no consensus even that it's a definition - the wording of the introduction sentence (—dash—when it refers to more than one topic covered by Wikipedia articles) is so ambiguous in itself that it could equally mean "Disambiguation in Wikipedia is the process of resolving the conflicts that arise when [...] a term refers to more than one topic" - this doesn't imply that it's a definition of the word "ambiguous". PRIMARYTOPIC has zillions of problems as a guideline, but those won't be solved with an attempt to redefine a word but with a honest discussion of the concerns that its bad wording and unclear purpose cause.
Born2cycle knows that the meaning of this proposed clarification has been disputed many times in the past month. There have been extended discussions of the implications such meaning would have (and several arguments as to why using it is a bad idea), so it wouldn't hurt that he (she?) would explain the claims that "it's supported by consensus, really, and it's such a minor change after all", and why he thinks such redefinition of an English word can be made by policy without an actual warning note in the text clarifying "when we use the word 'ambiguous' we're imbuing it with a technical meaning that doesn't correspond with the one you'd find in any dictionary". Diego (talk) 21:12, 31 January 2013 (UTC)
- The complete wording of the first line is:
  Disambiguation in Wikipedia is the process of resolving the conflicts that arise when a single term is ambiguous—when it refers to more than one topic covered by Wikipedia articles.
  It's certainly possible that the "when it ..." clause does not define "ambiguous", but defines the scope of its application. But applying different definitions for that makes no sense either, and the changes discussed here ensure consistent usage and application, regardless of what exactly this phrase is defining. --Born2cycle (talk) 05:05, 1 February 2013 (UTC)

Oppose. I have serious concerns about this RFC, both procedural and substantive:

Procedural concerns:

A change on 19 January is wrongly taken as a stable basis from which to proceed. There was a discussion involving very few editors, and then the underlined text was removed (diff, repeating earlier non-consensual removal):
"There is no single criterion for defining a primary topic. However, [t]here are two major aspects that are commonly discussed in connection with primary topics: ..."
The discussion was brief and not advertised, but the change has serious consequences. Personally, I did not have time to take things further. That change is one step toward narrowing the primary-topic criterion; the present RFC builds on it, making it seem that there are only two considerations that are ever raised or relevant. Already the provision is often misread and misapplied as if that were so; this change would entrench that, and allow editors to cite WP:DAB against bringing evidence from reliable sources outside Wikipedia.
Update: As I write, an editor has restored that deletion as non-consensual. I appeal to participants here to respect that restoration to the long-stable wording. Let any alteration to that part be discussed in this RFC also. I for one argue against its removal.
The RFC has been set up with a poor structure: one that favours participation by "the usual suspects", and discourages wide community participation. As I write, already the proposer has interpolated large paragraphs that demand equally long responses. Newcomers with something to contribute are discouraged. Even knowing where to vote is difficult! Watch: it will get harder and harder. We have seen this again and again at WT:TITLE and elsewhere: Disorder and sprawl can favour control by a zealous clique.
The wording of the RFC proposal is not neutral (as required: see WP:RFC) but prejudicial. In particular with this text:
"So, I'm seeking support for this minor clarification (really redundancy) to make sure 'ambiguity' is interpreted throughout this page consistent with how it is defined in the first line of the page."
It is not a minor clarification. If there are differences in how "ambiguity" appears to be defined on the page, it is neither "minor" nor a mere "clarification" to have one single mention dominate, forcing changes in another six mentions. The wording at those six mentions (ambiguity interpreted broadly, not restricted to Wikipedia articles) has been hotly disputed in the past, here and at WP:TITLE. It is disingenuous to pretend otherwise.
None of the wording in the section concerned with primary topics mentions ambiguity; but the proposal purports that it somehow does. Since WP:PRIMARYTOPIC is very frequently appealed to independently of the rest of the text of WP:DAB (indeed, often incorrectly and misread), let it be considered in that independent role. Let us not pretend that WP:DAB is a coherent, well-understood, and integrated whole. "Disambiguation", disambiguation page", and "disambiguated" do occur in the text the proposer quotes; but these are highly technical and specific terms as used on Wikipedia. By way of contrast, the wording and intent of the criteria for deciding if there is a primary topic and which topic might be primary are autonomous, and reach beyond Wikipedian technical usage. The provisions on the page are descriptive (again: "There is no single criterion for defining a primary topic. However, [t]here are two major aspects that are commonly discussed in connection with primary topics: ..."). This autonomy is well illustrated by the second of the two descriptively mentioned criteria: "A topic is primary for a term, with respect to long-term significance, if it has substantially greater enduring notability and educational value than any other topic associated with that term." There is no mention of or appeal to ambiguity here: there is talk of topics "associated with that term", regardless of whether such association involves ambiguity. The proposal is misleading on these matters.
Equally, the proposal is misleading because it ignores this subsection, which counters the tight limiting tendency of proposed wording and which needs to be read along with the text cited in the proposal (my underlining and layout):

            === Determining a primary topic ===
            There are no absolute rules for determining whether a primary topic exists
            and what it is;
            decisions are made by discussion among editors,
            often as a result of a requested move.
            Tools that may help to support the determination of a primary topic in a discussion
            (but are not considered absolute determining factors) include:

            * Incoming wikilinks from Special:WhatLinksHere
            * Wikipedia article traffic statistics
            * Google web, news, scholar, or book searches ...

[Substantive concerns to be added soon.]

Noetica^Tea? 23:37, 31 January 2013 (UTC)

On Noetica's point #3... If there are differences in how "ambiguity" appears to be defined on the page.... No. There are no differences in how "ambiguity" is defined on the page. It's defined once, at the top, the first line of the page. The differences are in how some people interpret it later on the page, particularly at PRIMARYTOPIC. No one is denying that that wording has been debated in the past. Never-the-less, all this particular change does is ensure that "ambiguity" is interpreted on the page consistently with how it is defined on the top. --Born2cycle (talk) 04:58, 1 February 2013 (UTC)

Oppose I see no benefit in adding five new uses of the term "Wikipedia" in two successive sentences. This is not a "minor clarification". Omnedon (talk) 02:44, 1 February 2013 (UTC)

I demand clearly defined areas of doubt and uncertainty. Sorry, but I just can't help but giggle a bit at this, this is a discussion that would only happen at Wikipedia. I am not convinced that there is a serious problem here that would be resolved by repeating the word "Wikipedia" as many times as possible. However, there is an urban legend if you look in a mirror and say "Wikipedia" seventeen times, Jimbo will appear and "delete" you. Beeblebrox (talk) 18:01, 1 February 2013 (UTC)
A term is ambiguous if it is ambiguous to any reasonable group of readers, for any reason, regardless of current Wikipedia articles. Terms that are well known as dictionary definitions, for example, should be considered ambiguous as search terms. Big, for example. --SmokeyJoe (talk) 21:44, 1 February 2013 (UTC)
You are proposing a change to Wikipedia disambiguation non-article pages (which disambiguate Wikipedia articles for navigational purposes), to turn them in to another type of list article (things that provides encyclopedic information on their own to the reader). Yes, your definition of "ambiguous" is accurate, but, like the definition of "consensus", Wikipedia's guidelines use it differently than the English definition. -- JHunterJ (talk) 13:34, 2 February 2013 (UTC)
I don't propose any change to DAB pages, and don't see why you think I do. Could you explain yourself? --SmokeyJoe (talk) 06:15, 3 February 2013 (UTC)
Wikipedia:Disambiguation sets its context: resolving the ambiguity of topics covered on Wikipedia. You'd like to change that context, at least for the primary topic section, to also include topics not covered on Wikipedia. The project page also says that disambiguation pages are navigational pages, not articles. If we include information in the encyclopedia not covered elsewhere in the encyclopedia, then these are articles (like set index articles), and might need references and be subject to notability, etc. -- JHunterJ (talk) 12:44, 3 February 2013 (UTC)
This discussion is happening because there's no consensus for that interpretation of Disambiguation's context, and never has been, not even when you take into account the opening paragraph - which defines when to create disambiguation pages, but doesn't have anything explicitly limiting its scope. It doesn't even describe current practice; DAB pages include content about topics currently not in Wikipedia, such as links to Wiktionary and red links to non-existant articles (accompanied with a blue link to a related article that may or may not describe the ambiguous term). Restricting DAB pages to include only things currently in Wikipedia would be a change on itself, one having a wide impact throughout the project. Diego (talk) 14:00, 3 February 2013 (UTC)
The existing consensus is for that context (it's not an interpretation -- it's what the page says: "when a single term is ambiguous—when it refers to more than one topic covered by Wikipedia articles"). I realize there are a lot of editors here who are not familiar with the project and its history that lead to that consensus and who disagree with the primary topic and WP:PRECISION guidelines, and consensus may have changed. It's unfortunate that this discussion has been framed as it has been. The cross link to Wiktionary is separated from the list of ambiguous topics on Wikipedia, and the red links include blue links in the description to topics mentioned on other articles in Wikipedia (MOS:DABMENTION). Wikipedia disambiguation pages disambiguate Wikipedia topics that could have the ambiguous title; this is not a change to the consensus or to the page, even though there are editors who disagree with it or are unfamiliar with it. Continuing with that consensus would not have a wide impact throughout the project. Changing that consensus (and also WP:NOTDICT to include dictionary meanings directly or WP:NOTDIR to provide a directory of information everywhere on the Internet) would have a wide impact throughout the project. -- JHunterJ (talk) 17:53, 3 February 2013 (UTC)
Resolving the topics that are ambiguous frequently involves dealing with ambiguous meanings that WP editors don't want to write about but readers can reasonably expect to read about - such as dictionary definitions or less popular yet notable topics; or topics that do not share the exact term but are still easy to confound. Solving those situations is common practice at DAB pages, and yet they are not covered by the imperfect, limited and ambiguous "definition" at the first paragraph. No one is suggesting "trying to disambiguate everything" as you wrote, nor deprecating NOTDIR; that's a strawman argument. Links to Wiktionary are part of DAB pages, and NOTDICT explicitly encourages them. The strict limit of this proposal that you defend would involve removing all links to Wiktionary either at DAB pages or soft redirects, since neither of those are Wikipedia topics; that's certainly not the current consensus. Diego (talk) 22:24, 3 February 2013 (UTC)

P.S. And properly speaking, knowing the detailed story of the project is strictly irrelevant. What matters is the existing consensus among editors that are active at the current time. If some editors want to rely on some particular implications of the text as it's written, they should be concentrating on explaining the benefits of that approach that were learned from practice at the time, not on using the guidelines as if they were a sort of Revealed Text to follow by faith. It's up to the actual participants to evaluate if those purported benefits are still applicable to the current project; it may very well happen that the lessons from the past are not relevant anymore. Diego (talk) 22:41, 3 February 2013 (UTC)
Where you see Revealed Text, I see the consensus formed and refined. It may very well happen that the previous consensus loses relevance, but it won't be because some editors can't form new consensus in the pursuit of their Revealed Truth that disagrees with the previous consensus here and at WP:PRECISION. The benefits of using Wikipedia disambiguation pages to disambiguate topics covered on Wikipedia articles is reader navigation. The benefit of using the single topic on Wikipedia that is intended by most readers, moreso than all other Wikipedia topics for the title combined and much more than any single topic on Wikipedia, is efficient reader navigation. -- JHunterJ (talk) 13:07, 4 February 2013 (UTC)
Well, precisely the origin of this RFC comes from cases were the frequent practice of using exclusively page view statistics was not providing the most benefit to readers. SmokeyJoe provides below a good analysis of the points where current guidelines don't define a clear consensus; the linked move discussions above document some of the problems at hand. The process to determine which single topic should not depend only on page view statistics; the consensus at PRIMARYTOPIC that it should be formed using input from a variety of tools describing modern usage, as well as significance at reliable sources describing long-term weight.

If as you say the intent of those who wrote PRIMARYTOPIC was to select exclusively among the relative weight of articles already present at WP, without any consideration to usage of the term in the real world, it was certainly not what they encoded in the prose of the guideline. Moreover, that intent would present a a contradiction - that of failing to realize the stated goal of best serving readers since, in the cases discussed, the topic selected by editors as primary would not be the one more sought nor what readers would want to read when looking for the term. Solving that contradiction implies crossing the gap between what Wikipedia editors want to write about and what readers want to read about - and that can only be done by providing links to terms that are not currently covered as Wikipedia articles, but that readers looking up an encyclopedia want to read about. Diego (talk) 15:01, 4 February 2013 (UTC)
There's an intermediate position: it should indeed not be limited narrowly to current WP articles, but neither should it be extended into anything in the whole world. We are after all talking about all this in the content of the encyclopedia. What I think we should consider is the possible confusion between present and reasonabley likely WP articles. We should plan a little way into the future. DGG ( talk ) 05:33, 2 February 2013 (UTC)
WP:CRYSTALBALL. There's no harm in letting future Wikipedia's disambiguation pages handle future Wikipedia's ambiguities, while there's detriment in the pointless arguments that will come up over how far into the future and which not-currently-ambiguous topics will become ambiguous within that horizon. -- JHunterJ (talk) 13:34, 2 February 2013 (UTC)
We need to step back though, and think about the problem that is being addressed. Another way of formulating it might be that if users are reasonably likely to be confused when trying find something, we should try to minimise that confusion. That has many different aspects, only some of which are relevant to the disambiguation policies. It may be, for example, that an article editor has failed to understand that they are writing for an international audience and has not made clear that the article is about a particular local application; that might be put right by altering the lead or expanding the article, but amending the title might be appropriate. An editor might know when writing an article that there are several notable people, or buildings, or whatever with the same name which do not yet have articles; we need to be clear what they should do to help users avoid wasting time reading about the wrong one (and I am not convinced that there is a consensus that it is always right to create a stub article if we know little or nothing about a subject). In some cases the information that is being looked for is going to be in Wiktionary, or Wikivoyage, or Wiki Commons, or somewhere else and WP needs to evolve ways of suggesting that users try elsewhere. And so on. Of course we cannot solve every user's needs or anticipate every possible confusion, but we know that users come from all over the world, that Wikipedia is growing, that even simple words like city, gallon, ton, have different meanings in different parts of the English speaking world even before we run into the problems of translation or more sophisticated concepts. Disambiguation has to fit into all that in ways that make sense to editors and users. --AJHingston (talk) 15:03, 2 February 2013 (UTC)
That's a sensible and welcome approach - unfortunately, the confrontational approach of the current is not adequate for pursuing that discussion. I think a stance-based discussion like the one at Wikipedia:Requests_for_comment/Article_feedback, where each editor states their views and main concerns on the matter, would be more welcoming to constructive discussion and make easier to assess the real problems are and where the real consensus lies. Diego (talk) 16:53, 3 February 2013 (UTC)
Oppose – this is just B2C's next step in trying to remove "precision" as a consideration in naming, by redefining ambiguity to mean only article name collision, through the backdoor of the wording of disambiguation guidelines. He has been pushing this way for years, and it's still not OK. I'd go the other way, restore the wording of precision to what it was a few years ago, when it said things like "do not write or put an article on a page with an ambiguously named title as though that title had no other meanings" and "Name an article as precisely as is necessary to indicate accurately its topical scope" (see Wikipedia talk:Article titles/Archive 36#Some history of the "Precision" provision). Dicklyon (talk) 21:39, 2 February 2013 (UTC)

Oppose the yellow stuff above doesn't appear to clarify anything. --Joy [shallot] (talk) 11:56, 3 February 2013 (UTC)

Oppose. This seems to be an attempt to force all articles to conform to a single standard, which may be easy to interpret, but which clearly won't serve the needs of the community. The disambiguation policies at Wikipedia favour clarity over consistency, and discourage narrow, ethnocentric views of topics. This proposal would have the opposite effect, and lead to more conflicts between editors. Being "right" is less important than being helpful or informative. P Aculeius (talk) 14:34, 3 February 2013 (UTC)

Support JHunterJ's arguments make sense to me here; we must focus on the stated purpose of dab pages. The edit, although I cannot foresee why it makes such a huge difference here, does promote that purpose, and (+$0.02 here:) I think something should be said for attempting consistency. Reminds me of all the brouhaha over what the definition of "is" is. Carlossuarez46 (talk) 23:29, 4 February 2013 (UTC)
Carlos, the issue at hand is that the change is being proposed to block moves of obscure topics out of their respective PRIMARY base names, using the following argument:
1. Articles of pop culture named after common English idioms are located at the base name (see Big or Brand New).
2. Editors propose moves to place a disambiguation page (with a link to the Wictionary definition) at the base name for those terms, arguing that the "dictionary" meaning has wider use and long-term significance and thus the pop-culture item is not primary.
3. Counter-arguments are made citing that Wikipedia is WP:NOTDICTIONARY (a guideline that applies to content, not links, but whatever) and thus the more common use doesn't matter to establish the primary topic.
4. Choosing a primary topic exclusively between existing articles shows that the pop-culture topic has more relative popularity with respect to the others (excluding the word meaning), so it should be used as the primary topic and placed at the base name.
Is this whole chain of reasoning what you want to support? If it is, please state it directly. If not, then the RfC wording has succeeded at its attempt to obfuscate what is really being proposed. Diego (talk) 07:04, 5 February 2013 (UTC)
Of course, that's not the whole chain of reasoning. You omitted the counter-argument recognizing that when there is Wikipedia coverage of the dictionary meaning then the dictionary meaning is primary (Apple), but when there's no Wikipedia coverage of the dictionary meaning ("big", "brand new"), then obviously it's not Wikipedia's primary topic on Wikipedia for the title that is ambiguous on Wikipedia, because Wikipedia disambiguation pages disambiguate Wikipedia topics that could have the ambiguous topic. The easy solution, rather than trying to Wikilawyer the guidelines (taking some of the points out of their context, while claiming that those who simply read the whole of the guidelines are treating it as Revealed Text) would be to add Wikipedia coverage for these dictionary topics rather than trying to turn navigation pages into informational articles. -- JHunterJ (talk) 12:09, 5 February 2013 (UTC)
In that case you'd be missing the counter-counter-argument that the primary topic for a term should not depend on what is the current status of Wikipedia but determined with respect to what readers expect to find for that term and what reliable sources cover it, as the wordings topic sought when a reader searches for that term and [topic that] has substantially greater enduring notability and educational value than any other topic associated with that term suggest. The main point of contention is whether readers are best served by presenting them with information about what they look for, or what we as editors have decided to cover. The current introduction sentence (that "disambiguation solves problems when a term refers to more than one topic covered by Wikipedia articles") certainly doesn't decide the concern of PRIMARYTOPIC one way or the other, since PRIMARYTOPIC definitely addresses information that exists outside of Wikipedia articles. Funny how you accuse others of Wikilawyering, when the whole exercise of this proposal is to force a sentence written in one context (an introduction describing when disambiguation is needed) to a different one (a specific guideline determining what article merits having the simple term in exclusive).

I'm sure you'll have some counter-counter-counter-argument of your own. But since none of those arguments are part of the current guideline text, my point stands - that any particular position that is not literal interpretation of the current wording is not to be defended as "the current existing consensus". That includes the position that the second sub-clause of the intro sentence that describes DAB links applies to every possible consideration of the rest of the guideline that describe other types of content. Diego (talk) 14:01, 5 February 2013 (UTC)
So, you're taking it as Revealed Text and filtering it through the lens of your foregone conclusion? Rather than simply reading it as it's written, context and all, as an attempt to describe what is indeed the current consensus. -- JHunterJ (talk) 14:31, 5 February 2013 (UTC)
Final stance to make crystal clear what I mean: What you claim to be the current consensus is not the current consensus, because what you say is not written in the guideline. So we have as much right as you to assess the meaning of what it's actually written. (I was going to make some remarks about a High Priest interpreting the Revealed Text, but I'll better restraint myself). Diego (talk) 14:51, 5 February 2013 (UTC)
It still crystal clear: you are treating it as Revealed Text, except where it doesn't fit your conclusion. If you would like to restrain yourself, that would involve not posting those parenthetical digs, rather than implying that experience & familiarity with the guidelines & their history equates to high priesthood. -- JHunterJ (talk) 17:03, 5 February 2013 (UTC)
Diego, I'm still not persuaded. I find it quite possible that a primary meaning may change (for, example if the word itself is notable and an article is created for it) that bumps a conventional meaning of the word (Lunatic is an example - where the Wiktionary meaning is the same as Wikipedia's article - not the meaning (commonly ascribed) to "Lunatic"). These will be rare, I bet. Carlossuarez46 (talk) 17:23, 5 February 2013 (UTC)

Support. Disambiguation pages are a navigation tool; they are necessary because our search function, article title implementation, and the English language all intersect awkwardly. In an ideal world, we would be able to tell exactly which encyclopedia page a reader is looking for without resorting to disambiguation pages. But even in that ideal world, we cannot possibly be expected to help readers find non-encyclopedic topics that may share a name with encyclopedia topics. If a reader searches in an encyclopedia for the word "big", it is perfectly reasonable to assume that they are not looking for a definition of the word "big" (for which they should search a dictionary) but rather for an encyclopedic topic with the name "big". Since, you know, they're searching an encyclopedia. Powers ^T 18:48, 15 February 2013 (UTC)

There is a problem, though, in that we cannot define 'encyclopedic' as simply being what is currently in Wikipedia. There are still gaps, and the notability criteria are often arcance and inconsistent between topics. The point of reference should surely be the user, not Wikipedia. --AJHingston (talk) 21:55, 15 February 2013 (UTC)

There are gaps, but we're intelligent enough to distinguish between topics that are likely to have an article someday and those that never will. Powers ^T 00:29, 16 February 2013 (UTC)

rfc comment: Oppose Other things exist, even if we don't have an article on them yet. Killer Chihuahua 17:45, 20 February 2013 (UTC)

Ambiguous application of "ambiguous"

I find this RFC confused by ambiguous application of the word ambiguous.

The way I see it, there are three spearate applications.

(1) Primary topic. A title is a primary topic if it is not ambiguous to any worthwhile group of readers that the title refers to the topic. Eg. Earth. It may be ambiguous with soil among people without any modern astronomical awareness, but we dismiss them as not our audience.

(2) Disambiguated titles. Eg ? (film). There is only one article, but PRIMARYTOPIC fails because the title is ambiguous. Titles should be sufficiently precise that you can tell from the title (the url) what the topic is likely to be about.

(3) Disambiguation pages. These are needed where there are three or more topics, the existence of a PRIMARYTOPIC being irrelevant to the question. There are multiple ambiguities among Wikipedia articles, including immediately forseeable articles allowed as DAB page redlinks.

--SmokeyJoe (talk) 00:07, 4 February 2013 (UTC)

I'm not sure I disagree what you're trying to say, but the perspective in Point 1 seems a little skewed, if you're serious (I can't honestly tell). Who determines how "worthwhile" a group of readers is? Do we have groups of readers (as distinguished from vandals) who aren't worthwhile? And your point about earth/soil is improperly argued: the issue isn't whether "earth" is a valid contemporary synonym for "soil" (it is, though mostly in literary language), but whether in that lexical guise it's an encyclopedic topic (it isn't; it's a dictionary entry). Cynwolfe (talk) 16:39, 4 February 2013 (UTC)

Is that last point right though? Soil is a legitimate encyclopedia topic - the proper question ought to be whether users are reasonably likely to search for it under earth. Surely it is in the approriate DAB page for that reason. There will be topics where users will not be familiar with the alternative word, or with use of their search term for the topic. Words with more than one meaning, and topics with more than one name, pose a real problem because we cannot assume a shared cultural experience. That is one reason why the 'pop culture' debate arises, it is not just about elitism. --AJHingston (talk) 19:19, 4 February 2013 (UTC)

No, I agree with you: as a gardener, I see "earth" in the sense of "soil" as a useful orientation item in a dab list. I was mainly commenting on the mode of argumentation, which just asserted "'earth' doesn't 'really' mean that." I see that a lot. Obviously when this kind of discussion breaks out, it does "really" mean that to a significant number of people, or there'd be no argument. We're not here to impress specialists with how hypercorrect we are; we're here to serve the kind of general reader who actually uses Wikipedia. It's just that lowercase "earth" to mean "soil" probably wouldn't be the article title, and certainly not the primary topic—but yes, I certainly do think dabs should contain items that are not the article title but are possible search terms that might need disambiguating. Cynwolfe (talk) 01:07, 5 February 2013 (UTC)

Hi Cynwoldfe,
My main thrust is that I wish that everyone would be clearer, in the above RfC, as to whether they are talking about issue (1), (2) or (3). There are strongly connected, but not equivalent.
On (1), I think than "ambiguous" means "ambiguous, without restriction to just Wikipedia articles, to anyone". Then, digressing a little from seriousness, thinking that "anyone" might be too loose, I suggest a hypothetical group that we might not bend over backwards to meet. What I was thinking, is that someone searching for earth, as in soil, coming to a page on the planet, should be expected to not be astonished because nearly everyone knows that earth is planet. (capitalisation issues aside, few search engines use capitalisation). I think that we should make sure that very few readers are astonsihed by the page they arrive at on following a link or a search result.
More generally, I feel and sense from others that (1)PRIMARYTOPIC has issues requiring attention; (2) stems directly from (1) but is not itself a point of contention (we disambiguate the title if the title is not the PRIMARYTOPIC (not meaning to imply that a PRIMARYTOPIC necessarily exists)). Finally, I don't think (3) is at all contentious, but I read some criticism of (1) being defended in terms of (3). --SmokeyJoe (talk) 01:21, 5 February 2013 (UTC)

I think that the capitalisation point is actually a nice example of the difficulty. If I were looking for an article on Soil I would very likely capitalise it instinctively because that is how I think of it as a course or subject title. I would not expect to have to guess whether the article title is capitalised, or be automatically be taken to an article about a particular book or TV programme of which I might never have heard. The fact that it is not ambiguous within Wikipedia because initial capitalisation distinguishes between the articles is of no help to me as user. It is a bit like asking users to guess whether words have been hyphenated, where US and British usage differ. The within Wikipedia point runs across your three categories. In the case of category 2, we can have an instance where there are, for example, two historic buildings with the same name in different parts of the world. If only one has an article, should the title be disambiguated? If the answer is that we are only concerned with ambiguity in what is currently on WP, then logically not; if we are concerned with ambiguity for the user then I think we should. Or does the community want to say that there should be a disambiguation, but editors should always create a stub article for the ambiguous topic (which creates its own problems)? If we only want to deal with the question of the primary topic that has to be explicit, though we still have to be address the difference between internal and real world ambiguity. --AJHingston (talk) 15:53, 5 February 2013 (UTC)

SmokeyJoe, first, I don't understand your objection. The RFC above is proposing wording changes in the section about only one of the three applications you list: (1) Primary topic.

Second, I disagree with your description of Primary topic. You write: A title is a primary topic if it is not ambiguous to any worthwhile group of readers that the title refers to the topic. Many (probably most) titles of primary topic articles are ambiguous; that is, those titles have other uses on Wikipedia. The only primary topics that don't have ambiguous titles are the ones with unique titles, and those are not usually referred to as "primary topics", since they are the only use of the term in question, and primary is a relative term that presumes others exists. While Earth is ambiguous, and not only "among people without any modern astronomical awareness" (the planet and soil meanings are both legitimate and distinct), we believe most people searching with "earth" are looking for the planet article, so that's what we put at Earth (as opposed to a redirect to Soil or a dab page).

Third, for Disambiguated titles, you say, "Titles should be sufficiently precise that you can tell from the title (the url) what the topic is likely to be about." Not quite. More accurately: "Titles should be sufficiently precise that someone familiar with the article's topic (though not necessarily an expert) can recognize the title (the url)."

On Disambiguation pages, I would say they are needed when there are three or more topics that exist on WP. While it's true that "immediately forseeable articles" are also allowed as DAB page redlinks, they shouldn't count as one of the minimum three until they actually exist. --Born2cycle (talk) 19:15, 5 February 2013 (UTC)

Born2cycle, I think PRIMARYTOPIC needs a complete rewrite, and your suggestion is either insufficient or further confusing. I think Earth is fine as a PrimaryTopic because the ambiguous topics are directly related and no reasonable reader would be astonished to find themselves on the current PrimaryTopic page on following a link to "earth".

On Disambiguated titles, we seem to be close, but I think we should never assume an expert reader, we should assume much less, with respect to titles and ledes.

On Disambiguation pages, that's a very very minor difference, as there are fractionally very few worthwhile redlink titles to be found. --SmokeyJoe (talk) 21:36, 6 February 2013 (UTC)

Totally agree about never assuming an expert - but assuming familiarity with the topic is absolutely essential in terms of how "recognizable" we make our titles, otherwise we will need to change the vast majority of our titles!

Please explain what is insufficient or further confusing about my suggestion, especially the confusing part. --Born2cycle (talk) 21:44, 6 February 2013 (UTC)

Proposal for new variety of quasi-disambiguation page being discussed

The discussion at Wikipedia talk:Policies and guidelines#Accepting "WP:Multiple-cross-reference page" as a guideline might be of interest to disambiguators. older ≠ wiser 14:17, 13 February 2013 (UTC)

Primary topic does not mean "is the title"!

SmokeyJoe, I'm really confused by this edit with edit summary "A dictionary word, or swear word, or neologism, and other things may make a unique topic an inappropriate title".

First, I can't even parse "may make a unique topic an inappropriate title". Topics are the subjects of articles. They can't be titles! Topics are what the article is about. Titles are almost always the names or descriptions of those topics.

Of course I agree that "a dictionary word, or swear word, or neologism, and other things" may not make appropriate titles, but I don't understand how that explains removing this statement:

When a word, name or phrase refers to only one topic on Wikipedia, then that topic is the unique use, and so also the primary topic for that term.

Can you give a specific example of what you're talking about?

Here are some examples of "words, names or phrases" that refer to only one topic on WP, and so those topics are primary for these terms: 2004 Birthday Honours, Abidemi Sanusi, Wallace's Isle, Castillo de San Pedro de la Roca, First Market Bank, Cloud.bg, Motema, Eupithecia nuceistrigata, Bring 'Em Back Alive, The Origin of Fire, Sausar, Group signature, Adaptive response...

All of those listed above are articles titles. Here are some terms which are not titles, but are redirects to their respective primary topics: Oprah, City of New York, Perdicula asiatica, Love making, Pedalcycle, ...

Can you provide any examples of "words, names or phrases" that refer to only one topic on WP, but for which those topics are not primary for those terms? Even one example that is contrary to the statement you removed? --Born2cycle (talk) 22:08, 6 February 2013 (UTC)

I haven't found one yet. I guess that I am uncomfortable with the strength of the statement and the possibility of it implying a MUST for a particular title when there may be a good reason for a different title. It appears to mean that titles should be as brief as possible, while on highly specific subjects a short title can imply a broad article. A rewording might be good. --SmokeyJoe (talk) 04:17, 8 February 2013 (UTC)

The other way around, can you prove your claim to be correct? Can you prove that every topic "X (A)" residing at "X" and there being no "X (B)" in Wikipedia, is the general primary topic for X? Pedro Gonzalez-Irusta (talk) 22:45, 6 February 2013 (UTC)

Of course. If there is no other topic on WP to which "X" refers, then not only is the topic of that article primary for "X", but the use is unique, by definition.

Any way, hitting SPECIAL:RANDOM any significant number of times will produce examples of which the vast majority are consistent with the statement quoted above. When zero examples can be produced that contradict it, I suggest that proves the claim to be correct about as well as any claim on WP can be proved to be correct. --Born2cycle (talk) 22:56, 6 February 2013 (UTC)

"Of course." - If you can, then please provide the prove. Pedro Gonzalez-Irusta (talk) 22:58, 6 February 2013 (UTC)

I'm still not sure exactly what you're asking. Regardless, I'm not going to waste my time trying to prove anything about "general primary topics" (whatever that is) which have nothing to do with determining primary topics on WP. --Born2cycle (talk) 23:05, 6 February 2013 (UTC)

Your SPECIAL:RANDOM method only looks at the set of topics covered in Wikipedia. But the criteria for a primary topic don't restrict themselves to only that set. Pedro Gonzalez-Irusta (talk) 23:13, 6 February 2013 (UTC)

Why do you believe that "the criteria for a primary topic don't restrict themselves to only [the set of topics covered in Wikipedia]"?

Here is why I don't believe that's the case. First, the whole point of PRIMARYTOPIC is to make an exception of disambiguating a title that would normally be disambiguated due to it being ambiguous in the WP sense set out on the first line of WP:D... ""when it refers to more than one topic covered by Wikipedia articles." If the title does not refer to more than one topic on WP, then PRIMARYTOPIC and everything else on WP:D, including guidance on how to disambiguate a title, doesn't even apply to the situation.

Second, if we ignore the context of WP disambiguation and inexplicably treat PRIMARYTOPIC as some kind of criteria independent of the problem of disambiguating WP titles, and choose to "disambiguate" titles against some vague concept of topics not covered in WP, we are opening an incredibly large can of worms. What exactly is this general set of topics? How notable to do they have to be to be included in this set? How are these questions decided... based on what criteria?

The criteria for topic inclusion based on notability is not perfect, but it's fairly well defined and refined. For the purposes of title determinations and disambiguation, traditionally, we conveniently assume it is perfect. That allows us to not have to re-open all these basic questions for every title consideration. We limit the scope of what we have to consider to the very well defined set of topics covered in articles on WP. What you're proposing reverses all that. To what end? --Born2cycle (talk) 00:32, 7 February 2013 (UTC)

There are many topics that meet the vague and not well defined criteria of Notability. They will be included if editors spent time on doing so. And then, they may have to move articles around. Village A1 is not yet in Wikipedia, but Village A2 is. Start moving and fix links. Footballer A1 was important in 2010 and he was the only by that name, but in 2014 there were two others by that name. Move Move. The can of worms is Primarytopic if you don't have the full set of topics that may be included in WP even under current Notability policies. Pedro Gonzalez-Irusta (talk) 01:26, 7 February 2013 (UTC)

Part of the responsibility of creating a new article on WP is choosing a title for it, which includes reviewing the titles of articles about uses of the same name, and, possibly, changing some of those accordingly as a result of such review. That's a perfect time to take into consideration the impact of the use in the new article with respect to the other uses, not doing so based on speculations about what may or may not be an article in the future. --B2C 17:05, 22 February 2013 (UTC)

Avoid this one?

There is discussion at 2013 Southern California shootings about a rename as well as a merge discussion with Christopher Dorner. I was wondering if any of the wise regulars here can help them decide on a good title for the former and a good title if they are combined. Feel free to avoid, run away and hide, revdel, and/or oversight this post.--Canoe1967 (talk) 14:58, 23 February 2013 (UTC)

Katz

Resolved

This dab page has two dab templates at the bottom that look identical. I am not sure which to remove to avoid the duplication. No biggy.--Canoe1967 (talk) 16:32, 19 February 2013 (UTC)

I've cleaned it up. -- JHunterJ (talk) 19:45, 19 February 2013 (UTC)

Great work as usual. I noticed that Jonathan Katz is in the Katz (surname) list, but should his character Dr. Katz, Professional Therapist be included in fictional section in the Katz dab as well? Readers may not know both of his names when searching.--Canoe1967 (talk) 20:21, 19 February 2013 (UTC)

My suspicion is that readers not knowing both names but seeking this topic would enter Dr. Katz and land in the right arena. But it might be returned to the dab in the See also section, as a partial title match. -- JHunterJ (talk) 20:26, 19 February 2013 (UTC)

I never thought to search for it. If they can't find it with what they are searching for there is always help desk. Hopefully a request there won't cause someone to mess up your work. Thanks again, resolved.--Canoe1967 (talk) 20:51, 19 February 2013 (UTC)

Am I clairvoyant or what? Didn't take long for someone to mess up your beautiful work.--Canoe1967 (talk) 15:48, 24 February 2013 (UTC)

paste and pasta

Not sure what the guidelines are for this. Should these two pages, Paste and Pasta (disambiguation) be merged in to one?

This is causing a bit of an issue on wikidata and the handling of interwikis. Danrok (talk) 20:18, 22 February 2013 (UTC)

They are two completely different terms in English. I don't how merging would help readers. older ≠ wiser 20:26, 22 February 2013 (UTC)

No, they should not be merged with each other or with Pasty (disambiguation) or Piste (disambiguation). Only if two titles have almost identical lists of topics ambiguous with the titles would be merge them. -- JHunterJ (talk) 21:15, 22 February 2013 (UTC)

I never knew that when you cut an object on a computer you pasta it on something else. Something doesn't feel right here. Simply south...... catching SNOWballs for just 6 years 21:23, 22 February 2013 (UTC)

No they should not be merged, they are not the same. Why is paste (glue) missing? Surely we should have an article on that. Hopefully Wikidata will fix the interwiki link problems over time. Vegaswikian (talk) 22:44, 22 February 2013 (UTC)

I created Paste (glue) and Paste (adhesive) that both redirect to Adhesive and added an entry for Paste (adhesive) on the Paste dab page. --B2C 23:18, 22 February 2013 (UTC)

I don't see any case for merging Paste and Pasta (disambiguation). Aside from a shared etymology they have little in common. It looks to me like the article Wheatpaste should be the target of Paste (glue) and Paste (adhesive). I don't think I've ever heard of "wheatpaste" before. I think a case could be name for renaming Wheatpaste to Paste (adhesive), but that's another issue. SchreiberBike (talk) 03:04, 23 February 2013 (UTC)

Agreed. --B2C 17:31, 24 February 2013 (UTC)

I think the adhesive should be the primary topic of the title, Paste. I never hear the paste part of the computer function outside of "cut and paste". I certainly wouldn't merge paste and pasta. bd2412 T 18:02, 24 February 2013 (UTC)

Example of why "historical significance" consideration in PRIMARYTOPIC is a problem

The discussion at Talk:Madonna_(art)#Requested_Move:_.E2.86.92_Madonna demonstrates how the "historical significance" consideration of WP:PRIMARYTOPIC serves no purpose except to muddy the waters. Regardless of the historical significance of the art piece or any other use of that term, no use of "Madonna" should be at Madonna, unless it can be demonstrated that most people searching with "Madonna" are seeking that article's topic. Period.

I propose we remove all references to "historical significance" from PRIMARYTOPIC.

The actual significance of any use, historical or otherwise, is reflected in google results, link counts and page view counts - information that is objective. The traditional primary topic criterion -- much more likely than any other, and more likely than all the others combined -- implicitly encompasses the "historical significance" consideration. Having the "historical significance" consideration explicitly called out separately only confuses matters, as is demonstrated at that discussion. Since "historical significance" can be evaluated only purely subjectively, it makes PRIMARYTOPIC even more vulnerable to disagreement and WP:JDLI argumentation. The historical significance criterion is important, but since it's implicit in the usage criterion, I think the usage criterion is all we should mention explicitly.

Comments? --Born2cycle (talk) 17:20, 31 January 2013 (UTC)

NOTE As Kauffner points out below, there is no reference to "historical significance" in WP:PRIMARYTOPIC. The current text says: "In many cases, a topic that is primary with respect to usage is also primary with respect to long-term significance. In many other cases, only one sense of primacy is relevant. In a few cases, there is some conflict between a topic of primary usage and one of primary long-term significance. In such a case, consensus determines which article, if either, is the primary topic." Editors may consider how this boo-boo reflects on the general quality of this proposal. Johnbod (talk) 17:51, 1 February 2013 (UTC)
NOTE 2. I dropped a notice of this discussion at the Classical Greece and Rome project, because members of that project have participated in dab discussions as they pertain to classical antiquity. There are of course other projects with useful perspectives who could be notified. Cynwolfe (talk) 17:36, 2 February 2013 (UTC)

Oppose Fixes a non-problem. That people introduce, on the looks of it unsuccessfully, an argument in a discussion is not a reason to change a policy that works well. Johnbod (talk) 18:45, 31 January 2013 (UTC)
- If something in policy provides the only ostensibly reasonable grounds for dubious proposals, and provides no grounds for sound proposals that are not also provided by something else in policy, then I would say that something does not "work well". In other words, I'm saying there is no topic that is primary per historical significance, and should be the primary topic, that is also not primary per the usage criterion. On the other hand, there are topics that could be argued to be primary per "historical significance", but are not and should not be primary, like Madonna (art). These are the ones that are not primary by usage. Therefore, we should use only usage to determine whether a topic is primary. If I'm wrong about this, actual counter-examples should be easy enough to provide to refute what I'm saying. --Born2cycle (talk) 19:26, 31 January 2013 (UTC)
  - I think "historical significance" should be a factor to be weighed, but it is just one of many. In the Madonna case, it is outweighed by the multiplicity of meanings and the popularity of a different meaning. bd2412 T 03:56, 1 February 2013 (UTC)
    - What do you think of my point that historical significance is implicitly accounted for in looking at usage? I mean, the higher the historical significance of a topic, the more likely it is that people will be searching for it, relative to other uses of the term in question. Does that make sense?
      For example, there quite a few Thomas Jeffersons... (see Thomas Jefferson (disambiguation), but the most historically significant one gets the most usage, and so is at Thomas Jefferson. Why do we need to look at historical significance separate from usage? Doesn't that just create more reasons for people to disagree? What problem does that solve? --Born2cycle (talk) 04:13, 1 February 2013 (UTC)
Support. The guideline says, "long-term significance", not "historical significance." If the criteria applies only to Buddhist subjects and to Hindu subjects, and not to a major genre of European art, it's obviously not working to promote educational value. The way the guideline is currently written confuses many people into thinking that a primary topic should meet both the long-term significance and usage criteria, although I don't think that is what was intended. My interpretation is that only guideline-based reason to oppose making the art form primary for Madonna is if you think the singer should be primary. Kauffner (talk) 05:10, 1 February 2013 (UTC)
Comment. To speak to the example you raise, (without re-arguing that debate here), without the long-term significance criteria, Madonna (entertainer) would have a fairly simple case of being the primary article, based on it's very high usage relative to other Madonna (x) articles. I don't think that is the right answer in that case, nor in many similar cases. I think both criteria (usage and long-term significance) are valuable, and weighing both together usually leads to the best outcome. Perhaps the guideline could be clarified by revising the last sentence to something along the lines of, "While an article does not need to pass both tests to be selected as a primary article, where there is a significant conflict between the topic of primary usage and the topic of primary long-term significance, it may be best not to select either as the overall primary article."--Trystan (talk) 06:00, 1 February 2013 (UTC)
- I really don't see who benefits from not having the most viewed/referenced use, by far, of "Madonna" at Madonna. I suggest the most neutral POV to adopt in these decisions is that of a Martian tasked to serve human WP users, and whose job is evaluated according to how many are served well by her decisions, as compared to how many could be. Based on that objective criteria, insisting on making Madonna a dab page should lead to her firing. --Born2cycle (talk) 16:55, 1 February 2013 (UTC)
In my opinion, even the sleaziest pop culture topic is more educational than a DAB. If the "long-significance" criteria will be used as a reason to promote DAB pages to primary topic status, as seems likely, I say junk it. Kauffner (talk) 17:25, 1 February 2013 (UTC)

I take a completely different perspective -- where there is no clear primary topic, readers are better served with disambiguation page than making an arbitrary choice. As I've said before, I would support making a disambiguation page the default where there are multiple topics and any claim of primary topic should need some supporting evidence and if contested, consensus. older ≠ wiser 17:44, 1 February 2013 (UTC)

I agree with using DAB pages when there is no clearly primary topic, but the usage criterion handles this. When there is no topic that is much more likely than any other, and more likely than all the others combined, to be the one being sought, then there should be a dab page. With Madonna we have a clear primary topic case. And, yet, because of "long term significance", not because any other uses are nearly as likely to be sought, we have a dab page at Madonna. How are the users served by this? Fire the Martian (that's a reference to my comment just above - in case anyone missed it)! --Born2cycle (talk) 17:54, 1 February 2013 (UTC)

The purpose of a title is to tell readers the name of the subject, not to make any choice for the reader. So the idea should be to get as many articles as possible titled under the actual name. If there is a conflict, the article that gets more page views or has more educational value should get priority. What this does to traffic depends on Google's search algorithm, which we can't control and shouldn't try to game. I think it is clear from the page view stats that when readers arrive at a DAB unexpectedly, they don't use it. So there isn't any point to using the primary topic rules to get more readers to go to DABs. Kauffner (talk) 03:49, 2 February 2013 (UTC)

Support. For the record, my first preference would actually be seeing WP:PRIMARYTOPIC go away forever altogether—whatever its benefits are, they are more than offset by mounds and mounds of pointless wasteful bickering across many and many articles about which usage is more "primary" than another. However, since axing the whole thing is not the subject of this particular discussion, I'd gladly settle for at least removing this one clause which, in my opinion, significantly contributes to those mounds and mounds of pointless bickering I mentioned earlier (and because the fewer factors we need to weigh, the sooner we can get back to actual editing). If nothing else, the "historical significance" clause is just too damn vague; it is hardly ever that two people interpret it the same way! I'd also gladly go with the suggestion by (old and unwise :)) Bkonrad above, and while I don't think it would solve the problem, it would still be a significant improvement over existing practices.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); February 1, 2013; 18:08 (UTC)
Comment. I always thought the purpose of the "historical significance" caution was to ward off pop-culture blips that quantitatively distort search results in the short term. Madonna is an exceptional example, since even without the complicating factor of Ms. Ciccone, we'd fork Madonna (art) and Mary, mother of Jesus. At any rate, it's hardly true that "With Madonna we have a clear primary topic case". I'd say we have an excellent example of not having a clear primary topic, with or without the "historical significance" or "long-term significance" clause. So Madonna is a bad case for changing policy. Could someone provide another example where "historical significance" provoked unseemly arguments? Cynwolfe (talk) 18:16, 1 February 2013 (UTC)
- Of course "Madonna" means Mary, but almost exclusively in the context of the art form. The most notable RM where the long-term significance issue has come up is avatar. In my opinion, this article was promoted to primary topic status less because of its educational value than because avatar is a dictionary word. One I was involved with is Yui (singer). At the time of the RM, the singer was only topic people typing in "Yui" were likely to be looking for. Yet she was still judged to be of insufficient educational value to be primary. Adele (singer) is a similar situation, although the guideline cited was WP:RECENTISM. Kauffner (talk) 03:49, 2 February 2013 (UTC)

Yes, "recentism" is another way to express "historical significance". There are also instances when the "most common" meaning of a word is non-encyclopedic (die). If there's a serious debate over what the primary topic should be, creating a dab is the solution. I agree that the wording could be improved, but not by simply asserting that whatever search term is most popular at a given moment is the primary topic. As a side note, the intellectual distinction between "Madonna" as an art genre and "the Madonna" as a religious icon is not so definitive as we secularists might imagine. It can be difficult historically to draw a hard line between "pure" art and a work created specifically as an aid to religious veneration in a place of worship. This was one of the areas in which Protestantism quarreled with the "idolatry" of Catholicism. At any rate, if I understand better the intention of this proposal, I'll switch to express an opinion rather than merely commenting. Cynwolfe (talk)

Oppose. If the wording needs improved, it should be in the direction of "when in doubt, create a dab." Although I recognize that some Wikipedia editors will dance angels on the head of a pin forever just for sport, the existence of a debate among reasonable people is prima facie evidence that there is no clear primary topic. It's only "clear" depending on your perspective, and the article title in the tug o' war should be the dab. One indication of "primary topic" might also be that there is no good or effective way to disambiguate the term; that is, it wouldn't occur to my hypothetical reasonable person to specify the topic in that way. Most adults would recognize that Madonna doesn't always mean Ms. Ciccone, and wouldn't be surprised to be taken to a dab. As for educational purpose, I've often learned some interesting things when alerted that I linked inadvertently to a dab. It does no harm to learn things you didn't set out to learn. Cynwolfe (talk) 17:04, 2 February 2013 (UTC)
- Yes, dabs are not bad or even evil. They are good and informative. To add one point, if there is a question primacy for two similar articles, the more general or overall or most encompassing one should be at the base name since that one would always be right and knowledgeable editors can always change the bad links to a better one. So, for example, when you have a province and a city sharing a name that don't cover the same area and the city is totally within the province, then the province should be at the main name space if the primary topic is unclear. Vegaswikian (talk) 20:11, 2 February 2013 (UTC)
Oppose – disambig pages are good, useful, orienting, and educational. This change would support Kauffner's odd viewpoint "even the sleaziest pop culture topic is more educational than a DAB", which would make WP a lot sleazier indeed. Dicklyon (talk) 21:46, 2 February 2013 (UTC)
- There is a policy against making sleazy pop culture articles primary topics? If so, we need to do something about Kesha, Britney Spears, and Lindsay Lohan. Kauffner (talk) 11:24, 3 February 2013 (UTC)
Hey folks, I see the light now. What's up with articles, anyway? Whatever their benefits are, those are more than offset by the fact that they distract readers from our excellent collection of disambiguation pages. That's where Wiki's real educational value is. Did you know that Encyclopedia.com, IMDB, AllMusic, AllMovies, and Encarta don't even have DABs? Britannica has something equivalent, but they are nothing like ours. They look almost like search results, and not that many topics are covered. They don't have them for the stuff that really needs disambiguating, like "big". It's not like their readers can enjoy a long list of partial title matches or anything like that. It could be a selling point: Wikipedia -- Not just an encyclopedia. A free collection of disambiguation pages anyone can edit. Kauffner (talk) 01:44, 3 February 2013 (UTC)

Do the encyclopedias you mentioned cover as many subjects as Wikipedia? Britannica has no article for the city of McLouth, Kansas, but Wikipedia does. Do the databases you mentioned cover all subjects (not just movies or music), like Wikipedia does? Wikipedia is in fact different than any of those others. They all have their strengths, and I use several of them; but they are not all the same. Having said that -- I presume you are saying that we should not have disambiguation pages, right? Omnedon (talk) 02:13, 3 February 2013 (UTC)

The attempt at reductio ad absurdum does a disservice to what has been a serious consideration of Born2Cycle's proposal. Cynwolfe (talk) 03:35, 3 February 2013 (UTC)

Strong Oppose. The policy is there for a valid reason. Historical usage is important and a valid criterion for determining which article, if any, should have the unqualified title. In this case, the popularity of the entertainer is a strong argument for not giving primacy to the historical usage of the term, but at the same time, the historical usage is widespread and well-known enough not to be completely eclipsed or rendered secondary. That's why I think it's better to use disambiguation to help direct readers to the articles they want, as well as providing some context by doing so. I understand that for some editors, the popularity of Madonna Ciccone is enough to justify according her primacy, while for others, historical usage should establish primacy. The fact that reasonable people could disagree on this issue (and clearly do) is one of the strongest arguments for disambiguation in article titles, as well as using disambiguation pages when readers might be looking for information on different uses of a term. P Aculeius (talk) 04:32, 3 February 2013 (UTC)
@Omnedon: IMO, articles should get priority. So most DABs would end up in the form "Foo (disambiguation)". Then they would be out of the way. The ultimate solution is software that allows multiple instances of a title. Then we could move past this issue. Kauffner (talk) 07:27, 3 February 2013 (UTC)
Oppose. The long-term significance aspect is an important one for every serious encyclopedia. In the Madonna example, the primary topic is clearly Mary (mother of Jesus). That's where all the other topics are derived from: Madonna (art) and whoever is named after the original. If the original Madonna were not better known as Mary in English, we would have a conflict between the two principles. (That happens occasionally, must be dealt with flexibly, and is not really a problem. Kauffner's proposal was simply wrong because Madonna (art) is definitely not the primary topic. That's the second most important topic w.r.t. long-term significance, but that does not mean it automatically wins when the primary topic doesn't need the title.) But the original Madonna doesn't need the title, so there is no primary topic and Madonna is a disambiguation page. That's exactly as it should be.
We have similar situations with Ubuntu and Ebert. The proposal would lead to a situation where the operating system would get precedence over the philosophy after which it is named, and the last name of the first German President would become a redirect to an American film critic. That would be ridiculous recentism. Hans Adler 09:41, 3 February 2013 (UTC)
- The whole concept of letting proper names be a "primary topic" and residing at the ambiguous name is recentism. Pedro Gonzalez-Irusta (talk) 23:04, 6 February 2013 (UTC)
Oppose - link counts and google results are not actually particularly objective, it's still perfectly reasonable to tell people to assess historical significance as opposed to thinking popularity metrics are a magic bullet. --Joy [shallot] (talk) 12:00, 3 February 2013 (UTC)

Comment. Perhaps have a section in this article that discusses what happens if there IS a conflict between the rules...? Srsrox (talk) 20:42, 4 February 2013 (UTC)

Oppose - although first in time is not necessarily first in right; it must be considered: Troy, with a current population of ZERO, is certainly a primary topic as opposed to the various people, fictional characters, and other towns/cities/chess moves/etc. so named; this, even though "Troy" was probably never used by the locals (or the Greeks who wrote about it). Carlossuarez46 (talk) 23:35, 4 February 2013 (UTC)

Oppose mostly for the reasons ably stated by older ≠ wiser and Cynwolfe, but also because sometimes taking a step back and saying, what did I think about this topic five years ago and what will I think about this topic in five years changes the whole perspective. The U.S. Post Office (before the USPS) had a good idea when they said, "No living person shall be honored by portrayal on U.S. postage." Too bad the USPS doesn't believe that. Postal Service Will begin Honoring Living People on Stamps N.Y. Times 26 September 2011. --Bejnar (talk) 10:55, 8 February 2013 (UTC)
Weak support This provision is rarely argued and rarely wins arguments anyway. I've only really seen it used well once. --BDD (talk) 00:56, 14 February 2013 (UTC)
Oppose: My opinion mirrors what was expressed by Cynwolfe. If there is a debate then simply create a dab, problem over. J04n(talk page) 11:21, 22 February 2013 (UTC)
Oppose per Hans Adler, "The long-term significance aspect is an important one for every serious encyclopedia." Granted, one wonders if and when Wikipedia will become a serious encyclopedia, but some of us do aspire to that. First Light (talk) 00:39, 26 February 2013 (UTC)
- If we had a single editor or an editorial board of a few people that made these decisions, then we could aspire to having titles that reflected a priority for long-term significance over usage sometimes, as they could develop standards that they consistently applied in deciding how to weigh usage against long-term significance. But since a different set of people makes each title decision, we don't have that luxury. In order to avoid what are now practically random results (I suggest one could not predict the outcome of these decisions much better than flipping a coin would), we have to have clear, simple and perhaps simplistic rules that are less subject to the whimsical preferences of those who happen to be participating. That means sacrificing things like giving long-term significance direct consideration.
  Until some time in 2011, the only exception to giving usage consideration was for vital articles. At least that exception was well defined and objective (is the article on the VITAL list or not?). Then we opened up the long-term significance consideration to potentially any title. I suppose another way to deal with it is come up with more objective criteria that could be used to decide whether long-term significance should be weighed, but I don't know how you do that. That's why I've always opposed adding this highly subjective consideration to WP:PRIMARYTOPIC --B2C 01:30, 26 February 2013 (UTC)
  - There is so much that is subjective about writing a high quality encyclopedia—that subjectivity is part of thousands of discussions every day over giving due weight to certain aspects of an article, determining a Neutral Point of View, and much more, including titles. That's just what it takes to write good articles, and of course their titles also. To say that such discussions are based on "whimsical preferences" and produce "random results" is obviously something I would disagree with you on. It just isn't possible to write a serious encyclopedia (including the naming of titles) based entirely on data-driven objective data points. First Light (talk) 01:45, 26 February 2013 (UTC)
  - Until some time in 2011, the only exception to giving usage consideration was for vital articles. This is at best a misleading half-truth. Prior to 2011, page traffic was not given any special attention either. It was merely one tool that could be used. And the guidance explicitly stated: There are no absolute rules for determining how likely a given topic is to be sought by readers entering a given term; decisions are made by discussion between editors, often as a result of a requested move. Just a cautionary word for those who might fall for the illusion (delusion) of certainty offered by B2C. older ≠ wiser 02:14, 26 February 2013 (UTC)

Improperly titled disambig

Hi, the disambig Labor Party (United States) improperly has a parenthetical title. Could a dab-naming maven (JHunterJ, perhaps?) take a look? Thanks, --JaGa^talk 23:41, 24 February 2013 (UTC)

Fixed merged to Labour Party. -- JHunterJ (talk) 11:40, 25 February 2013 (UTC)

"Usage" criterion in WP:PRIMARYTOPIC: Correctly deciphering statistics?

I failed to convince people to support disambiguating the film The Boys in the Band, based on the original play. Probably the high number of views must have convinced others to oppose it. However, I am still not convinced that the high numbers should be primary topic of the same title. We have 90+ or 100+ people everyday surfing into the film, while the play gets low views (i.e. 30+). Why would stats of novel and film Doctor Zhivago (500 vs. 1,500) make the title more ambiguous, while stats of the play and film (30 vs. 100) does not sufficiently make the title ambiguous? --George Ho (talk) 21:24, 25 February 2013 (UTC)

The root cause of this problem is our giving any direct consideration whatsoever to so-called "long-term significance". That sets us up for absurd situations, where the decision is based entirely on which title is favored by those who happen to be participating in a given RM discussion, since each participant is free to "weigh" the competing criteria any way they want.

If we went solely by usage -- determining only whether one use is much more likely to be sought than all others -- then we would have a much more objective criterion and would virtually eliminate this problem.

Note that going solely by usage does not dismiss "long-term significance"; it simply puts the determination of how significant that "long-term significance" is on the dozens, hundreds or thousands of users sufficiently interested in the corresponding topics to view those articles (the more significant the "long-term significance", the more likely it is to be viewed), instead of on the typical handful or so of WP editors who happen to show up at a given RM discussion.

Trying to make sense of decisions like that are based on inherently conflicting guidance like this is pointless. The problem is primary topic having competing criteria, and until that's fixed, this problem will not be fixed. --B2C 21:51, 25 February 2013 (UTC)

The same issue is preventing rational consensus development at Talk:Madonna_(entertainer)#Requested_move. --B2C 23:58, 25 February 2013 (UTC)

~~I couldn't use the "long-term significance" criterion in that discussion. Both film and play are equally significant for a long term. --George Ho (talk) 03:23, 26 February 2013 (UTC)~~

Yet you used it. You argued that since the long-term significance of the two uses was equal, there is no primary topic. Never mind that one has a much higher page view count than the other. --B2C 05:11, 26 February 2013 (UTC)

My blooper again. Well, can we re-analyze the criteria right now and determine how numbers should make one topic primary over the other? How can I re-establish that "usage" is useless in the next proposal? How long must I wait until I do another proposal? I can't make same arguments again, unless I could pull an "Anne Hathaway". --George Ho (talk) 05:22, 26 February 2013 (UTC)

I don't understand why you're bothering at all. The film use is obviously and not surprisingly significantly more likely the one being sought when people search with "The Boys in the Band". So it is the primary topic. What's the problem? --B2C 05:28, 26 February 2013 (UTC)

... Well, I don't know if I should keep fighting for a change or another cause. Honestly, the play will go on future performances, while the film... is very dated. Also, per-day stats (200 vs. 50) should be put into consideration more than 90-day ones. Even highly popular topics with same name, like Take Me Home Tonight (film) isn't primary. The film (1,000 per day) is currently more popular than the song (100 per day), but neither is primary. Can anybody tell me the difference between "Take Me Home Tonight" and "The Boys in the Band"? --George Ho (talk) 05:40, 26 February 2013 (UTC)

WP:OTHERSTUFFEXISTS. It's actually possible that the "Take Me Home Tonight" arrangement could be improved, rather than taken as a proof of its numbers. You like per-day stats, because they give the illusion that the difference between the usages is small. per-hour stats (if they existed) would be even better at that, but we aren't looking for micro-trends. -- JHunterJ (talk) 12:35, 26 February 2013 (UTC)

Any other criterion to determine primacy of similar names?

If 90 days stats is more reliable than per-day stats, then I don't know if I have a chance to have the "usage" criterion revised. If "long-term significance" is a better criterion than "historical significance", then I don't know what else is remaining in my hands. I have "ambiguity" (how broad the word or phrase is?), "familiarity" (is one topic more familiar than other), and "quality" criteria (Good or bad article?), but none can overcome predominant two criteria. --George Ho (talk) 01:22, 27 February 2013 (UTC)

Daily stats are a guide and not a determining factor. They should be used to assist in the determination and not used to decide a question. One other point that we ofter forget about these numbers. When the topic name is ambiguous, even if we have a primary topic, these page view numbers must be considered in light of all of the other possible articles that it is ambiguous with. Those other names may or may not be listed on the dab page. Vegaswikian (talk) 01:49, 27 February 2013 (UTC)

You lost me there, Vegaswikian. If there are other articles to consider, they need to be listed on the dab page; if they're missing, they should be added by the considerer. -- JHunterJ (talk) 01:59, 27 February 2013 (UTC)

Dab pages list articles with a common name. They do not always list articles under unique names even if they are better know by a different name. If I have a misimpression here, I need to update some dab pages. Vegaswikian (talk) 02:52, 27 February 2013 (UTC)

If they are missing topics that are ambiguous with the title, those topics should be added. I'm having a hard time coming up with an example that fits the unique name/better known different name description, though, so maybe you are thinking of something I'm not. -- JHunterJ (talk) 12:02, 27 February 2013 (UTC)

There's another criterion often mentioned when determining primary topics, namely when one of the topics is the original use of the term and most other topics derived their names from the first one (this is frequently mentioned for movies based on a book, for example). This is sometimes conflated with the long-term significance criterion, but is really an independent one. Diego (talk) 09:50, 27 February 2013 (UTC)

Can you clarify? Doctor Zhivago is currently a dab page because neither film nor novel is primary at this time. --George Ho (talk) 10:01, 27 February 2013 (UTC)

The "original use" is often raised in local discussions, but it has no broad consensus. Sometimes the original use is primary, and sometimes it isn't; usage and long-term signficance determine which. You're right, it would be an independent criterion from the long-term significance, if there were consensus for considering it. -- JHunterJ (talk) 12:02, 27 February 2013 (UTC)

List of standardized parentheticals?

Okay, I know I saw something like this the other day (it was something like a bot-generated list of most common qualifiers), but now I can't find it. I am looking for some sort of guideline on how to choose the "correct" parenthetical qualifier for a disambiguation article title. For example, there is a page at Eddie Rodriguez about a baseballer. However, there are also both a Filipino writer/director/actor and a Texas State Representative by that name. If I was to try creating pages for them, (assuming they are proved notable) I suppose the first one should be called Eddie Rodriguez (actor) (since he acted in 145 films, and directed/wrote much fewer), but should the second one be Eddie Rodriguez (politician) or Eddie Rodriguez (representative) or Eddie Rodriguez (congressman) or what? --Atlantima (talk) 21:29, 27 February 2013 (UTC)

We have generally left the choice of disambiguating qualifiers to the content-area projects, so the film project specifies (actor) or whatever and the politics project specifies their preferred qualifier. -- JHunterJ (talk) 22:11, 27 February 2013 (UTC)

When I'm in need of guidance for a disambiguating term and I can't find a naming convention that specifies one, I go to a relevant category page and look at the other entries. In this case, Category:Members of the Texas House of Representatives indicates that the most prevalent disambiguator is (politician), or, if there is another politician with the same name, (Texas politician).--ShelfSkewed Talk 22:29, 27 February 2013 (UTC)

What is the primary topic for "trot"?

Odd case.

My instincts strongly suggest the topic most likely to be sought when someone searches with trot is the horse gait, but page view and WP:GOOGLETEST evidence surprisingly indicates that another use, a form of Korean music, is also highly likely to be sought, so highly that there is no primary topic.

Talk:Trot_(horse_gait)#Requested_move

--B2C 00:18, 1 March 2013 (UTC)

Here is a case were your instincts may serve you better than a random stats dump; the Korean music is named after the physical action of trotting after all. Also you're not supposed to go forum shopping like this - a neutral notification would have been better. Diego (talk) 06:37, 1 March 2013 (UTC)

2nd order disambiguation by birth date - RFC

Be advised that I have started an RFC at Wikipedia talk:WikiProject Disambiguation#2nd order disambiguation by birth date - RFC.--TonyTheTiger (T/C/BIO/WP:CHICAGO/WP:FOUR) 21:54, 4 March 2013 (UTC)

This discussion has been closed and replaced by Wikipedia_talk:Naming_conventions_(people)#RFC-birth_date_format_conformity_when_used_to_disambiguate.--TonyTheTiger (T/C/BIO/WP:CHICAGO/WP:FOUR) 07:42, 6 March 2013 (UTC)

Should the Pope Francis article have a hatnote?

There is a discussion at Talk:Pope Francis#Hatnote required about whether there should be a hatnote at the top of the article. This discussion would significantly benefit from additional participation. Thryduulf (talk) 21:53, 13 March 2013 (UTC)

Wisdom needed

Resolved

Vancouver has two time signals. The 9 O'Clock Gun and the Heritage Horns (link). Should a new article Vancouver time signals be created to include both with re-directs? The heritage horns are notable locally and by tourists but may not warrant a stand alone article. They are now only in Electra Building (Vancouver), but could also be linked from Canada Place where they live now as well as time signal and O Canada. Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL.--Canoe1967 (talk) 19:26, 18 March 2013 (UTC)

It shouldn't disambiguation at all, but a reasonable redirect target is 9 O'clock gun. That article is actually about a time signal - the others are about buildings that happen to have one (or in one case, used to have one, so it can't seriously be considered as a target). If you make an article about Heritage Horns, then you'd have a legit two-topic disambiguation page. However, I would think that until there is more material (that is, more than two signals), Vancouver time signals should redirect to something like time signals in Canada or list of time signals, unless you have material about more than just those two. Ego White Tray (talk) 01:55, 19 March 2013 (UTC)

I think the Pan Pacific is actually part of Canada Place. I may add sections Pan Pacific and Hertitage Horns to Canada Place and just create re-directs for now. I realize it isn't a dab issue but this page seems to have the most wisdom on matters such as this. I have contacted Canada Place about getting a sound file for the horns. They should be able to confirm that Pan Pacific is part of their complex. The creator of the horns is quite notable and has an article here as well.--Canoe1967 (talk) 10:00, 19 March 2013 (UTC)

Establishing a WP:PRIMARY within the brackets

We keep seeing at WP:RM move proposals seeking to shorten (reduce the precision) of parenthetical disambiguation, like this randomly generated example:

Let's assume for the sake of argument that the American is actually much more notable than the Brit (which he probably is). Is the current text of WP:DISAMBIGUATION clear about whether we do/don't establish who is the more notable in brackets. My understanding, given that no-one will ever type "Mark Smith (racing driver)" is that there is no benefit to such moves, and it reduces utility of both Google results and the R/H search box autocomplete. Can someone comment on this please? In ictu oculi (talk) 06:08, 25 March 2013 (UTC)

I don't think that titling decisions should entertain discussions of relative notability. We have trouble enough defining sufficient WP:N-notability without wanting to pretend that we can quantitate degrees of notability. Deferring this poor measure to ghits is an abrogation of our aim to create reference knowledge accessible to all people. Ghits are biased and non-representative of our readership.

Titles should be precise enough that no reader is ever astonished on arriving at the page. Where there is ambiguity, each should be disambiguated to equal precision, not all but the most notable.

My understanding is that titles are not written to facilitate the search function. The search function is quite capable of looking beyond the title, and searching the article text. My view of titles is that the title, as it appears elsewhere, embedded in another article or written externally as a link to a particular page on Wikipedia, should tell the reader what the article is about.

I therefore thnk that WP:PRIMARY should be reduced in impact, and that article titles should err on the side of being too precise. I don't think that WP:PRIMARY should have any applicability within parentheses in a title. --SmokeyJoe (talk) 07:05, 25 March 2013 (UTC)

I agree. Once parentheical disambiguation is necessary, the goal should be to disambiguate adequately between all uses of the term, though there may be difficult cases where different criteria are used for disambiguation. I am also not too happy about London (electoral district). --Boson (talk) 08:35, 25 March 2013 (UTC)

Not the first time has come up on this talk page (here's a similar discussion that springs to mind), but I can't recall a concrete consensus being reached. Personally, I agree with you that if an article is not the primary topic it should have a unique disambiguation and I'd support adding something of this effect to the guideline. Jenks24 (talk) 10:10, 25 March 2013 (UTC)
Something to this effect is present in some of the naming guidelines. IMO, that's where it should be; disambiguation will work once the titles are different, and the choice of which differentiators to use is best left to the content projects and their naming conventions. -- JHunterJ (talk) 12:58, 25 March 2013 (UTC)

I personally think that the reduction in disambiguation is poor, since it will result in even more contentious renaming battles than already occur, since as of right now, people already make a big fuss over which article should live at the undisambiguated title. If we reduce disambiguation for disambiguated titles as well, these battles will be ever more frequent, and result in more move wars. If it isn't the primary topic, we should treat all other uses as needing sufficient disambiguation to distinguish each from each other. The choice of disambiguatory term is also rather arbitrary for Wikipedia. In the example provided, it uses "racing driver", and "British" or "American", but someone else may use "United States" and "United Kingdom", or "US" and "UK", and "racing driver" may be replaced by "driver" or "pilot" or may use the current series/discipline they're in "F1 pilot" or "rally driver", etc. -- 65.92.180.137 (talk) 04:41, 29 March 2013 (UTC)

Well it seems there's not absolute agreement here. I understand JHunterJ's comment to say that the question of whether to establish a primary within brackets should be left to projects. I have noticed in fact that (song) seems to be a particular category where project participants seem determined to argue over who gets the prime "(song)" spot, leaving (Walter Mitty song) (John Smith song) (Joe Bloggs song) as 2,3,4 runner ups. Sort of an X-factor contest for the "(song)" spot. But is there really value to the project as a whole in having some projects do this? A lot of these projects only have 2-3 active editors and when 1 leaves, the view changes and renames happen in the other direction. In ictu oculi (talk) 07:35, 1 April 2013 (UTC)
If the project as a whole thinks that all of the content-area projects should standardize on this, I think it's a discussion for WP:VPP or WT:NC. Formulating overriding rules within the dab project for their naming conventions is IMO going to cause more arguments, not fewer. And I have no objection to that potential broader consensus. -- JHunterJ (talk) 11:08, 1 April 2013 (UTC)

URLs on category pages

There is a discussion at Wikipedia talk:Categorization#Reference resources about whether WP:External links should be permitted in the descriptions on category pages. Since I propose that cat-page descriptions be handled similar to dab descriptions (e.g., useful description, but no reliable sources and no external links), perhaps some people here will be interested. If you have an opinion, please comment there. WhatamIdoing (talk) 21:50, 28 March 2013 (UTC)

Matters for resolution pertaining to the Frank Burns (disambiguation) page

A discussion germane to the one above, regarding criteria to be applied in the choice of WP:PRIMARYTOPIC in human name disambiguation pages, as well as the specificity of the disambiguating qualifier to be appended to a fictional character who is a member of the cast of a TV series, is currently active at Talk:Frank Burns (disambiguation)#Requested move. —Roman Spinner (talk)(contribs) 13:48, 10 April 2013 (UTC)

Need advice/help with disambiguation page

It all started by searching for a medieval papal legate, which I knew only by his ecclesiastical name (in latin) "Pelagius". In the disambiguation page, his article was not there, so I assumed it was not created. I looked around the fifth crusade article and finally found his article under his native name of "Pelagio Galvani" (Pelagio is the Leonese version of "Pelagius" kind of like "Theodore" being an english version of the Greek "Theodoros".

So, I found out that under the disambiguation page of "Pelagius" there was a see also for "Pelagio" for those named as such. I am wondering what are the guidelines for disambiguation pages for cognates? For instance, should there just be one disambiguation page that redirects from "Pelagius" Pelagios" "Pelagio" and every other cognate, then listing the various people in sections by whichever name they are most known for? Because I almost was unable to find an article just because of there being three disambiguations for a single name.

I was thinking of even putting the Pelagius Galvani into the "Pelagius" disambiguation, thinking there would not be much harm in it being in both the disambiguation page for "Pelagius" and Pelagio". but I did not want to do this without seeking input from others. So, anyone have any comments/advice/ideas?75.73.114.111 (talk) 23:11, 13 April 2013 (UTC)

As it sounds helpful, I advise just doing it. If there were some reason to not do it, which I think is unlikely, then someone will say so. --SmokeyJoe (talk) 01:14, 14 April 2013 (UTC)
A line in Pelagius (disambiguation) like the below might work.

Pelagius or Pelagio Galvani (c. 1165—1230), Leonese cardinal, canon lawyer, papal legate and leader of the Fifth Crusade

Thanks, SchreiberBike (talk) 02:47, 14 April 2013 (UTC)

Sounds like a good plan, thank you for your help!75.73.114.111 (talk) 23:52, 15 April 2013 (UTC)

Get rid of WP:PRIMARYTOPIC

Wondering around RMs, I'm only becoming more convinced that WP:PRIMARYTOPIC does far more bad than good. It is very poorly explained, with people straight off making opposite readings. Probably this is because it is not well defined. It is a huge invitation to assert a majority bias. It draws people to study ghits, which are poorly aligned with the goals of the project. It creates confusion and controversy in what should be the simplest of cases. Better, it would be, to just disambiguate when there potential for ambiguity, on the part of any reasonable group of readers. I think it should be got rid of as unhelpful. --SmokeyJoe (talk) 13:27, 8 April 2013 (UTC)

I'm uncertain that I understand what you're proposing. As a couple of examples, do you advocate that Dog (disambiguation) be relocated to Dog and that Barack Obama become a disambiguation page listing Barack Obama II and Barack Obama, Sr.? —David Levy 13:38, 8 April 2013 (UTC)

I have to say, if this were a formal proposal, it would have my full support. Saying that WP:PRIMARY topic is "unhelpful" is putting it mildly: the amount of time Wikipedians waste on discussing which topic is more "primary" than the other is simply mind-boggling, and the results of such discussions often remain contentious anyway. Yet all that's often needed in such discussions is a little common sense. I have difficulty imagining someone coming up with serious reasons to turn something like "Barack Obama" into a dab page, but in most cases it's not so clear-cut. At any rate, even if something like "Dog" becomes a disambiguation page, the harm would be rather minimal, and if doing so saves up editors' time otherwise wasted on counting ghits, popularity votes, and abstract arguments about "educational value" or "enduring notability", I'd say the net positive is well worth it.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); April 8, 2013; 14:05 (UTC)

I don't quite understand your position. We can't stop recognizing the existence of primary topics without making Barack Obama a disambiguation page. It certainly is true that many cases are less clear-cut than that one is, and it's reasonable to opine that we should adjust the threshold to draw such a distinction only in the most clear-cut cases, but abandoning the entire concept behind WP:PRIMARYTOPIC would mean never drawing the distinction; a disambiguation page would occupy the base title for any term requiring disambiguation. —David Levy 14:29, 8 April 2013 (UTC)

David, to clarify, I'm not for stopping to recognize the existence of primary topics altogether; I'm for removing of a bunch of artificial criteria which are vaguely defined, interpreted by different editors differently, and ultimately lead to wasting enormous amounts of time which could be put to better use elsewhere. Note that under any arrangement there is nothing preventing editors from continuing to discuss potential moves, but getting rid of the artificial nonsense PRIMARYTOPIC currently endorses would simplify the RM process significantly. If there is substantial disagreement as to which topic is primary, keep the dab page at the primary location. If there is a clear consensus that some topic is indeed primary, then so be it. It can't be simpler than that (and doesn't have to be) and it doesn't lead to disambiguation pages occupying all possible "primary" nodes either. Assessing whether there is a consensus or not is a lot simpler than assessing a bunch of esoteric criteria often created and enforced by people who do little productive work here beyond endlessly arguing policy and coming up with new creative way to regulate everything. If such people can be shut off, everyone else would benefit.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); April 8, 2013; 14:56 (UTC)

David, We can stop recognizing the existence of primary topics without making Barack Obama a disambiguation page. There is no policy requiring making Barack Obama a disambiguation page except for the existence of the guideline section with the shortcut PRIMARYTOPIC.

I want to abandon the superficial, not-proven-useful section in the guideline, not abandon an entire underlying concept. But if the concept is to remain explicit, it needs to be better than what’s there now. Removing WP: PRIMARYTOPIC does not imply that we are writing into policy that everything that was once written under WP: PRIMARYTOPIC was false. --SmokeyJoe (talk) 01:16, 9 April 2013 (UTC)

David, We can stop recognizing the existence of primary topics without making Barack Obama a disambiguation page.

No, we can't. In such a scenario, there would be no reason to treat Barack Obama, Sr. differently than we treat Barack Obama II.

There is no policy requiring making Barack Obama a disambiguation page except for the existence of the guideline section with the shortcut PRIMARYTOPIC.

Let's set aside our polices and guidelines for a moment. Without the concept of "primary topics" (whether written down or not), there would be no justification for assigning the base title to a particular Barack Obama's article. The very act of doing so is an acknowledgement that said article's topic is primary.

I want to abandon the superficial, not-proven-useful section in the guideline, not abandon an entire underlying concept.

Oh. Then why did you just advocate that we "stop recognizing the existence of primary topics"?

But if the concept is to remain explicit, it needs to be better than what’s there now.

I noted above that the guideline can be modified. —David Levy 03:52, 9 April 2013 (UTC)

Hi David. On Obama: I mean that we don't need to be this explicit in recognising the primary topic to continue with current practice. Other than that, I'd like to leave Obama as not a typical page.

Then why did you just advocate that we "stop recognizing the existence of primary topics"? Trying again then... I want to stop recognising the something under the term "Primary topic". I would prefer to speak in terms a topic that overshadows all others in terms of notability/recognisability, especially when the other topics derive from the first. A problem with "primary topic" is that it sounds like it is something that is very well defined, as in there are not shades of primariness.

Modifying the text: Yes, it could be modified. Roughly speaking, do you think that is needs (1) wholesale rewriting; (2) work; (3) tweaking only, it's basically OK. --SmokeyJoe (talk) 04:34, 9 April 2013 (UTC)

Trying again then... I want to stop recognising the something under the term "Primary topic".

Thanks for clarifying. I was referring to the underlying concept, not the term used to describe it (which certainly can be changed if we come up with something better).

Modifying the text: Yes, it could be modified. Roughly speaking, do you think that is needs (1) wholesale rewriting; (2) work; (3) tweaking only, it's basically OK.

My impression falls in the range of 3, but I'm open to the possibility that more extensive change is called for (depending on what problems exist and what solutions are offered). —David Levy 09:34, 9 April 2013 (UTC)

(ec) OTOH, I've only been more convinced of the need for WP:PRIMARYTOPIC, after seeing some of the poor criteria editors sometimes use in RMs. And as David Levy notes, doing away with it leads to worse absurdities that you're finding with it, so it does more good than bad. -- JHunterJ (talk) 14:30, 8 April 2013 (UTC)

JHunterJ, can you expand on how cutting this section (with the option of starting again with different words) will (definitely) lead to worse absurdities? I don't think you can, because guidelines don't define practice, they reflect it. Also note that I am not asserting absurdities. On the whole, we make reasonable decisions despite the confusion injected by WP:PRIMARYTOPIC. My point is that for editors looking for guidance, they are being poorly served. I maintain that it does little good, that correct decisions make that are consistent with PRIMARYTOPIC were going to be made that way anyway (based on sources, etc). --SmokeyJoe (talk) 01:22, 9 April 2013 (UTC)

No need. This section will simply fill with sound and fury, signify nothing, and then be archived. If your point is that editors are being poorly served, there may be improvements that stand a snowball's chance of finding consensus, but leading with "Get rid of WP:PRIMARYTOPIC" ensures this won't be the discussion that finds them. -- JHunterJ (talk) 01:33, 9 April 2013 (UTC)

JHunterJ, I have been looking at this for a long time. I really think that removing the section will be a dramatic improvement without bad consequences. At worst, people who like to bluelink the allcaps oneword may have to think their rationale out more clearly. You say "it does more good than bad". You normally say sensible things. I would really like to understand why you say this. If you think I am just pot stirring or game playing or otherwise making noise without point, please respond on my talk page. --SmokeyJoe (talk) 01:51, 9 April 2013 (UTC)

Our guidelines are intended to document how we do things. If the one in question is poorly written, it should be fixed, not eliminated.

The logic that editors will continue to behave sensibly without such documentation can be applied to the entire Manual of Style or even every Wikipedia policy and guideline. Editors who regard them as needless bureaucracy (and wish to delete them en masse) actually have done so. You're presenting essentially the same argument on a smaller scale. —David Levy 03:52, 9 April 2013 (UTC)

I think the MOS is over-written. It probably happened because there are so many pages of them. On Wikipedia Policies and guidelines, I think they are generally well balanced on depth of detail. Am I on a limb here?

I think the one in question is built on such a poor word choice, it would be easier to start again. I guess that you would like to see a replacement first. --SmokeyJoe (talk) 04:41, 9 April 2013 (UTC)

In my view, nothing else makes sense.

Our guidelines are descriptive, not prescriptive. We don't do the things described at WP:PRIMARYTOPIC because they're written there; they're written there because we decided to do them. If we decide to do something else instead, we can adjust/replace WP:PRIMARYTOPIC accordingly. Until such time, the documentation should remain. Removing it wouldn't magically change how we do things or solve any problems related thereto. It would amount to nothing more than obfuscation. —David Levy 09:34, 9 April 2013 (UTC)

I would be more inclined to go in the opposite direction and say that for pages like Battery and Mercury and Springfield, we should just pick a "most likely" meaning, put that topic at that title, and move the disambig to the "Foo (disambiguation)" title. bd2412 T 17:16, 8 April 2013 (UTC)

Could you clarify your point, BD2412? Mercury, for instance, seems an excellent example of how impossible it would be to choose which page could stand without a disambiguator (that is, which one would be the primary topic). I don't even know how you'd "prove" whether the planet or the element was the most likely intended destination. Cynwolfe (talk) 17:32, 8 April 2013 (UTC)

There I would contend that a chemical element is more fundamental a topic than a planet, which is in turn a more fundamental topic than a god or a record company. This disposes of popularity questions, and focuses on which topic is primary from a standpoint of importance to an encyclopedia. However, I would still have incoming links to the page piped through Mercury (element), just so we could check the unpiped links to be sure they intended that meaning of Mercury. bd2412 T 20:34, 8 April 2013 (UTC)

Both the element and the planet names derive from Mercury (mythology). Ususally, the inclusion of the word “fundamental” indicates that the statement is false due to excessive simplification --SmokeyJoe (talk) 00:12, 9 April 2013 (UTC)

That's exactly the wrong approach, in my view. Our goal is to provide access to the encyclopedia articles sought, not to favor topics that we consider more important. The latter doesn't serve readers, and it's the last thing that will prevent arguments among editors.

Also, requiring the use of redirects in running prose would be a never-ending headache. —David Levy 03:52, 9 April 2013 (UTC)

"Mercury" is a textbook example of a term with no primary meaning. —David Levy 03:52, 9 April 2013 (UTC)

We can't get rid of WP:PRIMARYTOPIC, for as others have noted that would mean making Barack Obama a dab page, but we can surely improve it.

The problem with WP:PRIMARYTOPIC probably starts with the ambiguity of the meaning of the word "primary". For example, just above, BD2412 interprets the chemical Mercury to be "primary" over the planet, because the chemical is "more fundamental". Now, "fundamental" is a synonym for "primary" and an appropriate substitute in some contexts, but I've never before encountered it being used in primary topic discussions, and I thought I'd seen everything. "Most fundamental" is NOT what "primary" means in the context of "Primary Topic"!

What "primary" means in "PRIMARY TOPIC" is "most commonly used". So, what if we renamed Primary Topic to something like, "MOST COMMON USE TOPIC"?

Regardless of what we call it, to discuss it, it's critical to understand why we have it. The reason is, and has always been, to serve our readers best. The quintessential example, perhaps, is Paris. Even though there are many uses of "Paris", most people searching for "Paris" would be looking for the city in France, so we put that topic at Paris. Thus we have the traditional definition of "primary topic" of a given term:

it is highly likely—much more likely than any other topic, and more likely than all the other topics combined—to be the topic sought when a reader searches for that term.

The relatively recent addition of the second criterion to the definition, long-term significance, has made things worse, in my opinion. Not only is this criterion far more ambiguous/subjective than the traditional "likely" one is, it often suggests a different answer, thus creating, rather than resolving, conflicts. What kind of guideline is that?

I suggest that primary topic arguing has gotten much worse since the addition of long-term significance, and we could go a long way towards improving the situation by removing it.

As to arguments about "popularity" and counting ghits, it wasn't all that bad before "long-term significance" was added. Without it, unless popularity including counting ghits makes it clear that one of the uses of a given term is "much more likely than any other topic, and more likely than all the other topics" to be the one being sought, there is no primary topic, period. Hence Mercury is, and should be, a dab page. --B2C 21:19, 8 April 2013 (UTC)

Also concerning: "educational value." How and, more importantly, why are we supposed to define it? "volume of coverage in reliable sources" would be a reasonable alternative which is quantifiable for the purposes of debate and should be agreeable to everybody. For the record I don't agree with throwing out primary topic. Marcus Qwertyus (talk) 21:42, 8 April 2013 (UTC)

And if you can simply define some type of criteria, how do you objectively determine it? Vegaswikian (talk) 21:58, 8 April 2013 (UTC)

Some good responses.

The best outcomes I see with WP:PRIMARYTOPIC, it is redundant to WP:COMMONNAME, with no other topic challenging for the same COMMONNAME.

I do not have a problem with Dog. Dog could be at Dog (Canis lupus familiaris), but "Dog" is the clear COMMONNAME with no other topic has a serious, even time or place, claim to be at Dog. If the article were at Dog (Canis lupus familiaris), Dog should redirect straight to it, without any need for a PRIMARYTOPIC argument.

Barack Obama is an extreme case. A hot, sensitive, promotionally written article that I would not touch. In general, I think biographies should be more precisely titled, whether by parenthentical birth-death and/or main occupation. If this were a good idea, I'd begin with long dead people, then presidents of small countries, and would not propose a rename of the most significant person to the most significant group of editors at this time. I don't think it would even lead to a helpful debate.

Mercury is an excellent example (as I guess there are few emotive interests) of a title where you can imagine separate people in different times and places arguing that their use (bias) is "PRIMARY".

The sort of article RM that have motivated me are human commercial products, such as songs and films. Including "Big".

At the very least, I think the shortcut "WP:PRIMARYTOPIC" should be abandoned as historically confusing, and if replaced (NB we do not NEED ALLCAPS ONEWORDS for every concept), it should be something much more tightly defined, so that we can agree on what it means, and ordinary editors can interpret unambiguously. However, I still feel that every appropriate use of PRIMARYTOPIC (the relative few of them) is redundant to a better explanation for the best title. --SmokeyJoe (talk) 22:52, 8 April 2013 (UTC)

I do not have a problem with Dog. Dog could be at Dog (Canis lupus familiaris), but "Dog" is the clear COMMONNAME with no other topic has a serious, even time or place, claim to be at Dog.

"Dog" is the common name of multiple subjects. One of them is much more well known than the others are, which makes it the primary topic.

If the article were at Dog (Canis lupus familiaris), Dog should redirect straight to it, without any need for a PRIMARYTOPIC argument.

We don't redirect "Foo" to "Foo (disambiguation term)". If you believe that we should, I strongly disagree.

In general, I think biographies should be more precisely titled, whether by parenthentical birth-death and/or main occupation.

Why? —David Levy 03:52, 9 April 2013 (UTC)

""Dog" is the common name of multiple subjects". Really? For example. Are we working with different interpretations of WP:COMMONNAME, specifically that a COMMONNAME has prevalence in reliable English-language sources? Vernacular is excluded. To my reading, every other usage on the disambiguation page involves another word or some contextualisation before use.

We don't redirect "Foo" to "Foo (disambiguation term)"? I don't see a problem in principle, but I haven't deeply considered the question. I think we are agreed to leave Dog at Dog anyway.

"In general, I think biographies should be more precisely titled, whether by parenthetical birth-death and/or main occupation.

Why?" A: Because biographies are a major growth area for the project, and recognisable names are often not unique identifiers. We therefore have to disambiguate many of them somehow. Middle names, sometimes, but middle names, if they exist, may not help recognisability. Parenthetical birth and death dates are likely to help recognisability because the searcher probably has some idea of the historic period they belong to. The searcher probably also has some idea of the occupation. I think this should be done, it is often done, and I think it should be done more. See William Henry for example. In the listing, it is natural to disambiguate and sort by birth and death years as well as occupation. I think this would work well in titling. At the moment we sometimes disambiguate biographies oddly, such as William Henry (brother of Patrick Henry). My preference would be to recommend a systematic method, allowing the article writer freedom to accept or reject the recommendations. The exceptions to usual disambiguation are: (1) living people, and (2) supernotable people (where WP:PRIMARYTOPIC would apply). (1) doesn't bother me, I'm happy to see more precise disambiguation happen later. (2) feels wrong to me. We have a arbitrary-seeming line of supernotability, and assume that no one needs precision for the supernotable. This has some downsides, including providing poor examples of titling for non-regular editors. --SmokeyJoe (talk) 05:13, 9 April 2013 (UTC)

"'Dog' is the common name of multiple subjects". Really? For example.

"In general, I think biographies should be more precisely titled, whether by parenthetical birth-death and/or main occupation. Why?" A: Because biographies are a major growth area for the project, and recognisable names are often not unique identifiers. We therefore have to disambiguate many of them somehow.

Perhaps I misunderstood. I interpreted your statement to mean that we should routinely append such information to biographical articles' titles (even when disambiguation isn't needed). —David Levy 09:34, 9 April 2013 (UTC)

I do think that we should routinely disambiguate articles even if not strictly needed, for ease of recognition and consistency, as a recommendation. I think that the precision used should be consistent across biographries, and should not be inversely proportional to importance. For example, William Henry (brother of Patrick Henry) should be moved to William Henry (1734–1785, politician) (subject to an ongoing format discussion at Wikipedia_talk:Naming_conventions_(people)#RFC-birth_date_format_conformity_when_used_to_disambiguate), even if he is the only notable William brother of a Patrick Henry. I recommend moving George Washington to George Washington (1732–1799, US President) and George Washington (Washington pioneer) to George Washington (1817–1905, Washington pioneer). Consistency is nice, but I have in mind good service for readers such as the children in Borneo who have difficult access and no cultural association with these people. I think that titles should feed an indexing system, and that 50 characters is an appropriate title length. --SmokeyJoe (talk) 11:59, 9 April 2013 (UTC)

You certainly are entitled to your opinion. You're aware, I assume, that the above is far removed from anything for which consensus has been established. —David Levy 13:05, 9 April 2013 (UTC)

Moderately far, yes. But I do like the idea. However, it is tangential to the main goal here of clarifying PrimaryTopic. Cutting it would clarify WP:DAB and dissuade non RM regulars from clumsily quoting it, but it seems you are going to require me to work out what exactly the underlying concept is. Ok. My opinion is directly relevant in declaring an opposition to an effect of WP:PRIMARYTOPIC, which is reduced precision for more popular subjects. --SmokeyJoe (talk) 13:35, 9 April 2013 (UTC)

Fair enough. I don't agree that our titles should be utilized in the manner that you've described, but I understand the underlying concern. —David Levy 13:45, 9 April 2013 (UTC)

I would like to see rejected arguments based on likely searches. Arguments of likelihood are arguments from majority bias.

Regarding the quintessential Paris, looking though the disambiguation, I see that no other topic has a reasonable COMMONNAME argument to be at just "Paris". The second best would be Paris (mythology), and for this subject I note that no one refers to this Paris without the subject being first contextualised. So Paris can be justified from policy without reference to PRIMARYTOPIC. Although on the other hand, it would hurt no one if Paris were at Paris, France, and it would help people who do no immediately know Paris, foreign children, for example. I remember someone believing Paris to be the capital of Europe. --SmokeyJoe (talk) 23:01, 8 April 2013 (UTC)

New York is another example for what is the primary topic. There are at least 3 articles that could make the claim, but how do you prove that the city or the state is the primary topic? Vegaswikian (talk) 23:16, 8 April 2013 (UTC)

If I had my druthers, we'd rename Georgia (country) to the native Sakartvelo and be done with that disambig. However, looking to names for which the "primary topic" could be more subject to dispute, what about Apple and Grape? Orange and Lime are disambigs. What about Europe and Asia and Florida and Kansas and Chicago? bd2412 T 23:38, 8 April 2013 (UTC)

"New York State" versus "New York City". That one could have gone either way. The original colony was, what is now, the city, certainly not the current state according to defined boundaries. I'm sure the people of the city refer to as "New York", and the upstate people refer to their state as "New York". Normally, in my experience, it is pre-contextualised as to whether the conversation is about cities or states. I don't think there is anything in the current WP:PRIMARYTOPIC that helps decide the question, and so yet again WP:PRIMARYTOPIC is serving no useful purpose. I think the guiding principle for New York should be the choice made by the original author, Trimalchio (talk | contribs) at 01:09, 1 October 2001, subject to a possible later clear (not rough) consensus to change. NB. This is actual practice.

On Sakartvelo, WP:PRIMARYTOPIC topic is useless due to not advising on whose perspective to consider it from. This one looks like a worthy RM discussion. I see Talk:Georgia_(country)#New_move_proposal and a completely useless reference to "primary topic", as if the user thinks it means something.

As I said above, we are not suffering disputes over Apples and Grapes, Oranges and Limes, and so PRIMARYTOPIC is not existing to help them. The problems are worst with recent human products. Europe, Asia, Florida, Kansas and Chicago? What about them? None were named with reference to PRIMARYTOPIC. --SmokeyJoe (talk) 00:04, 9 April 2013 (UTC)

SmokeyJoe, "Arguments of likelihood are arguments from majority bias." Indeed. Since in this case (choosing article titles) the goal is serving the majority best, arguments from majority bias are appropriate.

"Although on the other hand, it would hurt no one if Paris were at Paris, France". Very few titles could hurt anyone. That's rarely a consideration in deciding titles. The problem is having multiple choices all of which could be reasonably supported... how do we decide which one to use? Yes, yes, I know, "by consensus", but I mean, how does each person decide which one to use? Based on what criteria, exactly? If you and I apply the same criteria to the same situation and come up with different titles, we're not looking at the criteria in the same way - the criteria is vague, and we have no objective means to resolve such differences, except to improve clarity in the criteria.

This is why I advocate more and better criteria. It would not be okay if Paris, France met our criteria just as well as Paris, for that would mean anyone could reasonably call for a move periodically, without end. Thankfully, that's not the case: because of common name, primary topic and concision, we clearly prefer Paris to Paris, France, and that's a good thing because it means this case, and countless others, are settled (at least as settled as anything can be on WP).

If we loosen the criteria, like by getting rid of primary topic, we would introduce more ambiguity and thus more conflict into the title decision process. That would be a bad thing. In contrast, by removing the long-term consideration criterion from primary topic, we would be removing ambiguity and thus reducing conflict... a good thing. --B2C 00:09, 9 April 2013 (UTC)

B2C. Is the most direct service of the majority the goal of choosing article titles? I think not. The majority, especially as measured by ghits, are already well served. We should aim instead to serve all humanity. To serve broadly means making special efforts to serve those disfavoured by the prevailing bias. We should think of readers who do not know how to search. We should think of readers who have only a print version of selected pages.

When speaking of Paris, you seem to me to be too heavy on the need for criteria. This criteria-based decision making feeds into expert-pleb style governance, disempowering the ordinary editor. Disempowering the ordinary editor is to stifle the future growth of the encyclopedia. Instead, we should trust the editors who build to be doing things sensibly. On the question of Paris, the page naming decision by Zundark at 11:45, 6 November 2001 should stand until there is a clear consensus that something should be changed.

“If we loosen the criteria, like by getting rid of primary topic”. That’s where you are mistaken. WP:PRIMARYTOPIC is not serving by providing defined criteria. Define it better and it may work better. However, I think it is in a rut.

I disagree with removing the concept underlying long-term consideration criterion. An encyclopedia is best considered as a historical document, things are best written from a historical perspective, and titles should reflect the perspective of the content. “Long term consideration” is a variation on WP:NOTNEWS and Wikipedia:Recentism.

“Anyone could reasonably call for a move periodically”. Certainly not. A consensus is required for a move. The longer the history of proposed moved, the clear the consensus required. “Periodic” calls for moves quickly becomes received as disruptive editing. --SmokeyJoe (talk) 01:53, 9 April 2013 (UTC)

How has such a simple idea become so complicated? If we want to get rid of Primary Topic why can't we? Mercury the element becomes Mercury (element), Mercury the god becomes Mercury (god), Mercury the planet becomes Mercury (planet), and Barrack Obama becomes Barrack Obama (44th president of the United States). Anyone who still doesn't know what they're looking for, even with a helpful drop down menu on the search bar, can go to a dab page.--Ykraps (talk) 00:08, 9 April 2013 (UTC)

The simple answer is that some editors believe that disambiguation is bad and predisambiguation is worst. Then you have those who will argue that the first article has a god given right to the main name space. Vegaswikian (talk) 00:12, 9 April 2013 (UTC)

If we get rid of Primary topic, we get rid of those arguments.--Ykraps (talk) 00:20, 9 April 2013 (UTC)

If that is due to improved acceptance of disambiguation not being bad, then maybe getting rid of the primary topic guideline is also not bad. But would editors accept this? Is it possible to get a consensus in support? Vegaswikian (talk) 00:26, 9 April 2013 (UTC)

I think most editors find PT an unnecessary distraction. Everybody seems content to go to the index if they want to look something up in a book. Nobody starts crying because it hasn't magically fallen open at the page they wanted.--Ykraps (talk) 00:35, 9 April 2013 (UTC)

Ykraps: Do you seriously advocate that we move Barrack Obama to Barrack Obama (44th President of the United States)? If so, why? —David Levy 03:52, 9 April 2013 (UTC)

So we can get rid of PT. Why shouldn't we move him?--Ykraps (talk) 06:18, 9 April 2013 (UTC)

We don't have to move Barrack to get rid of PT. --SmokeyJoe (talk) 07:12, 9 April 2013 (UTC)

Why? Where would entering Barack Obama in the search box take you?--Ykraps (talk) 07:33, 9 April 2013 (UTC)

Barack Obama is a very stable title regardless of the guideline section being discussed here. The connection between PT and BO is very weak, non-existent even. In practice, many people are well used to ignoring the actual text of PT. --SmokeyJoe (talk) 08:08, 9 April 2013 (UTC)

Barack Obama is a very stable title regardless of the guideline section being discussed here.

I agree that eliminating WP:PRIMARYTOPIC wouldn't result in a realistic possibility of the Barack Obama article being renamed. Our guidelines' text is descriptive (not prescriptive), and removing it wouldn't magically cause a convention that it describes to disappear. It would merely increase the difficulty experienced by editors seeking to learn about or explain said convention.

The connection between PT and BO is very weak, non-existent even.

On the contrary, WP:PRIMARYTOPIC documents ~~the~~ a principle on which the article's title is based. —David Levy 09:34, 9 April 2013 (UTC)

I know little of the Principles of PRIMARYTOPIC, but I think that's revisionism, and that it is much more likely that the original author of the article took the title from the title predominantly used in the sources at the time.

Obama's campaign website
State Senate biography of Obama
Obama's State Senate website

I think Meelar drew the information from his sources, which is what we should do.

I think it very unlikely that Meelar considered the likelihood of searches for the senator against searches for his father or other Obamas. --SmokeyJoe (talk) 11:25, 9 April 2013 (UTC)

I don't mean that the article's title was determined by consulting the section and applying its advice. I'm referring to the core principle that when a common name shared by multiple subjects is associated with one much more strongly than it is with all others combined, it's used as the title of that subject's article.

Had Meelar titled the article something other than "Barack Obama", it would have been moved to that title. Alternatively, if Wikipedia's standard procedure were to always use titular disambiguation in such cases, it would have been moved from "Barack Obama" to something else.

Of course, I should have written "a principle" (not the principle). The aforementioned WP:COMMONNAME obviously plays a major role. —David Levy 13:05, 9 April 2013 (UTC)

The descriptive not prescriptive mantra is not strictly true, unless using twisted definitions such as "policy describes ideal practice", where "ideal" means "what I would like you to do". In reality, much policy contains prescription mixed in, in this this case certainly. --SmokeyJoe (talk) 11:25, 9 April 2013 (UTC)

I don't think that you're fully understanding what's meant by "descriptive, not prescriptive".

Certainly, Wikipedia's guidelines are intended to advise users of how they should go about editing the encyclopedia. No one disputes that. We mean that the guidelines themselves don't dictate anything; they aren't like laws that take effect when enacted. They're intended to describe the practices on which the community has settled via consensus. We want people to do what's been written, but not because it's been written. It's been written because it's what we've decided to do. —David Levy 13:05, 9 April 2013 (UTC)

We don't have to move Barrack to get rid of PT.

That depends on what Ykraps means by "PT" — the guideline section or the underlying concept. My impression is that Ykraps means the latter. —David Levy 09:34, 9 April 2013 (UTC)

Why shouldn't we move him?

If you mean that the title "Barack Obama" no longer would lead to his article, that would be extremely unhelpful to readers. If you mean that the title "Barack Obama" would redirect to his article, such a change would provide no benefit.

Regardless, if titular disambiguation were called for, "Barrack Obama (44th President of the United States)" is excessively precise and inferior to various alternatives. —David Levy 09:34, 9 April 2013 (UTC)

We don't have to move Barrack to get rid of PT.

That depends on what Ykraps means by "PT" — the guideline section or the underlying concept. My impression is that Ykraps means the latter. —David Levy 09:34, 9 April 2013 (UTC)

Why shouldn't we move him?

If you mean that the title "Barack Obama" no longer would lead to his article, that would be extremely unhelpful to readers. If you mean that the title "Barack Obama" would redirect to his article, such a change would provide no benefit.

Regardless, if titular disambiguation were called for, "Barrack Obama (44th President of the United States)" is excessively precise and inferior to various alternatives. —David Levy 09:34, 9 April 2013 (UTC)

Barack Obama is a highly distinctive name. What about the many people named John Adams, Thomas Jefferson, Andrew Jackson, Benjamin Harrison, and Jimmy Carter? bd2412 T 12:25, 9 April 2013 (UTC)

What about them? —David Levy 13:05, 9 April 2013 (UTC)

Under this proposition, it seems that they would all be moved to disambiguated titles, with the undisambiguated title being a disambiguation page. I think that would be equally unhelpful. bd2412 T 13:18, 9 April 2013 (UTC)

Agreed. —David Levy 13:45, 9 April 2013 (UTC)

Having Barack Obama go to Barack Obama only makes it easy for people who are looking for the 44th president of the United States. If you are looking for another Barack Obama, then it’s actually quite difficult. Entering Barack Obama (disambiguation), a perfectly logical thing to enter if you are aware of Primary topic, takes you to a page that doesn’t exist, although from here you can find a link to Obama (surname); I don’t know why, to me that is a step backwards, I’ve already made it clear that I’m looking for a Barack Obama. Once at the Obama (surname) disambiguation page, if you’re lucky, you can find a link to the article you’re looking for. This of course assumes you have spelled Barack with one ‘r’ and not the two I initially thought.

Yes we should make it easy to find the most likely topic but not to the point where it makes finding the less likely, difficult.

As far as naming stability is concerned, moving Barack Obama to Barack Obama (44th president of the United States) is surely more stable in the long run. There may be a more notable Barack Obama in the future or even another US president by that name, but there will never be another Barack Obama (44th president of the United States). I appreciate that Barack Obama is an uncommon name in the English speaking world and so this is unlikely but what about another George Washington or Richard Nixon. There are already two ex US presidents called John Adams.

My main point is however, if we want to do away with Primary topic, it is perfectly possible to do so. And all the time saved from writing and rewriting (again and again) the guidelines, and bickering about which article should occupy that slot, can be spent actually improving the encyclopaedia’s content. Ykraps (talk) 15:59, 9 April 2013 (UTC)

We then have the problem of incoming links. Thousands of links have already been made to Barack Obama, with the expectation that clicking on that link will take the reader to the article on the 44th President, not some menu of other possibilities. The same can be said of John Adams, Thomas Jefferson, Jesus, Europe, Flowers, Sheep, Bananas, China, Pork, and so on. If we do away with a primary topic, all of those becomes disambiguation pages because of their many possible meanings. Will you be the one to fix the millions of links that are created, so that readers can be taken to the expected articles that now sit at the primary topic titles for these pages? bd2412 T 16:45, 9 April 2013 (UTC)

Having Barack Obama go to Barack Obama only makes it easy for people who are looking for the 44th president of the United States.

(an overwhelming majority)

If you are looking for another Barack Obama, then it’s actually quite difficult.

No, it isn't. Simply visit the Barack Obama page. The only other Wikipedia article about a person by that name is linked at the top.

Under this setup, most users arriving at the Barack Obama page have reached the desired article, while the rest must follow a link. If a disambiguation page were to occupy the "Barack Obama" title, everyone arriving there would need to follow a link to the desired article. How would this be beneficial?

Entering Barack Obama (disambiguation), a perfectly logical thing to enter if you are aware of Primary topic, takes you to a page that doesn’t exist, although from here you can find a link to Obama (surname); I don’t know why, to me that is a step backwards, I’ve already made it clear that I’m looking for a Barack Obama.

It's a keyword search.

If you wish to create Barack Obama (disambiguation) as a redirect to Barack Obama, I don't object.

As far as naming stability is concerned, moving Barack Obama to Barack Obama (44th president of the United States) is surely more stable in the long run. There may be a more notable Barack Obama in the future or even another US president by that name, but there will never be another Barack Obama (44th president of the United States).

That logic applies to topics in general, irrespective of whether their names are ambiguous. Do you advocate that we preemptively disambiguate every article's title (just in case)?

My main point is however, if we want to do away with Primary topic, it is perfectly possible to do so.

Has someone asserted otherwise? —David Levy 16:47, 9 April 2013 (UTC)

David, yes an overwhelming majority probably are searching for the US president and I refer you to my earlier comment, “Yes we should make it easy to find the most likely topic but not to the point where it makes finding the less likely, difficult”. The fact that PT exists prompts me to search for a disambiguation. Why would I search for something I’m not looking for? I appreciate your simple solution but I really needed that information before I started my search. Also I erroneously (apparently) assumed it was a keywords (plural) search and I considered ‘Barack’ to be a key word.

I am advocating that titles should be precise enough to unambiguously define the topical scope of the article (as per wp:precise), yes.--Ykraps (talk) 19:52, 9 April 2013 (UTC)

I wasn't accusing you of searching for something that you weren't looking for. I was explaining your approach's inconsistency with our actual practices and describing a simpler means of accomplishing the task at hand.

As noted at WP:TWODABS (part of the same guideline whose existence prompted you to seek a disambiguation page), "if there are only two topics to which a given title might refer, and one is the primary topic, then a disambiguation page is not needed—it is sufficient to use a hatnote on the primary topic article, pointing to the other article. (This means that readers looking for the second topic are spared the extra navigational step of going through the disambiguation page.)" —David Levy 21:36, 9 April 2013 (UTC)

Suppose we were to move Barack Obama to Barrack Obama (44th President of the United States); but suppose that some tremendous calamity caused the United States to dissolve, but a century or two on, a group of South American countries decided to form a national union, and also called this new country, "the United States of America". Suppose, further, then two centuries later, this nation were to elect its 44th President, and his name happened to be "Barack Obama". Now we have two people who could be identified as Barrack Obama (44th President of the United States). Do we need to preemptively disambiguate the current one to Barrack Obama (44th President of the United States, North America)? How about Barrack Obama (44th President of the United States, born 1961)? Either name could still yield a scenario where the title is ambiguous - if for example the new United States were formed in North America instead of South America; or if some future order of man were to initiate a new date for counting years, so that thousands of years in the future there is a new 1961, and a new Barack Obama is born who is destined to become the 44th President of a new United States. In fact, I propose that we initiate a new system of counting years right now: Before Wikipedia and After Wikipedia, with it currently being the year 12 AW. Then we can move Barack Obama to Barrack Obama (44th President of the United States, born 40 BW). bd2412 T 17:26, 9 April 2013 (UTC)

Bd2412, I appreciate what you say about incoming links, a problem indeed. With regards to your second concern, I think we can safely leave that for our ancestors to worry about; or are you just using this opportunity to show everyone what a great wit you are?--Ykraps (talk) 19:52, 9 April 2013 (UTC)

I assume that you mean "descendants". Indeed, we can safely allow such hypothetical concerns to be addressed later. The same is true of those that might arise within our lifetime. This is a wiki. We base it upon the world as it exists today (including knowledge of what's to come), and if something changes tomorrow, we edit accordingly. —David Levy 21:36, 9 April 2013 (UTC)

You are quite correct, I did indeed mean descendants. Thank you.--Ykraps (talk) 22:34, 9 April 2013 (UTC)

An alternate wording

A topic is the primary topic if this topic is clearly more important to the Wikipedia project than any of the other topics with the same title. If no topic is clearly more important, it is better for the title to be a disambiguation page. This would replace the current guidance, not add to it. Perhaps the importance ratings assigned by Wikiprojects could help in this determinations - a top-importance article is usually primary topic, I'd assume. I'm not honestly sure if this is any better, though. Thoughts? Ego White Tray (talk) 01:42, 9 April 2013 (UTC)

I think that dictionary words should be assumed to be very important to the project, alongside the importance of actual articles.

I’d prefer that we err on the side of precision over conciseness, up to 50 characters, for sake of any reasonable but unusual reader being able to guess the subject before downloading the article. --SmokeyJoe (talk) 01:59, 9 April 2013 (UTC)

I think we need PRIMARYTOPIC, but that it's applied way more often than it should be. This rewrite doesn't seem to fix that. It doesn't even make sense, since no other articles have the same title. Think about Big or Brand New. Many will claim that the film and the band are the most important topics to WP, since there's not much else competing with them. These terms are too ambiguous to not be disambig pages. How can we rewrite PRIMARYTOPIC to not tempt editors to make so many primarytopic claims based on simple popularity when the title is so ambiguous? Dicklyon (talk) 02:08, 9 April 2013 (UTC)

I said no other topics, have the same title, not no other articles have the same title. Example: a planet, element and a god are three topics that have the same title, Mercury, but the articles don't have the same title. See the difference? Ego White Tray (talk) 02:43, 9 April 2013 (UTC)

I strongly oppose this idea, which would neither serve readers nor simplify the process.

Our goal is to make it as easy as possible for readers to find the articles desired, not to funnel them to articles about subjects that we've deemed "more important". And such a concept would fuel endless debates among editors (and the aforementioned WikiProjects) regarding which topics are "more important to the Wikipedia project" (itself poorly worded, as it could be interpreted to favor topics related to Wikipedia or those in which Wikipedians are personally interested). —David Levy 03:52, 9 April 2013 (UTC)

I agree with the statements here from User:Dicklyon and User:David Levy. We should not funnel readers to the most popular pages (google already does that). --SmokeyJoe (talk) 05:19, 9 April 2013 (UTC)

I think I also agree. I have this fear that I'm missing something if the solution could be so simple. Importance is clearly problematic in the application here since it is totally subjective. Vegaswikian (talk) 06:11, 9 April 2013 (UTC)

I think objective guidelines can be reached by consensus. From an encyclopedic standpoint, scientific significance should predominate over commerce and pop culture. Taking Mercury for an example, I would contend that the element, being a building block of the universe, is the most significant use for an encyclopedia, followed by the planet, and then the god for its historic value, and then the space program, and then the car (the record company doesn't enter into it, because it's name is Mercury Records). bd2412 T 12:15, 9 April 2013 (UTC)

Again, our goal is to make it as easy as possible for readers to access the articles that they wish to read.

Persons typing "Steve Reeves" probably seek information about the bodybuilder/actor. It would be unhelpful to send them to the article about the computer scientist (on the basis that "scientific significance should predominate over commerce and pop culture"). —David Levy 13:05, 9 April 2013 (UTC)

I was addressing the significance of concepts, not people. The comparable discussion for people would probably be Madonna, for whom there is a person, and a much older usage in history and art. I suppose if Freddie Mercury had gone by his adopted surname alone, then we would have to fit Mercury the musician in with the element and the planet, and a specific person would go near the bottom of that group. bd2412 T 15:09, 9 April 2013 (UTC)

Distinctions like that are arbitrary. There is no clean way to draw the line. It's much simpler and less contentious to simply look at which topic associated with a given term is most likely to be sought - and if one stands out, it's the primary topic for that term. Period. After all, if the significance is really there for a given topic, it will be reflected in its likelihood of being sought. --B2C 15:23, 9 April 2013 (UTC)

But that is also arbitrary. What readers are likely to be looking for depends on so many things, and can vary for the same reader over time. To imply that we have a way to determine what that is for most readers is a delusion. Vegaswikian (talk) 16:58, 9 April 2013 (UTC)

Measurements do not have to be (and cannot be) 100% precise before they can be useful. Using available measures (reductions in uncertainty) is not arbitrary. -- JHunterJ (talk) 17:15, 9 April 2013 (UTC)

Importance is a very subjective and inherently contentious attribute. This is not the intended sense of "primary" in "primary topic", neither literally nor historically in practice. Flawed as it is, likelihood of being the one that is sought is a much more objective attribute. --B2C 15:23, 9 April 2013 (UTC)

Indeed. But that is not the same thing as what most people would expect the article to be about. So if, for example, a manufactured band on a US TV show is given a name which already exists in some other context, their article will probably be the target of most searches for that name, but everyone else will be left confused and even frustrated to get there. There might be no reason why they would even have heard of the band. Whether that is the right solution for WP is a subjective judgement. --AJHingston (talk) 15:47, 9 April 2013 (UTC)

If the article about the manufactured band is clearly getting much more attention, then it should be okay to confuse and even frustrate "everyone else", which will be comprised of a relatively small minority, by definition. Besides, a hatnote link, not to mention informative Google results, should really be enough to limit the confusion and frustration to a minimum.

If the manufactured band is not clearly getting much more attention, then putting the dab page at the plain name is what WP:D indicates.

This is pretty much how it used to work up to a couple of years ago, and everything was much less contentious then. The watering down of primary topic by the addition of the long-term significance consideration has created a lot of ambiguity and contention. --B2C 19:33, 9 April 2013 (UTC)

But the US teenagers in this hypothetical case are not representative of anyone else. The same is true of any other special interest group. My point about objectivity v subjectivity is that the decision whether to stay 'in universe' (to pick on a previous discussion), to give precedence to US users because they are the largest single group (as has been argued in other WP contexts), or to judge the primary topic purely on a statistical analysis of visits over a period or anything else is a subjective decision, and it is only after that objective criteria come into play. That is why we keep going around this - we all begin with different values. --AJHingston (talk) 20:41, 9 April 2013 (UTC)

I have to agree with AJHingston, who seems to me to be saying in essence that Wikipedia is an encyclopedia, not a fansite where articles achieve primary status by the number of likes. What is the evidence for thinking that seekers of articles on pop stars and TV shows are stymied by the mounds of Elizabethans and philosophical terms and astral bodies in their search path? Cynwolfe (talk) 23:26, 9 April 2013 (UTC)

Alternatively, what is the evidence for thinking that seekers of articles on Elizabethans and philosophical terms and astral bodies are stymied by the mounds of pop stars and TV shows in their search path? -- ArglebargleIV (talk) 23:44, 9 April 2013 (UTC)

But that's the point: when in doubt, disambiguate, and stop battling over which is the primary topic. Cynwolfe (talk) 00:46, 10 April 2013 (UTC)

And how do we deal with the editors that see disambiguation as bad and that we need to have a primary topic? I believe that I agree with you. However in practice consensus does not support this alternative. Vegaswikian (talk) 01:56, 10 April 2013 (UTC)

Disambiguation is not inherently bad, but erroneous links to disambiguation pages are. Right now, we have over 337,000 such links needing to be fixed, and implementation of this proposal would bump this number back up into the millions. bd2412 T 11:08, 10 April 2013 (UTC)

And even more inherently bad are the unknown (vast) number of articles with titles like Foo (something) with no hatnote or dab page entry to link them from "Foo"! But that's another saga. PamD 11:22, 10 April 2013 (UTC)

@BD2412, 337,000 is a surprisingly small number considering the size of Wikipedia. It speaks to the effectiveness of the volunteers who patrol the pages listed on the reports. While I don't support the proposed alternative wording, I would support proposals to raise the bar for for primary topics and have a stronger default to disambiguation where there are substantive questions whether there is a primary topic. As it is, a page with 51% more traffic than all the other ambiguous pages combined can be a primary topic. Than means there could be a very large number of links mistakenly going to the primary topic that go unnoticed by the links to disambiguation page reports and only get fixed on the off chance that someone following the link feels inclined to go back to the page they left and edit the link. I'd say underreporting of incorrect links to ambiguously titled pages is a problem and that including more ambiguously titled pages to be patrolled is a good thing. older ≠ wiser 13:14, 10 April 2013 (UTC)

We have 337,000 today (instead of the 900,000+ we had three years ago) in large part because a handful of us have made hundreds of thousands of fixes. Were we to let up, the number would quickly shoot into the millions. Not a day goes by without some page being turned into a disambig with hundreds (occasionally thousands) of incoming links that go ignored by the editor who made the change. With respect to errant primary topic links, I fix these from time to time also. Apple and Mouse are usual suspects. I would propose that rather than make those titles disambiguation pages, we pipe links to them through unambiguous redirects like Apple (fruit) and Mouse (animal). That way, we can easily see which incoming links are unpiped, and can check those, and fix them if they are wrong. However, because most incoming links would be pointing to the right target, this would be a low priority compared to fixing links pointing to actual disambiguation pages, for which the incoming link is always wrong unless it is piped through a "foo (disambiguation)" redirect to the disambig. In short, I propose a third category. Pages like Barack Obama sit at their clear primary topic title and links go straight to the page with the presumption that they are intended to go there. Pages like Mouse sit at their primary topic title with links piped through a redirect for occasional checking. Pages like Bass disambiguate. Incidentally, there are a great many pages that should not be disambiguation pages at all because they are really just lists of variations of a subject (for example, this months most-linked-to list includes Bionic Commando, British poetry, Commoners in the United Kingdom, Law of the sea, Leaving Certificate, and Peace process, all of which should be general topic articles). bd2412 T 13:46, 10 April 2013 (UTC)

And as someone who has spent time going though that backlog, the links are indicative of the problem. A better place to review this is Category:Articles with links needing disambiguation. These are the articles that have links that other editors can not figure out what the target is. This is an indication of the problem we currently have when an article is in the name space and the title is ambiguous. These are links that from an article and other editors can not figure out what the correct target is. There is no simple way to find those inbound links when they go directly to an article. With a dab page at least those "bad" links are easy to find and can be corrected. So which is better, bad links that you can not find or links that you know need to be fixed? Vegaswikian (talk) 20:52, 10 April 2013 (UTC)

I believe that my proposition would address this problem. bd2412 T 02:09, 11 April 2013 (UTC)

Remove "long-term significance"

Many editors have suggested that the "long term significance" clause is problematic. This is a narrower issue that we might get an agreement on. Should it be removed? Ego White Tray (talk) 00:32, 10 April 2013 (UTC)

No. Long term significance is more important than usage. Long term significance is the most redeeming part of PRIMARYTOPIC, and the only part properly consistent with WP:COMMONNAME, which says that (contrary to the common interpretation of "common") we should use the name that is prevalent in reliable English-language sources. The usage clause invites examination of ghits and page visits and is inconsistent with reflecting the best sources. --SmokeyJoe (talk) 00:51, 10 April 2013 (UTC)

The interpretation of "commonly used" to mean "commonly used in reliable sources" is much better established (thanks primarily to PBS (talk · contribs)) than including "long term significance" as a direct consideration when evaluating primary topic (a relatively recent addition). Indeed, long term significance is inherently accounted for when looking at usage in reliable sources. Calling it out as a separate direct consideration is what the problem is, because it's a much more subjective consideration than looking at usage in reliable sources. Since looking at usage in reliable sources already accounts for "long term significance", giving it separate direct consideration is at best redundant, and often just provides a source of contention. --B2C 19:11, 10 April 2013 (UTC)

I believe that "enduring notability" is a valid consideration, but we're greatly overstating its role and muddying the waters with the highly subjective "educational value" language.

Rather than maintaining a separate "long-term significance" criterion (which seems to have triggered a great deal of conflict), we should qualify the longstanding "usage" criterion by explaining that it sometimes is reasonable to consider historical usage. We shouldn't necessarily focus solely on what usage predominates right now (which might prove ephemeral), but a notable subject's existence over a long period isn't a standalone factor to be weighed against common usage. (And the reference to "educational value" is largely useless.) —David Levy 01:00, 10 April 2013 (UTC)

I think this at best just slightly obscures the problem; it doesn't really address it. But your suggestion has inspired me to make the proposal below! --B2C 19:32, 10 April 2013 (UTC)

It would be better to make an AND instead of the OR. To declare a primary topic, a topic both be much more popular and have more long-term significance. Whenever these two points are in conflict, use a disambig page. The other problem is the phrase "more likely than all the other topics combined", which should be "much more likely than all the other topics combined". The way it is now, it has to be much more signifcant than any other single topic, but a bare majority over a wide ambiguity. Makes no sense. Dicklyon (talk) 03:47, 11 April 2013 (UTC)

It makes perfect sense as is. Given four uses, A,B,C and D, for A to be the PT, it needs be much more likely to be the one being sought than each of the others, and merely more likely than all 3 combined.

So if the likelihood of B C D is 20% each, then A at 40% is not the PT because 40% is not more likely than 60%. But if B C and D are 15% each, and A is 55% then it is the PT, because 55% is much more than 15%, and more than 45%, the combined likelihood of B C and D. --B2C 05:19, 11 April 2013 (UTC)

Also, if A is 55%, B is 35% and C and D are 5% each, A is not the PT because it is not MUCH more likely than B, though it is more likely than the other three combined (45% like in the previous example). --B2C 05:28, 11 April 2013 (UTC)

And if you had 55%, 5%, 5%, 5%, 5%, 5%, 5%, 5%, 5%, 5%, you'd make the first primary, even though it's a bare majority of accesses. Makes no sense (to me) to declare a primary topic for something that's the target in less than about 90% of the accesses. Even 90% is marginal, when with a disambig page we can more easily show people the landscape of what the term refers to and let them find the possibly several relevant articles that they may be looking for. Dicklyon (talk) 05:49, 11 April 2013 (UTC)

With a disambig page we can force them to look at the landscape first, yes. With a primary topic, we can more easily allow most of the readers to efficiently see the topic they sought, and use a hatnote to allow the smaller group of readers to see the landscape and let them find the relevant article they were looking for. -- JHunterJ (talk) 11:19, 11 April 2013 (UTC)

Agreed. That's why we should only use a primary topic when we are absolutely certain that the article in it is what most of the readers are looking for, even if they come from different backgrounds (i.e. without biasing it to just one single group of readers only because it's big - the biggest group may be favored, but only if it's heterogeneous). This means that the choice of primary topics should be made stricter than they currently are. Diego (talk) 12:12, 11 April 2013 (UTC)

That's sounds like "disagreed". We should use a primary topic when it serves the majority of a single group (the group is the readership as a whole). Trying to distinguish those majorities that count (different "backgrounds", whatever those are) from those majorities that don't count (for some reason a "single" group, even though the majority from different backgrounds is also a single group) would be a good cover for "primary topics when I'm in the majority, because obviously my majority is diverse, but no primary when I'm in the minority, because obviously that majority is just a single group". -- JHunterJ (talk) 12:54, 11 April 2013 (UTC)

I agree that primary topics give more exposure to the topic and that this is sometimes a good thing. But distinguishing which topics are likely to be sought shouldn't be done only from raw numbers of readers without any concern of who they are - what's missing in the current PRIMARY TOPIC is a recognition that people come to Wikipedia from very different perspectives, therefore only worldwide subjects are likely to be sought by any reader picked at random in the whole world. I like B2C's idea to give emphasis to reliable sources, it's a step in the right direction. I'm not suggesting to distinguish between "groups that count" and "groups that don't" - when there's a reasonable claim that two different groups would be best served by different articles, no primary topic should be used. I'm explicitly thinking of Madonna here, where the number of visits has been repeatedly rejected as the only criterion to be considered even if it's clearly overwhelming. Diego (talk) 14:00, 11 April 2013 (UTC)

I agree that it would be an improvement to require that a topic meet both criteria. Most of the debates seem to occur when the criteria each lead to different outcomes, as the guideline offers no real help on what to do in that case. The best option for the articles in almost all such cases is to have no primary topic, which settles the matter. It would also be an improvement to have primary topics much more likely than any other topic (i.e. around 90% of the usage, rather than 51%). I look at it as a gamble: does anyone need to check links to this title to make sure they will take readers to the right place? If 49% of the incoming links are actually meant for another article, we should be disambiguating the incoming links.--Trystan (talk) 13:50, 11 April 2013 (UTC)

Dicklyon, right. What this is really about is margin of error. In the case of A is 55%, and the others are 5% each, clearly there is no term that people are searching for more than topic A, and it's apparently more than half, so let's take them there. Why take everyone to the dab page?

In the case of A=55, B=45, C=5, D=5, the thinking is that it's not even all that clear that more people are looking for A than B (because of margin of error), so in that case we do go with the dab page. --B2C 16:50, 11 April 2013 (UTC)

So it is OK that 49% percent of people could be directed to the "wrong" page ... and that is acceptable. In what other field would a 49% failure rate be acceptable? older ≠ wiser 18:05, 11 April 2013 (UTC)

If 51% of people are looking for topic "A", and we instead direct them to a disambiguation page, then 51% have been taken to the "wrong" page. Is a 51% failure rate better than a 49% failure rate? Furthermore, is it a "failure" if instead of going to disambiguation page "Foo" that basically says, "you were probably looking for Foo A, but you may have been looking for Foo B, and by the way there is also Foo C, Foo D, and Foo E", the reader is taken directly to page Foo A, that has a note right at the top saying "You may have been looking for Foo B; for other meanings see Foo (disambiguation)"? For the majority of people who are looking for Foo A, there they are. For the large minority looking for Foo B, the link is right there, so they are no more inconvenienced than if they had been taken to a disambiguation page in the first place (one extra click either way); only the very small number looking for Foo C and beyond must face the relatively minor inconvenience of clicking the link to the disambiguation page. For an excellent example of this principle in practice, see Colorado. bd2412 T 21:40, 11 April 2013 (UTC)

Except that arriving to a disambiguation page is not nearly as problematic as arriving to the wrong article. There's the well known effect that navigating to an unexpected page will cause a disorientation that takes several seconds to recover from. A layout where all readers arrive to the DAB is more fair - everybody sees quickly that the page is a directory of links, and all need exactly one click to find their desired article; while when there's a primary topic, the worst case is that will need two clicks to find it, the first one of which is through the cumbersome format of a hatnote (you can visit WT:HATNOTE regularly to learn about the well-known problems it causes). The disambiguation page is inherently more recognizable, and easier to scan or read. Preferring the problems that the primary topic causes over a DAB is only justifiable when its prevalence is overwhelming. Diego (talk) 22:03, 11 April 2013 (UTC)

I don't think there is any consensus on the assertion that "arriving to a disambiguation page is not nearly as problematic as arriving to the wrong article", or that arriving on the wrong article causes disorientation. I, for one, have never experienced it. But this too is even indirectly addressed by primary topic. The idea is that it will only happen when the "right page" is a relatively obscure use of the term in question. For example, if someone looking for the relatively obscure "Paris" in Texas arrives at the article about the city in France, is that really a problem? At the very top it clearly states: This article is about the capital of France. For other uses, see Paris (disambiguation).

I don't see how it's a problem at all. --B2C 22:27, 11 April 2013 (UTC)

B2C, you've gone straight to the other end of the spectrum where it is likely that far far more than 51% of people will be expecting the city in France to be at Paris. For pages where the majority is much less certain, I agree with Diego that there is a disorientation involved in arriving at the "wrong" article rather than a disambiguation page. A disambiguation page at the base name is effectively signalling that the term is ambiguous and we (the editorial Illuminati) are not making any presumptions about what the reader might have been looking for. I'd like to see a rather higher bar for a primary topic than 51% and have a stronger default to disambiguation if there is no obvious primary topic. older ≠ wiser 23:29, 11 April 2013 (UTC)

I too would like to see a rather higher bar for a primary topic. I would also like it to be source-based, not pageview or ghit based or percent internal searches. A criterion like “whether a broad-audience international newspapers disambiguates the name when using it” could be good to include in the list of suggestions. A newspaper would not explain that the dog mentioned was an animal, but would indicate that the dog tool is a type of tool. Newspapers indicate whether mercury is an element, planet or god. Newspapers do not usually specify that Barack Obama, or Bush, or Bill Clinton is/was a US President, but I do think they do remind the reader who John Adams was (especially to an international audience). Therefore, Dog, Barack Obama, Bill Clinton are not disambiguated by us, but Mercury and and John Adams should be. This criterion would align the likelihood of bewilderment at arriving at the wrong article with the likelihood of bewilderment upon reading the newspaper. NB. This criteria is not catch-all. There is another reason that we disambiguate at George Bush, and that reason is not pageviews. --SmokeyJoe (talk) 01:31, 12 April 2013 (UTC)

I'm pretty much in agreement. My big concern is that when you have a topic that the news sources fail to disambiguate at all and use it for several uses, most notability raising a US mailing address to the equivalence of a place. One could argue that is is in fact a reason to have the dab page at the main name space. Vegaswikian (talk) 02:05, 12 April 2013 (UTC)

SmokeyJoe, the purpose of disambiguation in WP is disambiguation with other uses of the same name on WP, not to make the title recognizable to a broad audience. So a criterion like “whether a broad-audience international newspapers disambiguates the name when using it” is usually not relevant for primary topic determination.

Articles about barely notable topics are routinely placed at their generally unrecognizable names, when the name is unique or considered the primary topic. Countless examples can be found with SPECIAL:RANDOM (Southampton Street, APEN Agiou Athanasiou, Kottan ermittelt, Myeongbong Station). Thousands and thousands of titles not recognizable to a broad audience exist on WP. They're probably even in the majority of all articles. Is that a problem looking for a solution? I don't think so. So, generally, I don't see why whether a broad-audience international newspaper disambiguates the name when using it should be a consideration for determining primary topic, or any other reason.

However, that said, there is one exception. If one of the uses is generally not disambiguated by such newspapers, but all the others are, that's very strong evidence of that one use being the primary topic (even it's a fictional character and all the others are real person biographies). But, if all of the uses are disambiguated by newspapers, that doesn't mean there isn't a primary topic among them. --B2C 06:14, 12 April 2013 (UTC)

WP doesn’t make things “recognizable to a broad audience”, no, I’m saying we should use evidence of” recognizable to a broad audience” in disambiguation decisions. I think you don’t like this idea because you want to push/support simple titles for the most popular pages. Is that wrong/unfair? Most articles don’t face issues of disambiguation, and most subjects are not broadly recognisable. There’s no issue there. The issue here is with popular, short titles. If our sources disambiguate a topic title, so should we. We should always be guided by our sources. --SmokeyJoe (talk) 08:33, 12 April 2013 (UTC)

As noted earlier, a rather sizable gap exists between Wikipedia's actual consensus-backed article titling conventions and your personal preferences. This particular discussion's goal is to accurately and clearly describe the former, so I think that it would be more constructive to narrow the focus accordingly. I realize that you wish to effect change (regarding our actual practices, not merely our documentation thereof), but that's best discussed separately. —David Levy 09:55, 12 April 2013 (UTC)

Consensus can change, but I'm not even sure there ever has been consensus for how to deal with primary topics. I'm pretty sure there's no consensus right now for it, and the actual practices are diverse; so if we're going to craft a wording that will affect future practice, it better get feedback from as many editors' opinions as possible. I actually find “whether a broad-audience international newspapers disambiguates the name when using it” a really good criterion to add to the current batch of things we look for in topics with ambiguous titles; it could be a way to settle the current problems we're experiencing in consensus-building at RMs, although it probably won't be definitive. Diego (talk) 10:59, 12 April 2013 (UTC)

I find the idea confusing, but that's beside the point. Consensus certainly can change, and there's nothing wrong with proposing that we modify our conventions. But it's counterproductive to conflate such discussion with that of our current practices' documentation. —David Levy 11:34, 12 April 2013 (UTC)

David: errrrrr ... what???? Tony (talk) 12:41, 12 April 2013 (UTC)

I don't know what you're asking. —David Levy 13:02, 12 April 2013 (UTC)

If we're not going to get rid of the whole PT thing, then I think looking at the Wikipedia:Disambiguation#Determining_a_primary_topic guide is pretty sensible. Currently, it tells people to look at incoming links, page views and google. I think it very odd that it says nothing about sources. --SmokeyJoe (talk) 12:15, 12 April 2013 (UTC)

I think that we're again blurring the lines between WP:PRIMARYTOPIC and WP:COMMONNAME.

As noted above, I find this idea confusing. What, in your view, constitutes "disambiguation" on the part of reliable sources? Does their use of the phrase "President Barack Obama" or "US President Barack Obama" qualify? Are we to determine whether a description is included for the purpose of distinguishing the subject from others with the same or similar name? If so, how? —David Levy 12:35, 12 April 2013 (UTC)

That one's easy. If sources directed to a general audience always use the term "President Barack Obama" or "Barack Obama (the president, not his father)", that would be a telling sign that the name needs disambiguation (in there are several articles in Wikipedia using it). On the other hand, if sources for a general audience commonly refer to some "Barack Obama" without needing to distinguish him from other people, and expecting their readers to know who's the one they're referring to, that's strong evidence for the term to be used with a primary topic. Diego (talk) 13:41, 12 April 2013 (UTC)

It's normal to describe entities when writing/speaking of them, irrespective of whether they're well known or better known than other subjects with the same or similar names.

I would expect virtually all non-specialist publications covering the Rolling Stones to include a descriptive word or phrase ("band", "rock band", "musical group", "rockers", "recording artists", etc.), just as we do in our article's lead. This is standard writing/reporting. It doesn't mean that the sources seek to prevent confusion with the novel The Rolling Stones.

Obviously, that example is clear-cut. Other instances won't be. The above proposal is based upon an unreliable inference. —David Levy 14:58, 12 April 2013 (UTC)

And what's better: if sources directed to a general audience from different parts of the world use the same term without disambiguation for different topics, that would be almost definitive evidence that it can't be used with a primary topic. Diego (talk) 14:04, 12 April 2013 (UTC)

That makes sense to me, but it doesn't really deviate from our current practices or documentation thereof; if reliable, non-specialist, English-language sources frequently refer to different subjects by a particular name, a primary topic probably doesn't exist for that term, regardless of where these publications originate and whether descriptions are provided. —David Levy 14:58, 12 April 2013 (UTC)

Not necessarily. There are lots of instances where world-wide usage may be varied, but because of over-reliance on navel-gazing page traffic statistics one usage can be designated as primary. This is a type of systemic bias. older ≠ wiser 15:06, 12 April 2013 (UTC)

Can you cite some specific examples? —David Levy 15:13, 12 April 2013 (UTC)

These criteria being discussed, they would be suggestions to editors for points to inform the discussion, not factors feeding a formula, No? I wonder is David Levy is fearing a formulaic approach being pushed? --SmokeyJoe (talk) 01:51, 13 April 2013 (UTC)

As noted above, my main concern regarding your "If our sources disambiguate a topic title, so should we." proposal is that it seems to be is based upon the unreliable inference that such a description reflects a determination that confusion with similarly named subjects is likely. —David Levy 01:59, 13 April 2013 (UTC)

I don't see this as a major concern. I don't see the inference as necessarily unreliable, though odd things are always possible. I've mentioned major braod-audience newspapers as one type of source to look at. Probably, the sources cited in the article lede should be considered first. Why would you consider aberrant usage in these sources to be likely? Page view statistics can also produce aberrant results, yet the guideline advises their use without warnings. --SmokeyJoe (talk) 02:10, 13 April 2013 (UTC)

You appear to have misunderstood my concern. I'm not referring to aberrant usage. I'm saying that normal usage doesn't imply what you evidently believe it does. Please see my elaboration (including examples) above. —David Levy 02:18, 13 April 2013 (UTC)

I did read your posts and examples above. Your concerns read as something to consider in refinement of the actual wording inserted into the guideline so as to prevent these possible unintended readings. However, overall your concerns seem unfounded. Reference 1 of The Rolling Stones is a document titled "The Rolling Stones Biography", which is undisambiguated. Nowhere in the article is the least consideration of hillsides and round rocks. Everything I find on Dog (engineering) says something explicit about it being a tool, as if the reader might not have known. I think that everything you say is important, but not a problem to ensure it is avoided. On the other hand, your opposition to the mere mention of usage in reliable sources seems odd. I also don't understand you concern of blurring the lines between WP:PRIMARYTOPIC and WP:COMMONNAME. The two should be, could be, in complete harmony. Why should there be a dividing line? --SmokeyJoe (talk) 02:34, 13 April 2013 (UTC)

Your concerns read as something to consider in refinement of the actual wording inserted into the guideline so as to prevent these possible unintended readings.

I don't know what "unintended readings" you mean. You haven't answered my questions about what reading is intended.

However, overall your concerns seem unfounded.

They might be founded in an inaccurate understanding of what you've proposed. I'm not certain that we're on the same page.

Reference 1 of The Rolling Stones is a document titled "The Rolling Stones Biography", which is undisambiguated.

Firstly, the Rock and Roll Hall of Fame and Museum is a specialist source. Secondly, the Rolling Stones are described in the article as "a rock and roll band", "band" and "group". Perhaps that isn't the sort of "disambiguation" to which you intended to refer. It's precisely the type that I mentioned above, so it should be clear that it's what I had in mind.

Nowhere in the article is the least consideration of hillsides and round rocks.

Are you referring to explicit mentions of an alternative usage? Your following sentence seems to imply otherwise...

Everything I find on Dog (engineering) says something explicit about it being a tool, as if the reader might not have known.

This is what I mean. It's an inference on your part. You (correctly) regard that usage as much less common than another is, and this influences your perception of the sources' coverage. You see standard descriptions of the topic and assume that they constitute evidence that it's been deemed less familiar than another known by the same term. But you ignore comparable descriptions of the Rolling Stones — something explicit about them being a band. You cite the lack of "consideration of hillsides and round rocks", but you don't appear to demand that coverage of the tool type include explicit considerations of the mammalian subspecies referred to by the same name. You interpret the descriptions as evidence of what you already believe (and not evidence of what you don't believe).

I think that everything you say is important, but not a problem to ensure it is avoided.

I'll copy and paste a couple of my above questions:

Are we to determine whether a description is included for the purpose of distinguishing the subject from others with the same or similar name? If so, how?

On the other hand, your opposition to the mere mention of usage in reliable sources seems odd.

It's nonexistent. I find the assertion odd, given the fact that my suggested addition to WP:PRIMARYTOPIC is explicitly based upon "usage in reliable English-language sources".

I also don't understand you concern of blurring the lines between WP:PRIMARYTOPIC and WP:COMMONNAME. The two should be, could be, in complete harmony.

I believe that they are. Improvements are possible, but I see no conflict between the two.

Why should there be a dividing line?

They address distinct concepts. WP:COMMONNAME is about determining whether a subject is most commonly known as "x", "y" or "z". WP:PRIMARYTOPIC is about determining whether a topic referred to as "x" (irrespective of whether this is its most common name) is better known as "x" than other topics are.

For the type of tool discussed above, the common name apparently is "dog"; it isn't referred to as anything else with the same frequency (if at all). WP:COMMONNAME advises us to use that name instead of some other designation.

But for the term "dog", the primary topic (or whatever we're to call it) is Canis lupus familiaris, which is much better known as "dog" than any other subject is.

One determination has no bearing on the other, excepting the possibility that a subject has more than one common name, in which case the absence of a primary topic for one or more of these names might influence our article's title. (We prefer Cheque over Check (finance) and Apartment over Flat (domicile). The Vest article covers multiple meanings, whose individual articles reside at other common names.) —David Levy 09:55, 13 April 2013 (UTC)

The Rolling Stones (novel). None of the references that are immediately accessbile use the title in introduction. Instead, the title is introduced. For example, the external link reads: "First serialized in compressed form in Boys' Life Sep-Dec 1952 as Tramp Space Ship. This book makes reference to ". This is evidence against the title not needing disambiguation, vis a vis the other subjects. --SmokeyJoe (talk) 02:40, 13 April 2013 (UTC)

You're citing an entry from a database in which miscellaneous notes appear in that format. It doesn't imply what you perceive it to. I'm perplexed as to why you believe otherwise, given the fact that the entry begins with "Title: The Rolling Stones". —David Levy 09:55, 13 April 2013 (UTC)

You're persuading me that this is not as simple as I imaged. I appear to have blundered with the Rolling Stones. --SmokeyJoe (talk) 22:39, 13 April 2013 (UTC)

Make "long-term significance" implicit by clarifying reliable source qualification

I note that WP:COMMONNAME is clarified with the following important statement:

The most common name for a subject,[3] as determined by its prevalence in reliable English-language sources, is often used as a title because it is recognizable and natural.

I propose we do something similar for primary topic:

We remove the long-term significance consideration bullet from WP:PRIMARYTOPIC
We clarify that when looking at usage to determine primary topic, we're looking at "prevalence in reliable English-language sources".

This will eliminate the inherent conflict and contention created by having two explicit potentially contradictory criterion of unspecified priority, and yet will continue to properly account for the long-term significance consideration because it is implicit in reliable source usage.

Specifically:

Change this:

There is no single criterion for defining a primary topic. However, there are two major aspects that are commonly discussed in connection with primary topics:

A topic is primary for a term, with respect to usage, if it is highly likely—much more likely than any other topic, and more likely than all the other topics combined—to be the topic sought when a reader searches for that term.

A topic is primary for a term, with respect to long-term significance, if it has substantially greater enduring notability and educational value than any other topic associated with that term.

In many cases, a topic that is primary with respect to usage is also primary with respect to long-term significance. In many other cases, only one sense of primacy is relevant. In a few cases, there is some conflict between a topic of primary usage and one of primary long-term significance. In such a case, consensus determines which article, if either, is the primary topic.

To this:

There is one single criterion to consider when determining a primary topic:

A topic is primary for a term, with respect to usage in reliable English-language sources, if it is highly likely—much more likely than any other topic, and more likely than all the other topics combined—to be the topic sought when a reader searches for that term. The likelihood that a topic is the one being sought for a given term is determined by the prevalence of usage of that term to refer to that topic in reliable English-language sources.

--B2C 19:32, 10 April 2013 (UTC) slight wording change --B2C 20:49, 10 April 2013 (UTC)

I agree that the separate "long-term significance" criterion is ill-advised, but I don't see why we shouldn't address the concern that led to its addition by explicitly noting the relevance of historical usage. I understand your point that it's implied, but there's no harm in spelling it out. —David Levy 21:02, 10 April 2013 (UTC)

Yes, there is harm in spelling it out. Spelling it out creates an explicit separate consideration, one that can be evaluated only subjectively. It can suggest, based on one's biases, a different title than the one suggested by usage in reliable sources. Actual usage in reliable sources is a consideration that can be reasonably evaluated much more objectively than "long-term significance".

Two different criteria potentially suggesting different titles is precisely what leads to much of the disagreement and contention regarding primary topic determination.

If someone thinks that a certain topic is or is not primary because of long-term significance, they need to prove it by showing how it is or isn't referenced accordingly in reliable sources. Usage in reliable sources is ultimately all that matters, and, so, that's all that should matter, but by spelling out "long-term significance" as an explicit consideration, that suggests otherwise, creating potential contention. Totally unnecessary. --B2C 21:33, 10 April 2013 (UTC)

I'm not sure that you understand my position. I agree that we shouldn't have two different criteria or describe "long-term significance" as a separate consideration. I mean that there's no harm in expressly stating that a term's long-term usage by reliable English-language sources can be relevant. It might be implicit, but it's far from obvious. That's why editors felt the need to add the "long-term significance" criterion in the first place. It wasn't a baseless change; it was overcompensation. I view your proposal in a similar light. —David Levy 22:06, 10 April 2013 (UTC)

Please provide some specific wording, so others can better understand what you mean. --B2C 22:50, 10 April 2013 (UTC)

I'll give it a shot, but please note that the following wording isn't part of a formal proposal. As demonstrated above, it can be a bit tricky to prevent the inference that "long-term significance" is being presented as a separate/competing consideration, so brainstorming may be required.

Both historical and current usage in reliable English-language sources may be relevant, depending on the circumstances. When examining these trends, editors should avoid assigning disproportionate weight to past usage whose prevalence has declined or recent usage that may be ephemeral.

—David Levy 11:48, 11 April 2013 (UTC)

I like the direction here, but propose this instead:

Per WP:COMMONNAME, the preferred title for an article is the name most usually used in the most reputable and reliable English-language sources, subject to a number caveats further explained in that policy. If all these conditions are met, this title is called the “Commonname title”.
Where multiple articles appear to claim the same Commonnname title, the competitiveness of each claim should be judged by the prevalence of use in the majority of sources, of the non-disambiguated, non-contextualised use of the title in introducing the subject, the reputation and reliability of the sources, and the number of sources.

Where multiple articles competitively claim the same Commonname title, disambiguation is needed at the Commonname title.

P1. WP:PRIMARYTOPIC (a term I would prefer to substitute, and so am avoiding) is directly subordinate to WP:COMMONNAME. I would want to avoid tangential paraphrasing of WP:COMMONNAME regarding inaccuracy or non-neutrality in sources, reasons to not use a Commonname title for reasons other than ambiguity.

P2. A slight tightening of the WP:COMMONNAME criteria, specifically directed to ranking the strengths of claims of different subjects to the same title, calling directly and solely for examination of use in sources. Google hits, page views, incidental mentions in social media, academic worthiness, etc, are not suggested as useful criteria. I consider that under P2, Dog makes a competitive claim to Dog, while no other nearly does so due to sources not using the simple term “dog” without disambiguation from another dog and without contextualisation to a very specific subject. I consider that multiple subjects are competitive in making a commonname title claim for Mercury.

P3. A simple conservative statement strongly reflecting current consensus, but phrase in the reverse of the current WP:PRIMARYTOPIC. (I believe that a dictionary definition should diminish a commonname title claim, but that is a separate debate). --SmokeyJoe (talk) 23:36, 10 April 2013 (UTC)

Let's not conflate primary topic and common name. COMMONNAME is about title selection; PRIMARYTOPIC is about disambiguation.

Recall that the same topic may be primary for multiple names, only one of which (or even none of which) is the most common name for that topic. For example, Barack Obama is the COMMONNAME for that article, but the topic at that article is primary for 122 other terms.

An article may also not be at its most common name, because it's not the primary use of its common name, and so it's at a disambiguated title, but at the same time may be primary for any number of other terms. For example, Portland, Oregon is disambiguated (it's not primary for Portland), but it's also the primary topic for its nickname, Little Beirut.

Primary topic considerations have to be made equally, regardless of whether the term in question is also the topic's COMMONNAME. Equal consideration in primary topic consideration must also be given regardless of whether the term is a candidate for the article's title, or is just being considered as a redirect or dab page entry to the article. Much of SmokeyJoe's proposed wording just above does not reflect any of this. It is reflected in the current wording of primary topic, and in what I have proposed above. This must be retained. --B2C 00:36, 11 April 2013 (UTC)

I reject several of B2C's views, and note that his answer illustrates the clumsiness of "primary" to explain what we do. --SmokeyJoe (talk) 01:52, 11 April 2013 (UTC)

Views? What views? What clumsiness? Please explain. --B2C 04:57, 11 April 2013 (UTC)

If I may answer, first, there's your view that primarytopic is a relevant concept where there is no ambiguity, even though this WP term of art is defined only in the context of ambiguity, starting with "Although a word, name or phrase may refer to more than one topic, it is sometimes the case that one of these topics is the primary topic." Then there's your opinion that "Primary topic considerations have to be made equally." Maybe so, maybe not. I don't see a guideline or policy that says that, and maybe it makes more sense to be more flexible. But I don't really like Smokey Joe's proposal either, as the case that "multiple articles competitively claim the same Commonname title" is not the only situation where ambiguity exists and disambiguation is needed. For one thing, Commonname is about names; we also have ambiguity with common non-name words, frequently; like Run, which is Commonname for some things for generic in others. Dicklyon (talk) 06:01, 11 April 2013 (UTC)

Is the multiplicity of reasons for disambiguation the biggest problem? On that point, I agree with you. --SmokeyJoe (talk) 07:15, 11 April 2013 (UTC)

Dicklyon, differences about those views are largely moot, don't you think? For example, what does it matter whether we refer to the capital of CA as the primary topic of Sacramento, or not? Per your view, it is the PT only if there are other uses of "Sacramento", right? Per my view, we call it the PT regardless of whether there are other uses. Yet, whether we call it the PT of "Sacramento" or not, Sacramento is the title of, or redirects to, the article about that topic.

And I didn't say PT considerations have to made equally, period. I said they had to made equally without regard to common name. That comes directly from the definition of primary topic, which makes no reference to common name. So why should common name be given any consideration during primary topic determination? --B2C 16:40, 11 April 2013 (UTC)

To B2C:

Your apparent views I don't accept:

PRIMARYTOPIC is independent of COMMONNAME. (My view: Usually, Primary topic only comes into play after COMMONNAME has pointed two articles to the same title.)

PRIMARYTOPIC is well defined.

Sentences like "the same topic may be primary for multiple names" make sense. You think "primary" is a function of a name?

Your view that it is good to optimise titles for the most frequent users. (I think, where this is at any expense of another user, it is bad)

Your view that precision is best avoided in titles.

Clumsy is the word "primary" as used here, especially by you. In every normal use of the word, it is decisive and objectively defined (although the definitions may be debated). Primary sources, for example, are primary or they are not primary. There are not degrees. There may be qualifications, such as source A is primary for point X, but it either is or it is not. One source is not more primary than another. Whether this source is primary is completely independent of all other sources. In colours, a source is perfectly primary, or it is not. Primary means "first". It does not mean approximately 1.0. I suspect that I could understand you much better if you rephrase all of the above without using the word "primary". This DAB page has created an unhelpful jargon word. I've looked at the history, and the evolution of the use of the term at WP:D. It happened in steps, and has got worse since introduced in ~2008. --SmokeyJoe (talk) 10:29, 11 April 2013 (UTC)

I agree with Born2cycle's assessment of your proposed wording, but I also understand your concern regarding the use of the word "primary". If switching to a different term would reduce confusion/misunderstanding, we should do that. I hope that we can set aside tradition (something that Wikipedians often seem reluctant to do) and consider the possibility.

"Preponderant" is more precise, but that isn't a particularly familiar word. "Predominant" is a more common synonym, but it also can convey traits (power and authority) that we don't seek to reference. The same is true of "principal" (which can pertain to importance and value) and "chief" (which shares the unintended meanings of "predominant" and "principal"). I'm tempted to suggest "prevalent", but its definition ("widespread; of wide extent or occurrence; in general use or acceptance") seems a bit too broad. (We want to communicate that a usage is more widespread and of widest occurrence/acceptance.) —David Levy 11:48, 11 April 2013 (UTC)

A word like prevalent would improve the logic of the sentences. A simple substitution highlights some troublesome, artificial definitions. --SmokeyJoe (talk) 13:05, 11 April 2013 (UTC)

Does "prevalent topic" make sense? I suppose we're talking about the topic that is referred to most prevalently by a given term... does that make it the "prevalent topic"? --B2C 16:42, 11 April 2013 (UTC)

That would be my question to you and others who see useful meaning in the concept. I find it confusing, see holes in it, and may not be the best person to answer this. Prevalent is the best of a few words suggested, but I think it would be best to speak of a prevailing topic, and to speak of measures of prevalence relating, measure using sources. --SmokeyJoe (talk) 21:39, 11 April 2013 (UTC)

The concept is simply this: Just because there are several topics with a given name, does not mean that name cannot be the title or redirect to one of them. If one of them prevails in terms of usage of that term in reliable sources, then that name should redirect to, or be the title of, the article of the topic that so prevails.

Is that confusing? Is it not useful? What are the holes? --B2C 22:04, 11 April 2013 (UTC)

That is not confusing or with holes. If that were the language we could discuss differences more efficiently. I would like to discuss problems with the prescriptive "then that name should redirect to". But if we could first change the whole section to straightforward standard English, it would be good. --SmokeyJoe (talk) 22:25, 11 April 2013 (UTC)

We can work on the wording, but you suggested the concept was a problem. Are we good with the concept? --B2C 05:50, 12 April 2013 (UTC)

I have problems with the concept yes. But the wording problem is first and worse. The wording problem causes actual communication problems at RMs when editors usethe same terms/links with different interpretations. I'm open to the possibility that the concept problem will go away if the wording is fixed.

I've been looking at your proposal leading this subthread, and have been looking to reword it without "primary". --SmokeyJoe (talk) 08:25, 12 April 2013 (UTC)

After thinking about this, I believe your view here is very reasonable. Commonname determines what would be the desired name for an article solely based on the subject of the article and how it is used in reliable sources and in the wild. We also have cases where the common name is not even considered for an article since there are naming conventions that dictate the use of another name. Primary topic only comes into play if we have multiple articles with a claim, no matter how small, to a specific title. The fact that existing guidelines and policies may dictate a different name does not drop those names from consideration in a discussion of primary topic. Vegaswikian (talk) 23:25, 17 April 2013 (UTC)

I don't have a problem with having two criteria; determining PT is a complex decision and I don't think it can be simplified to a single criterion. (The guideline could be improved by providing some guidance on what to do when the criteria point to different articles.) I would have much more difficult interpreting a guideline saying "the likelihood that a topic is the one being sought for a given term is determined by the prevalence of usage of that term to refer to that topic in reliable English-language sources," because it strikes me as articifally defined. There are much better metrics of determining which topic is likely being sought, like article traffic and incoming links. Usage in reliable sources is also a good PT criterion, but it's separate from determining the likelihood of a topic being sought.--Trystan (talk) 13:47, 12 April 2013 (UTC)

Good point. I agree the definition should continue to be based on the concept of likelihood of being sought, and that prevalence of usage in RS is but one criterion that can be used to help determine that. --B2C 16:30, 12 April 2013 (UTC)

I'm thinking the question of whether PrimaryTopic, however described or named, will need an RfC to gauge community consensus on whether it considers (1) actual usage in sources, or (2) data on searches/page views, or some balance. My sense that that (2) is favoured here and (1) will be prefered by the community at large. --SmokeyJoe (talk) 01:45, 13 April 2013 (UTC)

It depends on the situation, but in all cases we should be using these various tools to determine the likelihood a given topic is the one being sought for a search of a particular term, compared to the other uses of that term on WP. In some cases looking at usage in RS may make more sense, in other cases link counts, in still others page views. But most of the time we should get the same answer from all these tools. If the tools don't all indicate that one topic is primary, without good explanation for any discrepancy, there probably isn't a primary topic for that term. --B2C 22:36, 16 April 2013 (UTC)

We should not be using likelihood of search, because in optimising access for the majority we neglect the needs of a minority. While we are a top ten website with massive traffic, this is not our primary objective. We shouldn't pander to teenage pop culture at the expense of dry academic subjects. Our primary objective (See Wikipedia:Prime objective if you like) is to provide access to all information for all people. Ignoring rarer uses is not conducive ensuring the breadth of this. Instead we should ensure all searches lead efficiently to the right page. --SmokeyJoe (talk) 01:01, 17 April 2013 (UTC)

I'm all for minority rights, but when the issue is whether a minority should be inconvenienced to benefit the majority, or vice versa, I fail to understand why we should favor inconveniencing the majority to benefit a minority. --B2C 14:58, 17 April 2013 (UTC)

Further, your objection to using likelihood of search seems to be predicated on the assumption that taking someone to the wrong article is substantially more problematic than taking them to a dab page, an assumption for which I know of no evidence which supports it, including no evidence that this notion is accepted by consensus. To the contrary, the entire notion of primary topic is based on the goal to minimize the number of clicks users looking for pages have to make. Therefore, if by far most of the people searching with a given term are seeking one particular topic, we should take everyone searching with that term to the article about that topic. From there, the minority looking for another topic is given directions on how to get to their desired destination. The alternative is to send no one searching with that term to the place they seek, by taking everyone to a dab page. That's not what we want to do.

And the hysterical hypothetical about pandering to teenage pop culture is often cited, without a single actual example of that ever occurring, so far as I know. --B2C 16:57, 17 April 2013 (UTC)

"Number of clicks" is not the only measure relevant to browsing and should be not be taken as such. The load size of the target page is what causes the major delay in navigation; for every user that is wrongly sent to the "primary topic" article, the user will have to wait until the useless article fully loads, then the DAB page, and only then the desired article; this is costly for mobile browsing and for users without broadband or at places with bad connections; while DAB pages are commonly pretty small and without images, so they should load quickly for all users. Therefore erring on showing a DAB page is inherently safer than sending people to the wrong article - in particular for articles that are considered primary topics by popularity alone, which are likely to be large and contain various images. Diego (talk) 22:14, 17 April 2013 (UTC)

Page load time was already a non-issue when these discussions first occurred over 10 years ago. Mobile access is made through the abridged mobile interface, so even if there was an issue there, it's already greatly mitigated by the mobile interface.

To start considering page load time as a factor in deciding how and when to disambiguate at this stage in WP's life is plain silly. --B2C 23:18, 17 April 2013 (UTC)

By the way, when I turn off WIFI and 3g on my phone, it takes 4 seconds to load (the relatively large) List of M*A*S*H characters#Frank Burns, and it also takes 4 seconds to load the (very small) Frank Burns (disambiguation) dab page. --B2C 23:26, 17 April 2013 (UTC)

WIFI does not mean there is no noticeable delay. Also the mobile interface seems to be more linear so that you need to scroll through several screens to get to what appears at the top of the article with a normal browser. So users don't know if the comment on the first line applies since they don't know any of the content until later. Vegaswikian (talk) 23:32, 17 April 2013 (UTC)

"Page load time was already a non-issue when these discussions first occurred over 10 years ago". Nonsense, I tell you from personal experience. In recent years, using dialup, 56kbs, loading a large page would take minutes, especially if also loading other pages, which is tempting after the first 10 seconds of wait. The preference option to not load the whole page when loading a diff was for me the most fantastic discovery. I routinely load the history and then the diffs to avoid delays. I've also used nomimal 100Mbs that ran at a trickle at peak times. 3G is also highly variable, but behaves much more in waits and spurts, meaning that any page may take time, but then the whole lot comes at once. I found dialup, although slow, the most reliable access, although I gave it up when browsers stopped allowing me to disable loading of images.

I think the majority should be inconvenienced to being sent to a dab page if otherwise it means sending others to a page, such midway down as a long character list, that could be bewildering. The character is an old closed topic. The senator's page is probably suffering from lack of exposure by being buried.

I consider the inconvenience of being sent to a dab page very very small. I actually like to be more aware of similar topics. In fact, as said before, sometimes the information on the dab page is everything I wanted.

"predicated on the assumption that taking someone to the wrong article is substantially more problematic than taking them to a dab page". That is right.

"the entire notion of primary topic is based on the goal to minimize the number of clicks users looking for pages have to make." I understand that. I think that potential for bewildering minority users should be listed as one concern, as a possible reason for disambiguating regardless of a popular PT. --SmokeyJoe (talk) 05:47, 18 April 2013 (UTC)

The preference option to not load the whole page when loading a diff was for me the most fantastic discovery I pray, tell me where do you activate that setting? Diego (talk) 06:07, 18 April 2013 (UTC)

Preferences, Misc, top check button. --SmokeyJoe (talk) 08:59, 18 April 2013 (UTC)

Thanks! It's been a time since I checked Preferences. This does improve checking changes to articles when on mobile. Back in topic, we can't assume that a majority of readers will connect under ideal conditions - the group of users behind problematic connections is significant, and it's likely to be people without a thorough digital culture, so every navigation help is important. Diego (talk) 13:35, 18 April 2013 (UTC)

Let me put it this way. Even 10 years ago page load time was not enough of an issue for it to be raised, discussed and considered, much less listed as a factor to weigh in determining primary topic. To introduce it now would mean reevaluating thousands if not millions of decisions, taking this factor now into account, for the first time. I fail to see why it should be considered now when it never was before.

Every now and then people express a preference for the type of serendipitous discoveries that can be made on dab pages, but consensus seems to consistently prioritize getting people to the page they seek with fewer clicks above that. I am not aware of any reason to believe consensus may have changed on this point. --B2C 15:31, 18 April 2013 (UTC)

I don't think there was ever a very strong consensus for primary topic based on the "number of clicks" rationale. I think the foundation was more along the lines of a principle of least surprise. I can't speak for Diego or others, but I've always thought a simple majority (51%) was far too low of a threshold. There are subjects with ambiguous names, such as "Paris", for which there would be near universal agreement that readers would not be surprised to end up at the article on the French capitol when entering "Paris" into the search box, or clicking on a link to Paris. There are many other ambiguously titled subjects where there might be a regional recognition of one sense as primary, but which would not be readily recognized elsewhere. There are no clear-cut objective criteria that I'm aware of for determining this, other than discussion and consensus (which I'm sure means you'll reject it out-of-hand as leading to JDLI reasoning). I'd favor adding language stating that if there is principled disagreement as to whether there is primary topic, we should default to disambiguation. The qualification of "principled" is intended to deprecate disagreement based solely on subjective notions. older ≠ wiser 16:21, 18 April 2013 (UTC)

That's right. When users expect the topic that is found under the base name, it's not a big deal; readers looking for a different topic will know to add extra parameter to their search. The problem exists when an article is shown for users that wouldn't expect it; if that happens with any likelihood, that article is by definition not a primary topic and shouldn't be there. Consensus is that the best outcome for each case is decided within the most relevant parameters for it, with no single criterion valid in all situations. I'd support clarifying this in the policy, as well as defaulting to DAB when no local consensus exist. Since policies are not prescriptive, a change in it doesn't imply a need to have a sistematic review of all current pages; only new discussions would be informed by the additional clarifications. Diego (talk) 16:52, 18 April 2013 (UTC)

The Potter Stewart solution

Here's a crazy idea:

Wikipedia editors have struggled to define what makes something a primary topic but we know it when we see it. In cases where there is doubt, this will need to be discussed at the relevant talk page.

Ego White Tray (talk) 22:26, 10 April 2013 (UTC)

Prefer "In cases where there is doubt, there is no 'primary topic'". --SmokeyJoe (talk) 23:38, 10 April 2013 (UTC).

And how will this play out? Shall we determine consensus for doubt? Or is a single doubter sufficient -- JHunterJ (talk) 23:41, 10 April 2013 (UTC)

Yes. It's a crazy improvement. I more prefer my above solution that points to ranking each claim by use in sources. --SmokeyJoe (talk) 23:52, 10 April 2013 (UTC)

I don't know about "doubt", but the current scheme of "In such a case, consensus determines which article, if either, is the primary topic" doesn't work well when there's no consensus. It should default to no primary topic. Dicklyon (talk) 06:04, 11 April 2013 (UTC)

I concur. There are two improvements we could made right now without changing the wording of the current criteria (nor needing us to get them exactly right): make "no primary topic" the default when there's no consensus, and make it clear that any criteria listed are not final - they're descriptive, not prescriptive, so the final decision should be influenced only by whatever the local discussion agrees to by rough consensus. An explicit reminder of WP:NOTBUREAUCRACY and WP:IAR will help alleviate current discussions, where editors (myself included) often try to use WP:PRIMARYTOPIC as law even if its wording doesn't have a strong consensus supporting it. It should also make it easier to achieve consensus at corner cases, where the current criteria are not helpful and some other local considerations would decide that case better. Diego (talk) 12:24, 11 April 2013 (UTC)

I agree. I see non regular RM participant apparently assuming a PrimaryTopic exists, and that editors have to decide on one. --SmokeyJoe (talk) 21:44, 11 April 2013 (UTC)

The problem with making "no consensus" default to no primary topic is often we don't have a good measure of that particular question. That is, usually we're talking about an RM proposal where some of the arguments for or against moving the article to another title may be made in terms of the topic being or not being primary for the current or proposed title. But most participants are often silent on the primary topic question.

For example, look at the current discussion at Talk:Frank_Burns_(disambiguation)#Requested move. The proposal itself is ostensibly based on WP:PRIMARYTOPIC, but few supporters weigh in one way or another on that particular issue. Even those few who mention "primary" mostly don't really argue either way. Dicklyon, for example, simply asserts, without basis, much less any reference to policy/guidelines, that a fictional character should not be "primary". Similar, George Ho just asserts the topic is not primary. Even bd2412, who allows that a fictional character may be a primary topic, asserts that he doesn't believe this topic is, without saying why. No mention of search likelihoods, or any of the tools mentioned at WP:PRIMARYTOPIC. It goes on and on. I suggest it would be fair to characterize most of their positions as, "I don't know or care if it's the PT or not, I just don't like that title for this article". On the Oppose side there are at least several who argue the topic is primary topic for the current title, and reference what WP:PRIMARYTOPIC says as support. So, what is there consensus on the PT question in this case? If you look at the weight of the arguments, you'd have to say it is the primary topic, since there is no sound argument based on what the relevant guidelines actually say that it isn't. But, at least for now, the supporters greatly outnumber the opposers. --B2C 23:56, 12 April 2013 (UTC)

"no consensus" default to no primary topic is often we don't have a good measure of that particular question. We need good measure of that question, no?

The Frank Burns discussion shows that this guideline is not in harmony with community consensus. --SmokeyJoe (talk) 01:40, 13 April 2013 (UTC)

No discussion with so few participants, including this one and the Frank Burns one, says anything about community consensus. And when the predominant sentiment in such a discussion is so blatantly contrary to what policy and guidelines have long said, particularly when those expressing this sentiments show little knowledge and understanding of the underlying issues, or the implications of their position, as is the case at Frank Burns, it's likely to be an anomaly. --B2C 22:52, 15 April 2013 (UTC)

Maybe I have a skewed view, being in the habit of browsing the RM discussions from the back end. But doing so, time and time again, PrimaryTopic is being thrown around like an ill defined buzz term. In this case, if PrimaryTopic means that Frank Burns, the redirect to a mid list character now decades finished, should occupy the undisambiguated title while several other similarly named articles should be found via three clicks and the very large MASH character list, then this guideline is plain wrong. However, I don't think, unlike you, that this guideline really means to says that. --SmokeyJoe (talk) 01:10, 16 April 2013 (UTC) actually, given the character in question, I'm not sure there's not a joke at play

It may mean that or not. It depends entirely on the likelihood that people searching with the term "Frank Burns" are looking for the MASH character as opposed to other uses. If the likelihood is high enough to meet the PT criterion stated there, then it is the PT, by definition. It is that simple.

Looking at it backwards, by deciding the guideline is "plain wrong" if the result is that link redirects to that section about the MASH character in that article, without regard to said likelihoods relative to each other, is, well, backwards. It's also making decisions based on undocumented and unspecified bases that have no known consensus support. --B2C 02:00, 16 April 2013 (UTC)

Explaining the likely bewilderment for someone wanting Frank_Burns_(politician) is so very far from WP:JDLI, and is a concern to which this guideline is silent. --SmokeyJoe (talk) 04:24, 16 April 2013 (UTC)

The politician's page gets 2-6 views per day. The Frank Burns redirect gets 30-40 views per day. That's an order of magnitude difference. Let's get real. Limiting bewilderment is an implied concern managed by looking at likelihood; it's not such an overriding concern that it trumps overwhelming likelihood evidence. --B2C 18:08, 16 April 2013 (UTC)