Wikipedia talk:Disambiguation

Educational value and origin

WP:PRIMARYTOPIC has (somewhat) recently been changed to state:

“ An exception may be appropriate when recentism and educational value are taken into account, especially if one of these topics is a vital article. ”

A discussion has arisen at Talk:Avatar (Hinduism)#Requested move to move it back to the primary topic page, Avatar. One sentiment repeated by several supporters of the move is that "It is fundamentally a Hindu concept" and "All of the modern uses of the word "Avatar" originate with the Hindu concept, so it's fundamentally the primary topic." One editor helpfully pointed out that origination is not a criterion in WP:PRIMARYTOPIC. Permit me to copy-paste my response here:

They do include a suggestion to prefer "educational value", which is clearly related to the "originated from" status. It is educational for people to realize that the word "Avatar" didn't originate with computers or movies, but from a much older concept that these recent technologies have drawn from in their use of the word. It's OK for them to go to Avatar and see something they weren't expecting...as long as we also have hatnotes to help them to get where they wanted to go should they choose to ignore the educational value of the article.

You could easily modify my comment, replacing the word Avatar with Nirvana, Ubuntu, Titanic, or whatnot, and get the same idea. I propose that we clarify the part of the guideline that says "especially if one of these topics is a vital article" to explicitly recommend making the "origin" article the primary topic, in most cases. I'm not trying to open this up to a vote just yet, rather, trying to get some feedback on what you all think of this proposal. ...comments? ~BFizz 17:16, 16 February 2011 (UTC)

I still disagree with this. Titanic doesn't redirect to Titan (mythology), for instance, and it shouldn't. Winston Churchill goes to the prime minister, not Winston Churchill (1620–1688). Boston goes to the Massachusetts city, not to Boston, Lincolnshire. The criteria on traffic, wiklinks, vitality, web searches, scholar searches, book searches, and news searches are sufficient, and will often (but not as a rule) lend primacy to the origin. -- JHunterJ (talk) 17:44, 16 February 2011 (UTC)

I wish I could have added to that discussion about educational value. I am also an editor of the Avatar page, and an advocate keeping the disambiguation page as the first page one sees when looking up the word "Avatar" rather than the Hindu definition. According to the new policy, they do have an argument. However, the more I think about it, I think the new policy is against one of the highest principles of wikipedia WP:NPOV. Who are we to say what is "educational"? The page Mercury is a good example of that. Is the planet more important, or is it the Roman God? Oh well. Oldag07 (talk) 08:09, 17 February 2011 (UTC)

• I feel that "educational value" is subjective and may lead to pointless arguments. The reason we have a primary topic is for the benefit of the majority of readers who are thus saved both a moment of disconcertion and having to click through to a disamb page and search there for the article they want and then click through again. I'm not clear what "educational value" is implying. Does a more developed article have more educational value than a stub? Yes. So would the implication be that primary topics are first chosen from Featured Articles, then Good Articles, then B class, C class, etc, regardless of what the reader is actually looking for? Are we as editors making a decision as to which articles readers "should" be reading, and directing them away from the articles they actually want to read. Unless there is a good argument for keeping "educational value" it should be removed. It may also be worth questioning the statement about "vital articles" as well. I think the vital articles project is worthwhile in attracting attention to subjects that are likely to be of enduring importance, but asking readers to click through a topic they are not interested in is not going to address any systematic bias, it's just going to be a nuisance. SilkTork *^YES! 10:13, 18 February 2011 (UTC)

"Unless there is a good argument for keeping "educational value" it should be removed." - See Wikipedia_talk:Disambiguation/Archive_33#Educational_value. ...comments? ~BFizz 17:30, 18 February 2011 (UTC)

• Educational value has absolutely nothing to do with how well developed an article is. If you read some of the other discussions on this page you can get familiar with the concept. --JaGa^talk 10:30, 18 February 2011 (UTC)

  It is of educational value to know the origin of a word we are using for something, be it a film or guitar synthesizer or social networking, especially when this word happens to be centrally important to a philosophy or religion (Hinduism is both) and happens to in use for thousands of years. I am not suggesting that we sould create new criteria and distictions by which to decide what is educational and what not. I understand the above mentioned concerns. However some terms are simply very obviously useful educationally and this term (Avatar) happens to be one of them. It is also highly relevant (and educational) to all the film fans to know that this term is a term carrying a huge significance for billions of people and why. Words have power, someone uses this power. We show where the power comes from. Else it is like hiding this significance by bypassing it. Hoverfish Talk 13:51, 18 February 2011 (UTC)

  We show the reader the information the reader is looking for. We do not show the reader the information that we feel they should have been looking for instead. In this case, the disambiguation page will direct readers most easily where they want to go, and if they are reading about the film and want to learn more about that significance, Avatar (2009 film)#Themes and inspirations will lead them to it. Primary topic is not a value judgment, an indication of importance, or an identification of word origins. -- JHunterJ (talk) 13:59, 18 February 2011 (UTC)

  Actually I think there is a certain amount of value judgment. Even before educational value, we gave vital articles more leeway. But who decided what was vital, and how? These are articles that the community has decided are of great importance to the encyclopedia. That's a value judgment. Educational value, really, just allows the community to decide the "vitality" of an article without being subject to membership on the vital articles list. --JaGa^talk 18:59, 19 February 2011 (UTC)

  I meant that not "winning" the primary topic position should not be taken as a loss of valuation of the topic. Too much of the arguments around primary topic placement stem from the view that it's an award, like "good article" status, to be sought (and fought for) as validation of the topic, when instead it's just a navigational aid for the benefit of the readership taken as a whole. -- JHunterJ (talk) 19:03, 19 February 2011 (UTC)

  Let me put it another way: someone asks for "Apple". This does not link to a disambiguation page but it links to Apple the fruit. Why? Did someone make an extensive search on how many people are looking for the fruit and how many for the computer? No, because we take it as obvious that Apple means the fruit, just as billion others take it as obvious that Avatar means for their side what in the Western world is known by the Hebrew term Messiah (though there are some differences in the two concepts). I heard in a relevant discussion (though I don't remember where it was) the argument that if someone makes a very high budget film called Jesus and then it becomes a big success, the word "Jesus" should also lead to disambiguation page. I find this way of thinking biased, anti-scholarly, counter-educative and unfitting for Wikipedia. Just like it would look very unfitting if the word Apple led to disambiguation page. I may not be very versed and JHunterJ may easily turn my arguments down, but I know what I'm saying and I know it is important for Wikipedia.Hoverfish Talk 14:46, 18 February 2011 (UTC)

  Note that Messiah does not redirect to Messiah (Handel), despite it being "one of the most popular works in the Western choral literature". Nor does it redirect to Messiah (disambiguation). This is a good example of what is meant by "educational value". ...comments? ~BFizz 17:28, 18 February 2011 (UTC)

  Or of primary topic in English-language Wikipedia. It was the primary topic before the educational value bit was added to the guidelines. -- JHunterJ (talk) 17:32, 18 February 2011 (UTC)

  But under which pre-existing criteria? Messiah only has about 1.5x the number of incoming wikilinks and page views that Messiah (Handel) has (though page views for (Handel) skyrocket to about 3x as much (~6k/day) as the PT (~2k/day) during December), and both topics, along with Messiah College, show up prominently in google searches. Clearly, it was chosen because of the basic idea behind the "educational value" guideline; even though it wasn't in place, it's just intuitive. ...comments? ~BFizz 20:41, 19 February 2011 (UTC)

  Educational value can be established by a review of authoritative sources like other encyclopaedias and dictionaries.--Redtigerxyz ^Talk 04:55, 20 February 2011 (UTC)

  Or by use of Google Scholar and Google Books searches. -- JHunterJ (talk) 13:01, 20 February 2011 (UTC)

I'm not sure a consensus is emerging here, nor clarity regarding the disambiguation primary topic function. Where I think there is agreement and understanding is that we disambiguate when there is more than one likely topic for a given search term/article title - and we use the disambiguation process as an aid to the reader to help them find the topic they want. To save time and least confusion, we give a primary topic function to the most likely search term if there is clear evidence that one term is going to be more likely. The majority of readers will then arrive on the right page. The minority will have to search further.

The way we determine the most likely search term is to decide through discussion which is the most likely topic, using incoming links, page hits and amount of books and articles already on that topic. This is considered reasonable evidence, and I don't think that such evidence is in dispute.

However, added to the guideline six months ago was the notion that articles listed on Wikipedia:Vital articles should be considered as part of the evidence. Discussion on the issue was inconclusive, though the use of vital articles as part of the determining process for deciding the primary topic was kept.

Part of the objection for using vital articles is that it adds a new thread to the disambiguation process. One in which we are not aiding the reader to arrive at the most likely topic, but at one we feel is most valued, and therefore we should promote.

More recently the notion of "educational value" was also added. This sits alongside the vital article argument that it would be nice to promote certain articles, because they are educational.

The addition was felt to help counter any criticism that Wikipedia gives undue weight to popular culture, and gained support. Though as others who were not part of that discussion are becoming aware of that addition, it's inclusion is being questioned. Which is where we are now.

I think that presenting the majority of readers with a topic they are not looking for, mainly because Wikipedia editors have decided that the article is one they should be reading instead, is not our purpose, and should not be a part of what we do. It should also be borne in mind that popular culture is represented in the vital articles, particularly at Level 4. Popular culture is as vital and important a topic as any other. I note above that Titanic is mentioned. The primary topic for that is currently at the ship, rather than the film. In vital articles level 4, the film is listed, but not the ship. It is the piece of popular culture that is considered the most important by vital article editors, while disamb editors feel it is the ship. Using value judgements will lead to disputes. Guidelines should provide a path away from dispute, not lead people into conflict.

However, I can see a value in making people aware of a vital article, or one that may have some educational value. It may be worth opening up a discussion to the community, in which vital articles are indicated on disamb pages. I think it may be a bit much to flag them up on a hatnote, but a comment on a disamb page might be worthwhile. We would be leaving the choice up to the reader, which would be appropriate.

Meanwhile, I would like to remove the recent additions from Primary Topic, as I feel it is inappropriate to take the majority of readers to a topic they don't want. Would people make it clear if they would object to the removal of: "An exception may be appropriate when recentism and educational value are taken into account, especially if one of these topics is a vital article. In such a case, consensus may determine that the article should be treated as the primary topic regardless of whether it is the article most sought by users." SilkTork *^YES! 13:44, 23 February 2011 (UTC)

I object. While the language might not be optimal, I think it represents an important consideration in determining primary topic. I feel taking as a goal for disambiguation simply to minimize the number of clicks is not particularly encyclopedic. Wikipedia is first an encyclopedia that happens to use the web as primary media. IMO, considerations based on encyclopedic value take precedence over decisions based on the media. The criteria for primary topic had previously had language that extended disagreement might also be an indication that there is not a primary topic. In a way, I think that was a clearer criteria than educational value -- though perhaps it needed tightening that disagreements needed to be based on reasoned principles rather than variants of WP:IDIDNTHEARTHAT. older ≠ wiser 14:59, 23 February 2011 (UTC)

I see the "educational value" as an improvement over the "vital" criteria. Perhaps instead of "origin" we should use the term "leech": if the choice of name for a thing (such as Titanic the film) was strongly influenced by an older thing (titanic the ship) then the older thing is probably the primary topic. When the "popular culture" thing draws heavily from the meaning of something else by adopting its name, that "something else" is clearly the true meaning behind the word. What legitimate encyclopedia would place the movie at Titanic with a hatnote saying "for the ship this film was based on, see ___"? Sure, more users might get there in one click, but it's just wrong for an encyclopedia. ...comments? ~BFizz 15:30, 23 February 2011 (UTC)

I am still opposed to any indication that age or ancestry should be considered, beyond what it already covered by its ongoing primary-ness based on other criteria. Often, the oldest or first will be primary, but because it has retained widespread general educational value (coverage in the kinds of things measured by Google Books and Google Scholar). -- JHunterJ (talk) 21:40, 23 February 2011 (UTC)

Age isn't what I'm talking about, though. Origin is. It just so happens that we haven't invented time travel yet, so something can not be derived from anything newer than itself. Origin isn't really measurable by # of hits on a google search, but it's certainly perceivable. If two completely unrelated things happen to share an ambiguous name, then the older one is not necessarily the PT. This is often the case with acronyms, where dab pages are almost always appropriate rather than a PT. ...comments? ~BFizz 22:15, 23 February 2011 (UTC)

Which is why I said "or ancestry". -- JHunterJ (talk) 22:23, 23 February 2011 (UTC)

Wider community input

When a search title relates to two or more articles should readers be taken to the most likely article they are looking for or the most educational? See above for some previous discussion on this issue, including links to prior discussions. SilkTork *^YES! 17:17, 4 March 2011 (UTC)

• Most likely. I support most likely as that is what most readers will be looking for. Deliberately directing readers to an article they are not looking for is a concept I am very uncomfortable with. And the arbitrary value judgements that would take place as to which article has the most educational or encyclopaedic value would be disruptive and lead to instability as one faction or other temporarily wins out in discussions and the search term would be redirected again and again. Our long standing system of using incoming links, page hits, scholarly sources, Google, and other means of determining the most likely is not a perfect system, but it does tend to produce solid data on which to make a sound judgement. SilkTork *^YES! 17:26, 4 March 2011 (UTC)
• Generally speaking the most likely, though I'm not adverse to introducing other criteria that would give a certain extra weight to more "serious" subjects over ephemeral ones (though I'm not sure how they should be phrased; I don't think "educational value" is meaningful enough).--Kotniski (talk) 17:29, 4 March 2011 (UTC)

  I agree that "educational value" is overly vague, and needs clarification. ...comments? ~BFizz 01:21, 5 March 2011 (UTC)

• base meaning - In cases where the most likely article is also the most "educational", the answer is clearly to redirect there. In cases where these two don't coincide, I believe the encyclopedic thing to do is consider how heavily the popular choice's name draws from the meaning of the educational choice. I'd say typically names of movies, TV shows, books, articles, etc should defer to the concept behind their name, unless it is a clearly distinctive name. Mythbusters need not redirect to myth, Harry Potter should clearly redirect to the media rather than a list of people with that name, but Pearl Harbor, Titanic, Atlantis, Twilight, and so forth should obviously redirect to the concept, and not the media named after it. Some names, however, are still ambiguous and should go to the dab, for example, Evita should go to the dab instead of Eva Peron or Evita (film). This is why I started the "origin" discussion above; perhaps "educational value" isn't a very clear way of stating the policy. However, origin isn't the silver bullet answer either. Radio should not redirect to Radio (film), though the film's name is only tangentially related to a Radio. However, I feel Wikipedia should typically encourage a "base meaning" spirit when considering primary topics. If there is an underlying base meaning to a word, then in most cases the search term should redirect to the base meaning rather than the popular one. It's just a guideline, though. There can and will always be exceptions to the rule. ...comments? ~BFizz 17:40, 4 March 2011 (UTC)
• The phrasing of the question is misleading. I don't think this is a simple binary distinction and I don't think we should include simplistic guidance banning placement based on educational value or other considerations besides article traffic. older ≠ wiser 18:14, 4 March 2011 (UTC)

  I agree. The issue is more complicated than a binary choice. Kaldari (talk) 21:22, 4 March 2011 (UTC)

  I also agree. The originally phrased question ignores the possibilities of 1) sending the user to the dab page, and 2) sending the user to a search results page. Few would support #2 but it is an option. ...comments? ~BFizz 01:21, 5 March 2011 (UTC)

• Most likely expected. Educational value is often useful for determining likelihood, as there are problems with equating simple traffic counts to likelihood of the article being expected (or perhaps not being surprising) to the reader landing there. -- JHunterJ (talk) 19:01, 4 March 2011 (UTC)
• Follow current guidelines. It's a bit more complicated than a simple either-or choice. I think the current guidelines do a good job of addressing the issue. Generally most likely is the best choice, but there are warranted exceptions. Kaldari (talk) 21:21, 4 March 2011 (UTC)
• Educational value. My view is that Wikipedia should be organized according to the view that everyone who can read it will read it i.e the structure and practises of the project should reflect the highest aspirations. If you start organizing it based on access patterns, and bearing in mind internet use is a relatively narrow demographic, it will begin to lose its focus as a universal encylopedia. Betty Logan (talk) 01:43, 5 March 2011 (UTC)
• Follow current guidelines and keep educational value as one of the metrics. Educational value is no more arbitrary then the currently approved and most commonly used metric in page move discussions: "Google hits". Using Google hits is the application of Original Research to a non-reliable source. "Most likely" is also quite arbitrary. The way that other Reliable Source encyclopedias title their articles might be the most accurate metric of educational value, since these are reliable sources not only in terms of content, but in terms of choosing titles for that subject. In the end, page move discussions must balance all of these things, but "educational value" certainly deserves more weight than "Google hits". First Light (talk) 01:45, 5 March 2011 (UTC)
• Base Meaning If I'm searching for something, it's because I want to find that thing. Get me as close as you can get me, but don't presume that I'm looking for something more educational. Let's face it, plenty of hits come from people wanting to look up the plot to a Simpson's episode. We don't want to alienate people by making it harder for them to find what they are looking for. Sven Manguard Wha? 02:41, 5 March 2011 (UTC)
  • Any idea on how to define base meaning? I can see several possible definitions that would take editors to the wrong article if base was to imply oldest or first. Vegaswikian (talk) 02:59, 5 March 2011 (UTC)
    • It's hard to describe, but I understand "base meaning" as the original concept from which another subject's name was derived. In the example of "twilight", the book/movie series' title comes from the time period known as twilight. /ƒETCHCOMMS/ 03:57, 5 March 2011 (UTC)
      • And there lies the problem in the details. So should Syracuse, Sicily be moved to Syracuse? Of to try another, if Jean Amelia Summerlin ever received an article, should Summerlin become a redirect to her article since she is the namesake for the mater planned community? Vegaswikian (talk) 07:15, 5 March 2011 (UTC)
      • Agreed. Boston (Boston, Lincolnshire), Winston Churchill (Winston Churchill (1620–1688)), and Bead (Rosary beads) are other titles that would be poorly served by placing the article on the "original" concept there. -- JHunterJ (talk) 12:48, 5 March 2011 (UTC)
• I agree with B Fizz and Bkonrad (older ≠ wiser). This is not as simple as it seems, and I think we should decide this on a case-by-case basis. However, I think it's pretty obvious we should leave Twilight at the concept of the natural occurrence and not the poorly-written fictional series, etc. It is one of those basic topics that should stay un-disambiguated. I also, as per Kaldari, don't see any need to change anything. We can have exceptions and we can have discussions for each contentious article title. /ƒETCHCOMMS/ 03:53, 5 March 2011 (UTC)
• Case-by-case Attempting to set a general rule based on a vague concept like "most educational" is doomed to failure. Let's just figure this out on a case by case basis.--Danaman5 (talk) 05:08, 5 March 2011 (UTC)
• Case-by-case I fully support Danaman's viewpoint here. Sweeping generalised rules like this in a fully contributory and discussable information source will lead to the limitation of knowledge. People's viewpoints about which is more 'educational' or more 'anything' could badly damage the structure of information. Paul Bedson (talk) 23:27, 9 March 2011 (UTC)
• Comment The current guideline doesn't make "most educational" a rule, even in general terms. Nor does it make it a one-to-one comparison, as the RfC states in bold, between "the most likely article they are looking for or the most educational." It's only one of several elements that need to be balanced and discussed to arrive at a consensus. It merely says that "An exception may be appropriate when recentism and educational value are taken into account, especially if one of these topics is a vital article." There are several qualifiers there: "an exception may be appropriate"; "recentism and educational value"; "especially if....vital article". First Light (talk) 05:14, 5 March 2011 (UTC)
• Most likely if it's genuinely clear cut, being careful to avoid recentism and temporary likelihood of search target. Otherwise go for the primary topic as is likely to be judged in the long-term. Judging which article is "most educational" is fraught with problems.--Michig (talk) 10:20, 5 March 2011 (UTC)
• Comment. It may be appropriate and less potentially disruptive if, in cases of doubt, a disamb page is set up rather than taking the reader to an unlikely article. Adding in considerations of "base article" or "most educational" introduce subjective opinions in what is after all simply an index. If there were no other possible topics then a search term would take a reader directly to what they were looking for, regardless of opinions of the educational worth of such a topic. Soggy biscuit goes direct to the appropriate topic rather than via biscuit. It could be argued that biscuit is the base term, and is of educational value as it informs the reader about the object that the activity is named after. Clearly such an argument would be silly, but is essentially no different to any such argument suggesting that I be taken somewhere I don't want to go because someone feels I need educating about it. Taking someone to a neutral page in which a list of related topics is given is of greater educational value than taking someone to the wrong topic. It is also less confusing. An index page is reasonable. The wrong article is unexpected. SilkTork *^YES! 11:12, 5 March 2011 (UTC)

• Pretty much all disambiguation is subjective though really, isn't it? Even an index page is subjective because that's still a decision undertaken by us based on preconceived notions, because 100% of readers will then be taken to a page they don't want to be at, rather than maybe 40% of readers. If we were going to have a completely objective view then we would probably just follow the precedent set by other encylopedias, which is what I think should be done in cases where analagous entries can be found. So in some cases article title designation could be made an objective process, and in others we are still left with our imperfect subjective processes, but basically it would get us to a stage where an entry in Brittanica could be found on Wikipedia under the same name. Betty Logan (talk) 12:19, 5 March 2011 (UTC)

• Most likely - the value of Wikipedia in the eye of the public will be greater the more likely a person is to find what (s)he's looking for immediately on the page (s)he loads first. עוד מישהו Od Mishehu 17:25, 6 March 2011 (UTC)

  But if we make it well-understood that Wikipedia sends you to the "educational" thing when you type it in, then users will expect to go to that thing when they type it into the search box. Wikipedia serves a different purpose than Google. It's Google's job to turn words into a link the user wants. It's Wikipedias job to be an online encyclopedia. ...comments? ~BFizz 20:36, 6 March 2011 (UTC)

• Most likely. This seems obvious to me.. It depends on what the reader wants to be educated on, so in all cases the most likely target IS the most educational because that's what the reader wanted to learn about. Although there are some exceptions, as in the case where "big tits" redirected to Great Tit.. I did think that was a much better target ;p -- œ^™ 22:26, 6 March 2011 (UTC)
• Most likely. If I use an encyclopedia I expect to be sent as close as possible to what I'm looking for. While an encyclopedia serves also to educate, it is not a textbook. Kudpung (talk) 06:44, 7 March 2011 (UTC)
• More educational, if by that you mean, I dunno, more scholarly or more mainstream of more... scientific, or more real-culture as opposed to pop-culture, or whatever. Like in an example given above, "Twilight" should go to the atmospheric phenomena and not the pop-culture book series. But if the "more likely" is really an order of magnitude more likely, then it would be OK to go to the more likely. Really it's kind of an art and needs to be done on a case-by-case basis. But with the default tending toward the more educational. In some cases maybe you have to throw up your hands and go to a disambig page, I guess. Herostratus (talk) 09:23, 8 March 2011 (UTC)
• Base meaning. The term people are most likely looking for may be influenced by various aspects regarding popular culture or some other recent phenomena. That's why dictionary / scientific definitions should always come first, with all other meanings thrown up in a disambiguation page. _EricLeb⁰¹ (Page | Talk) 19:08, 8 March 2011 (UTC)
• False dilemma The guideline does not require an editor to choose between "most likely" and "most educational". There's a strong emphasis on "most likely" but it also allows educational value to be a factor in consideration. That way, we don't have to move Apple Inc. to Apple or move Tool (band) to Tool. --JaGa^talk 23:51, 8 March 2011 (UTC)
• Rather than "likely" or "educational", I would place the preference on "real" over "fictional". If the name of something in the real world is used as well for a fictional representation of it (such as "Evita" or "Battle of Los Angeles"), then the name should be for the real one. If the name of a real and known-by-all thing is taken by an unrelated thing because they like the name or whatever (such as the names of many music bands), then the general concept should be used. MBelgrano (talk) 02:44, 9 March 2011 (UTC)
  • A very reasonable alternative in my mind. Avoids the issues with the other options. Vegaswikian (talk) 03:00, 9 March 2011 (UTC)
  • Introduces problems of what is "known-by-all" (which we already have to deal with in move requests even though it's not a criterion). Should Black Sabbath should be moved to Black Sabbath (band) so that the base name can lead to the Shabbat Chazon or Black Mass? Should Amadeus be moved to Amadeus (play) so the base name can lead to Wolfgang Amadeus Mozart? (If you can deflate either of those specific examples, note that there are others possible.) The question of "likely" and/or "educational" would also still need to be answered in the cases where multiple things are "real" (such as at Lincoln) or multiple things are "fictional", such as the recent request for Inception). Where "real" or "non-band" things are primary, I believe the criteria of "most expected", informed by both traffic and educational value. Traffic alone won't address the desire to avoid surprising readers; someone looking for the band Muse wouldn't be surprised (I hope) to land at Muse, even if the band article had more traffic. "expected" and "educational" covers the "known-by-all" well. -- JHunterJ (talk) 12:10, 9 March 2011 (UTC)
  • Agree with JHunterJ on this point. There are numerous counterexamples where a fictional topic IS by all measures the primary topic. Should Hamlet be moved to Hamlet (play)?— Preceding unsigned comment added by Bkonrad (talk • contribs) 22:52, 9 March 2011 (UTC)
  • Great idea in general. But there is also an idea of "trademark-ness" that holds sway. Some names could allude to other things, but are not commonly used in that way. JHunterJ's Amadeus example was good. ...comments? ~BFizz 00:09, 13 March 2011 (UTC)
• False dilemma For the reasons stated by JaGa. Alanscottwalker (talk) 17:19, 12 March 2011 (UTC)
• Follow current guidelines, but I think "namesakehood" (what some call "base meaning" above) should be added to the factors considered when deciding primary topics. --Cybercobra (talk) 00:42, 13 March 2011 (UTC)
• I hardly realize what is the discussion about. We know, even if a word is essentially ambiguous, there may be cases where we give the preference to one article, such as a primary (base) meaning. But these are quite rare cases, so Wikipedia should follow a general rule: if ambiguous, then a disambiguation page. It is a bit less handy for readers than alternative approaches, but otherwise we would have a high risk of incorrect internal links to articles which can remain undetected and unfixed for years. Incnis Mrsi (talk) 14:55, 13 March 2011 (UTC)

• It's my opinion that the guideline should more strongly endorse a disambiguation page as the default in cases where usage is ambiguous with some exceptions. AFAIC, internal traffic statistics are of value only when there is a very marked difference in traffic and the primary topic is not contra-indicated by other measures (such as Google Web, Google Books, Google Scholar, common sense of editors, etc). older ≠ wiser 15:29, 13 March 2011 (UTC)

• I'm also not sure you're asking the right question. It's not really about likely versus educational. It's definitely not about popularity. It's some combination of all of them which indicates it's some other criteria. Relevance? I think that's the criteria. It's just hard to define. It's the one that best fits the search term. Shooterwalker (talk) 12:43, 16 March 2011 (UTC)
• Wrong question. The correct question, IMHO, is whether we can determine at all, to a high likelihood, what the probable intended target of the search term is. Bear in mind that since the term is (by definition) ambiguous, we can never be certain what a particular reader thinks it should mean. Some terms, however, are so widely and commonly used throughout the English-speaking world for a particular purpose that we can reach a consensus that readers who use them very probably intend the common usage. If we cannot reach consensus on this question, then the title should contain a disambiguation page, not redirect to either a "most likely" or a "most educational" substitute. --R'n'B (call me Russ) 13:29, 16 March 2011 (UTC)
• Most likely as determined by consensus. Marcus Qwertyus 00:50, 18 March 2011 (UTC)

Motion to close or change rfc

This rfc was a poorly worded false dichotomy. What exactly are we discussing? Why are so many contributing to the discussion as if it were a !vote between "most likely" and "most educational"? What do these !votes even mean? I suggest we close this rfc, with the possibility of opening a separate, more clearly defined, rfc. ...comments? ~BFizz 22:31, 6 March 2011 (UTC)

We are discussing that we have always disambiguated by "most likely", and that recently a wording has been introduced to direct people not to what they expect to read, but to something that a Wikipedia editor has decided is "more educational". The new wording is: "An exception [to directing to the most likely article] may be appropriate when recentism and educational value are taken into account, especially if one of these topics is a vital article. In such a case, consensus may determine that the article should be treated as the primary topic regardless of whether it is the article most sought by users." My bolding. This wording says that editors may decide (in situations which are not made clear) that readers who are searching for one topic may instead be directed to another because somebody considers it to have greater educational value. Someone might decide that people wanting to check who directed Titanic would be better served by taking them to the concept behind the name instead. This brings to mind Dotheboys Hall:

"This is the first class in English spelling and philosophy, Nickleby," said Squeers, beckoning Nicholas to stand beside him. "We'll get up a Latin one, and hand that over to you. Now, then, where's the first boy?"

"Please, sir, he's cleaning the back-parlour window," said the temporary head of the philosophical class.

"So he is, to be sure," rejoined Squeers. "We go upon the practical mode of teaching, Nickleby; the regular education system. C-l-e-a- n, clean, verb active, to make bright, to scour. W-i-n, win, d-e-r, der, winder, a casement. When the boy knows this out of book, he goes and does it. It's just the same principle as the use of the globes. Where's the second boy?"

"Please, sir, he's weeding the garden," replied a small voice.

"To be sure," said Squeers, by no means disconcerted. "So he is. B-o-t, bot, t-i-n, tin, bottin, n-e-y, ney, bottinney, noun substantive, a knowledge of plants. When he has learned that bottinney means a knowledge of plants, he goes and knows 'em. That's our system, Nickleby: what do you think of it?"

"It's very useful one, at any rate," answered Nicholas

I think that Squeers' system is perhaps more useful than than of attempting to educate people by sending them to articles they didn't ask for. It's a novel idea though, and perhaps should be introduced at university and research libraries around the world. If a librarian has a doubt about which book is being ordered, they offer substitutes rather than deliberately giving a researcher the wrong text. SilkTork *^YES! 17:56, 8 March 2011 (UTC)

The guideline's wording has never stated that the 'most likely' should unquestionably be the primary topic. It has always stated that consensus should be reached, and page statistics might be a good indicator. Recently, the guideline was expanded to illustrate some situations where it might not be a good indicator. But the main idea of the guideline is the same: consensus for each particular case, rather than blindly following one criteria or another. This rfc takes tiny points of the guideline completely out of context and pits them against each other. It is juvenile and pointless. This shouldn't be a criteria war. Conflicting criteria can coexist, and consensus can decide which is most applicable when. ...comments? ~BFizz 19:14, 8 March 2011 (UTC)

"The guideline's wording has never stated that the 'most likely' should unquestionably be the primary topic." Yes it has. That is the purpose of primary topic. If there is a topic that is most likely to be what people are looking for, we direct people there rather than a disamb page. The disamb process was set up to act as an index page for when we have the same titles for different topics. The aim has always been to seek to deliver people to the page they are searching for. We don't, however, always know (or agree) on what might be the most likely, so we have used various data, such as links, Google hits, page views and the amount of books on a topic. This is not perfect, but is the best we have so far. Any attempt to improve the way we decide what is most likely is to be welcomed. Confusing the issue by introducing a conflicting and subjective judgement, and explicitly stating that we are deciding NOT to direct people to the most likely, is not in the best interests of our readers, and is certainty not what disambiguation is about.

"This rfc takes tiny points of the guideline completely out of context and pits them against each other." This RFC is seeing if there is consensus for a recent introduction to the guideline which introduces a conflict. The RFC isn't introducing the conflict, the recent new wording has done that by changing the purpose of primary topic from "most likely" to "regardless of whether it is the article most sought by users". That changes the whole purpose and ethos of the disambiguation project from helping direct readers to what they are looking for into directing readers into someone we think they should be reading instead.

There is a good intention behind the change, and the enthusiasm for the notion of the moral good has swept some people along. I hope, though, that people will pause and think about the implications of the change. SilkTork *^YES! 20:58, 10 March 2011 (UTC)

Well put. This is a pretty confusing RfC; the choice we're presented with doesn't represent what's in the guideline. --JaGa^talk 00:07, 9 March 2011 (UTC)

The Rfc is badly worded. Editors above agree that some judgment needs to be exercised in selecting titles. It's therefore appropriate to have in the guidelines things like "educational value" as one of multiple criteria to be discussed in reaching consensus. Wikipedia is based on consensus in editorial decisions, and this is an editorial, not a factual question, so consensus based on any and all relevant criteria, such as educational value, should be explicitly stated in guidelines. Alanscottwalker (talk) 17:12, 12 March 2011 (UTC)

Refocus?

A number of people are suggesting that the question is the wrong one. Perhaps we should be looking at the purpose of disambiguation and primary topic. Disamb was set up to deal with the situation in which a reader enters a search title and there is more than one possible article for the term. So what do we do? We initially create a disambiguation page in which the possible articles are listed. This is a sensible approach, and I don't think anyone disagrees with that. However, there are circumstances in which one of the articles is the most likely to be the one that the majority of people are looking for, and so we have the notion of primary topic. Where we have difficultly is in how we decide what the primary topic actually is. Perhaps the questions should be: 1) Do we keep primary topic? and 2) If we keep it, how do we decide what the primary topic is? Though there is a sense that some primary topics are very obvious The Sun, I would rather be taken to Sun (disambiguation) in that instance (and discover how many topics there are with Sun as part of the title), than be taken to RMS Titanic when entering Titanic and expecting Titanic (1997 film). There are ongoing disputes and problems with the notion of primary topic, and our systems of working out which is the primary topic are unsatisfactory. We could resolve that by always having a disamb page. Disamb pages are themselves informative and educational, and less frustrating than landing on the wrong page. SilkTork *^YES! 12:07, 17 March 2011 (UTC)

I disagree with the proposal of never having a primary topic, if that is the proposal. The focus is (or should be) how to determine which topic being placed at the base name would be most beneficial for the readership; if none would very beneficial, then there's no primary topic. Disambig pages themselves are not educational or informative; they are navigational. I do not empathize with the possible frustration of someone landing on Red when they intended Red (2008 film), for example, but I can empathize with the frustration of landing at Red (disambiguation) when intending Red. -- JHunterJ (talk) 12:38, 17 March 2011 (UTC)

There are a few extreme solutions to this problem. When a user types in a search term:

• Always take them to the search page [this is a horrible idea]
• Always take them to a disambiguation page
• Always take them to an article

None of these extremes is a good choice, imho. Currently, we have a very vague "let consensus decide" hybrid approach that applies either #2 or #3 on a case-by-case basis, defaulting at #3 in unambiguous cases. SilkTork seems to propose a more straightforward hybrid approach, where #3 is used for the unambiguous terms, and #2 is used for the ambiguous terms. While the clarity of this proposition is appealing, I personally feel disambiguation pages should "get out of your way" as much as possible.

I also don't think that taking our readers to the "wrong" page is such a bad thing. I return to my Twilight example; I don't care how many people expect to reach Twilight (book) by typing in "Twilight", those people deserve a wiki-slap-in-the-face by taking them to the page about the "real thing". I could say the same about Titanic. Reader, did you really forget that there's actually a ship called the "Titanic" after which the film was named? One way of determining the Primary Topic might be this: if we had to choose only one of the articles to keep, out of all the ambiguous choices, and delete the rest, then which one would it be? If there is a clear "encyclopedic" choice (not basing the choice on popularity), then that is probably the PT. Disclaimer: I'm not saying this measure is infallible. Some might say that Titanic (film) is more culturally significant than the events it portrays. ...comments? ~BFizz 15:47, 17 March 2011 (UTC)

None of the readership "deserves" a wiki-slap. The encyclopedia is a tool for the readers, not a tool for the editors. Determination of hypothetical "sole topicness" is not related to "primary topicness" -- sole topic should be determined by importance, but primary topic is not (or should not be) a measure of importance, age, or ancestry, except where those are submeasures of reader expectation (not editor valuation). -- JHunterJ (talk) 17:17, 17 March 2011 (UTC)

A little learning is a dangerous thing;

drink deep, or taste not the Pierian spring:

there shallow draughts intoxicate the brain,

and drinking largely sobers us again.

"Titanic" is the name of the film, which is based on the story of the RMS Titanic, which was named after the Titans. Titanic also relates to titanium, another thing named after the Titans, so it could be quite reasonably argued that titanic should go to the Titans. However, assuming that people are looking for one thing when they may be looking for something else is very problematic, especially when many of us are not fully knowledgeable about everything. We know the stuff we know, but we don't know everything, and rely on sources such as Wikipedia to inform us about topics we know little about. Within our own span of knowledge we can't always reasonably be expected to know exactly what other people are looking for. While a discussion among a broad range of Wikipedians will improve our knowledge base, we are aware that there is a systemic bias among the whole community, which increases when it comes down a handful of editors deciding on primary topics. If there is clear and broad consensus (it's bleedin' obvious) that a topic is most likely, then primary topic can serve a purpose, but if there are disputes and discussions, then perhaps we should not be casting around looking for reasons to have a primary topic. If we have to do a Google search or bring in notions of educational value, then we are moving away from the bleedin' obvious, and we are starting to impose our own biased (and sometimes poorly informed) points of view. The notion of disambiguation is to help readers find what they are looking for. If we are unsure then we should default to a disambiguation page. SilkTork *^YES! 12:02, 18 March 2011 (UTC)

I guess I differ with both of you in that I don't see it as a major inconvenience to the user, who has typed in an ambiguous search term, even if they didn't realize it was ambiguous, to be taken to a disambiguation page. (If the term is not actually ambiguous, of course, that's a whole different issue.) I would much rather have that happen than have many users be taken to an article they didn't want at all. I agree with JHunterJ that "primary topicness" is a matter of reader convenience, not a value judgment; where we differ is that I tend to give more weight to the convenience of the minority who are looking for the less-popular topic, and would go to greater lengths to avoid sending them to the wrong article. As I said earlier, there are some (perhaps many) technically ambiguous terms whose usage in English is so clearly associated with one particular meaning that not only should that title take readers to an article about that primary topic, but it would be absurd if it didn't. I'd put "Sun" in that category; somebody who types "Sun" in the search box looking for a record company or newspaper and ends up on the article about the star may be disappointed, but they could not reasonably claim to be surprised. On the other hand, where there is more than one topic that could reasonably be expected to be the target of a particular search term, for example George Clinton, that term should take the reader to a disambiguation page. At least to me, the inconvenience to those interested in the musician of having to click twice is outweighed by the inconvenience to those looking for the U.S. Vice-President or another notable person of that name having to figure out how to get to the disambiguation page and then navigate a second time to the correct article. How many readers doing historical research would arrive at the article about the musician, conclude "This stupid Wikipedia doesn't even have an article about a U.S. Vice President," and give up? How can we know? --R'n'B (call me Russ) 18:14, 17 March 2011 (UTC)

I agree with this. If the topic is clearly obvious let's take people there and have a hatnote to help those who are looking for something else. If there is reasonable doubt, let's take people to a disamb page and let them decide. Any case in which people are having a dispute should default to a disamb page. SilkTork *^YES! 12:06, 18 March 2011 (UTC)

I think Russ has essentially restated what I have always understood the intent of primary topic and disambiguation to be. Over time, "most commonly searched for" or similar language seems to have been interpreted by some to mean that the goal of disambiguation is to minimize the number of clicks for most users as measured by page traffic statistics. older ≠ wiser 12:15, 18 March 2011 (UTC)

(ec) By "clearly obvious" I mean that there should be no need to argue the case for primary topic - no need to bring in scholarly references or page hits. "Sun" should go to Sun, "Earth" should go to Earth, "Soil" should go to Soil, "Michael Jackson" should go to Michael Jackson, and "Beer" should go to Beer. If there's reasonable doubt or a reasonable challenge, then it shouldn't be a case of changing the primary topic, but of simply going straight to a disamb page. SilkTork *^YES! 12:20, 18 March 2011 (UTC)

Yet Wikipedia editors will argue over the primary topicness of things that other editors find obvious. All this would do is move the target from to arguing over what's reasonable. You cannot have editors saying "But I am being reasonable!" as justification for no primary topic. -- JHunterJ (talk) 13:55, 18 March 2011 (UTC)

Of course you can, but if they cannot provide reasons to support their subjective view of "reasonableness" then other editors may disregard it. WP:CONSENSUS does not not mean unanimity. --R'n'B (call me Russ) 18:10, 18 March 2011 (UTC)

Correct. Which is why there is no need for this "clearly obvious" bit, unless it is to make the disambiguation page more likely the base name page, in which case I oppose the change. -- JHunterJ (talk) 18:15, 18 March 2011 (UTC)

Suggested new wording

Following the discussions above, I propose we change the wording in Primary Topic.

The wording initially proposed has not been accepted.

Initial proposal

Following the discussions above, I propose we change the wording in Primary Topic from: Although an ambiguous term may refer to more than one topic, it is often the case that one of these topics is highly likely—much more likely than any other, and more likely than all the others combined—to be the subject being sought when a reader enters that ambiguous term in the Search box. If there is such a topic, then it is called the primary topic for that term. If a primary topic exists, the ambiguous term should be the title of, or redirect to, the article on that topic. If there is .... on one page). An exception may be appropriate when recentism and educational value are taken into account, especially if one of these topics is a vital article. In such a case, consensus may determine that the article should be treated as the primary topic regardless of whether it is the article most sought by users. 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. Tools that may help to support the determination of a primary topic in a discussion, but are not determining factors, include: Incoming wikilinks from Special:WhatLinksHere Wikipedia article traffic statistics Google web, news, scholar, or book searches (NOTE: adding &pws=0 to the google search string eliminates personal search bias) to Although an ambiguous term may refer to more than one topic, it is sometimes the case that one of these topics is highly likely—much more likely than any other, and more likely than all the others combined—to be the subject being sought when a reader enters that ambiguous term in the Search box. If there is such a topic, then it is called the primary topic for that term. If there is .... on one page). 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. If there is no clear consensus, or where there is more than one topic that could reasonably be expected to be the target of a particular search term, for example George Clinton, that term should take the reader to a disambiguation page. If there are prolonged or repeated discussions regarding which topic is primary, then the default position should be to use a disambiguation page. If there is reasonable doubt, such that evidence and subtle arguments need to be used rather than common sense, then the default position should be to use a disambiguation page. Here are the differences: Although an ambiguous term may refer to more than one topic, it is often [sometimes] the case that one of these topics is highly likely—much more likely than any other, and more likely than all the others combined—to be the subject being sought when a reader enters that ambiguous term in the Search box. If there is such a topic, then it is called the primary topic for that term. If there is .... on one page). An exception may be appropriate when recentism and educational value are taken into account, especially if one of these topics is a vital article. In such a case, consensus may determine that the article should be treated as the primary topic regardless of whether it is the article most sought by users. 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. Tools that may help to support the determination of a primary topic in a discussion, but are not determining factors, include: Incoming wikilinks from Special:WhatLinksHere Wikipedia article traffic statistics Google web, news, scholar, or book searches (NOTE: adding &pws=0 to the google search string eliminates personal search bias) [If there is no clear consensus, or where there is more than one topic that could reasonably be expected to be the target of a particular search term, for example George Clinton, that term should take the reader to a disambiguation page. If there are prolonged or repeated discussions regarding which topic is primary, then the default position should be to use a disambiguation page. If there is reasonable doubt, such that evidence and subtle arguments need to be used rather than common sense, then the default position should be to use a disambiguation page.] The aim is to reduce the likelihood of internal disputes, provide more stability, and provide the reader with the most obvious and least surprising target page. SilkTork *YES! 09:51, 24 March 2011 (UTC) The aim here seems to be to favour one side of a disagreement over the other - saying that if it's not clear which position is right or which is best supported, then we automatically come down on the side of those saying "there is no primary topic". This, to me, seems pretty irrational. Perhaps the current position, where we automatically come down on the side of those supporting the established status quo, is somewhat irrational too, but at least it provides stability and discourages people from continuously proposing unnecessary title changes without any particularly strong reasons. (I.e. the current practice achieves two of the stated aims of the proposal better than the proposal would; and I don't see how the third aim is achieved by the proposal either.) In summary, strong oppose.--Kotniski (talk) 11:06, 24 March 2011 (UTC) Comment I think it's time for SilkTork to drop the WP:STICK. I'm really sorry to say this, but that last RfC was nothing more than a waste of everyone's time. Please, get involved with the disambiguation project first - vote on requested moves, fix some dablinks, get a good solid understanding of how PRIMARYTOPIC works and is implemented - and then try your hand at rewriting policy. --JaGatalk 17:53, 24 March 2011 (UTC) From the above discussion there is no clear consensus on the wording, and so I am looking for consensus on new wording. That is appropriate and how Wikipedia works. Your comment here is not so much about the issue as about me. We can air any concerns you have about me on either mine or your talkpage, as I don't think this is the appropriate place for such matters. Let me know. SilkTork *YES! 12:56, 25 March 2011 (UTC) Oppose due to "If there are prolonged or repeated discussions regarding which topic is primary, then the default position should be to use a disambiguation page." I disagree, continue discussion does not always mean there is no primary topic. --Falcorian (talk) 20:12, 24 March 2011 (UTC) Oppose I see no reason to eliminate the non-exhaustive list of legitimate factors worth considering when deciding on primary topics. Similar to WP:EVENT, no single factor is typically decisive, but having a list is good for consistency purposes. I say this despite thinking the list itself should be amended. --Cybercobra (talk) 23:30, 24 March 2011 (UTC) Oppose. I think the original guidance to specifically consider recentism and educational value is important and should not be removed. We are, after all, an encyclopedia. Kaldari (talk) 20:47, 25 March 2011 (UTC) Oppose "...prolonged and repeated discussions..." can simply be due to persistence by a strongly-opinioned editor, for example. That doesn't indicate that there is no primary topic. Recentism and educational value are both important factors, along with the others that are currently mentioned. Reasoned discussion of each case, based on all factors, is much better than a more rigid approach. First Light (talk) 05:23, 26 March 2011 (UTC)

New proposal - wording to be added after the end of the first paragraph [....linked in combination on one page).]:

There are no absolute rules for determining how likely it is that a specific topic is the one sought by readers entering an ambiguous term; decisions are made by discussion between editors, often as a result of a requested move. Factors that may be considered in discussions include:

• Incoming wikilinks from Special:WhatLinksHere
• Wikipedia article traffic statistics
• Google web, news, scholar, or book searches (NOTE: adding &pws=0 to the google search string eliminates personal search bias
• Recentism
• Educational value, including if one of the topics is a vital article
• Systemic bias
• Base meaning

These are not determining factors, and other factors may be considered on a case-by-case basis. If there is no clear consensus, or where there is more than one topic that could reasonably be expected to be the target of a particular search term, for example George Clinton, consideration may be given to using a disambiguation page instead of a primary topic.

I have looked closely at the previous discussions and attempted to amalgamate the proposals that had support. This discussion/RfC started when the addition of "educational value" to the primary topic section was questioned. Opening up the discussion to the community showed limited direct support for "educational value", though "base meaning" was named often. The wording of the RfC was questioned as it set up "educational value" in opposition to "most likely", when people felt that it might be considered as part of the criteria, so the low direct support for "educational value" might be attributed to the RfC wording. The changes proposed are that "educational value" and "recentism" are now considered as part of the criteria rather than set up in opposition to "most likely". "Base meaning" has been brought in to the list as that appears to have existing support, and systemic bias has been added as a proposal.

Views are welcomed not just on the wording as a whole, but also on items within the wording, such as "base meaning" and "educational value". SilkTork *^YES! 13:20, 26 March 2011 (UTC)

• Question How exactly do you define "base meaning"? Also, what text are you planning to cut? --JaGa^talk 16:29, 26 March 2011 (UTC)

• Actually, could you give us your latest version of PRIMARYTOPIC with strikethroughs, to make the changes clear? --JaGa^talk 16:46, 26 March 2011 (UTC)

Oppose latest proposal on the basis that it does not give guidance, like the present wording does. IE. Is recentism considered pro or con a primary title? Is educational value considered pro or con a popular title? Are the other factors pro or con, when deciding on a title. In order for the guidance to be meaningful, it should guide one way or another: 'If this ... than generally that.' The proposed change gives much less guidance than the present wording.Alanscottwalker (talk) 23:45, 27 March 2011 (UTC)

• Good questions all.

• "Base meaning" was picked up from what people were suggesting in the discussion above. I included it as it appeared a popular choice, though it's not a criteria I support (I'm ambivalent), so I wouldn't be able to give an appropriate rationale for its inclusion or meaning. BFizz was the first to use the term, and used phrases such as "the meaning" and that films, etc should redirect to "the concept behind their name". ƒETCHCOMMS developed this further with: "the original concept from which another subject's name was derived. In the example of "twilight", the book/movie series' title comes from the time period known as twilight". This may also link in with R'n'B's comment that "somebody who types "Sun" in the search box looking for a record company or newspaper and ends up on the article about the star may be disappointed, but they could not reasonably claim to be surprised". Being taken to the primary or base meaning (the record company and the newspapers are all named after The Sun) is somewhat less confusing than being taken to a currently popular meaning.
• The text I proposed replaces all the text: "An exception may be appropriate when recentism and educational value are taken into account, especially if one of these topics is a vital article. In such a case, consensus may determine that the article should be treated as the primary topic regardless of whether it is the article most sought by users. 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. Tools that may help to support the determination of a primary topic in a discussion, but are not determining factors, include: * Incoming wikilinks from Special:WhatLinksHere * Wikipedia article traffic statistics * Google web, news, scholar, or book searches (NOTE: adding &pws=0 to the google search string eliminates personal search bias)"

The proposal brings in the "recentism" and "educational value" wording as part of the criteria rather than treating these separately, and removes the wording that suggests that an exception is used for these criteria, and any implication that these criteria are to be considered differently to other criteria.

The word "tools" has been replaced with "factors".

Some words or phrases are kept, but moved around. Full changes here:

Changes

There are no absolute rules for determining how likely [it is that] a given [specific] topic is to be [the one] sought by readers entering a given [an ambiguous] term; decisions are made by discussion between editors, often as a result of a requested move [[[Wikipedia:Requested moves|requested move]]]. Tools [Factors] that may help to support the determination of a primary topic [be considered] in a discussion[s], but are *not determining factors*, include: Incoming wikilinks from Special:WhatLinksHere Wikipedia article traffic statistics Google web, news, scholar, or book searches (NOTE: adding &pws=0 to the google search string eliminates personal search bias [ *Recentism* Educational value*, including *if one of the topics is a vital article* Systemic bias Base meaning] An exception may be appropriate when *recentism* and *educational value* are taken into account, especially *if one of these topics is a vital article*. In such a case, consensus may determine that the article should be treated as the primary topic regardless of whether it is the article most sought by users. These are *not determining factors*, and other factors may be considered on a case-by-case basis. If there is no clear consensus, or where there is more than one topic that could reasonably be expected to be the target of a particular search term, for example George Clinton, consideration may be given to using a disambiguation page instead of a primary topic.

• I'm not clear where the previous wording gave more guidance than the proposed wording. I included recentism as that was included in the previous wording, though I am unable to give guidance as there is none in the previous wording. It is included as one of the criteria that people may use. How people use the recentism guideline would presumably be down to the individual circumstances. Some editors may regard recentism as a positive asset as it brings in more readers, others may feel it is negative that many readers may wish to read about a current event rather than a more established topic. As with "base meaning", it is not a criteria I support. Indeed, it is a criteria that I am vaguely unsure belongs here, though I can appreciate that it is a criteria that may be brought up in discussions regarding primary topic.

I tried to be as neutral as possible in listing the criteria - to provide a list of criteria that people have named as being useful in discussions. How people use those criteria, and which criteria they use, would depend on circumstances. Generally, consensus is that guidelines reflect current best practise, rather than prescribe new ways of doing something. I am uncertain if all the criteria I have included do reflect current best practise, and a discussion which looks at the wording and picks out specific wording that is acceptable or unacceptable would be useful. The idea being to move forward on improving the current wording which, from the discussions above, is seen as slightly problematic. SilkTork *^YES! 11:26, 29 March 2011 (UTC)

• TLDR (I just came here following an item in WP:CENT), but the point should be something like "If there's an article that more than 50% of people typing [title] in the search box would want to get, then that's the primary topic for [title], otherwise [title] should be a disambiguation page." ― A. di M.​^plé​_dréachtaí 09:47, 5 April 2011 (UTC)
• TLDR again. I think there's an easy way to explain this. If it's an editorial decision, there has to be a shorter way to explain it. This entire guideline is horribly confusing. Shooterwalker (talk) 18:50, 9 April 2011 (UTC)

why r people making comments if hey havent read what its about? —Preceding unsigned comment added by 93.28.215.151 (talk) 21:07, 25 April 2011 (UTC)

Acronyms

I have recently been involved in some editing disputes about the inclusion of entries on disambiguation pages where the DAB page is an acronym and the target page did not mention the acronym. The pages in question are:

• BSB, with a entry for Backstreet Boys
• MJ, with entries for Michael Jackson and Michael Jordon

Details can be found on the Talk page and History for BSB, Backstreet Boys and MJ. In general, I remove entries from a disambiguation page if the target doesn't explicitly mention the acronym, citing WP:DABSTYLE, in particular "Include related subject articles only if the term in question is actually described in the target article." For acronyms, if the target article is a topic with which I am familiar, I might instead add the acronym to the target article if I think it is appropriate (and consistent with other Wikipedia policies). However the above-mentioned entries are causing me some angst. For BSB in particular, my deletion of the entry from the DAB page was reverted several times, on the grounds that it was a common abbreviation for the band. After discussion on the Talk page, I agreed that it was a sufficiently common abbreviation for the band, so (because nobody else had followed my suggestion to) I added the abbreviation to the target article (with a reference) - but that addition was reverted several times as being "unnecessary". I've tried to resolve the problem (of a link on DAB page to an article that doesn't mention the acronym, thus violating WP:DABSTYLE) on the talk pages, but it seems to me that the other parties are not interested in resolving the DABSTYLE violation, instead invoking WP:IAR and/or WP:COMMONSENSE. I know that IAR has its place, but it seems to me that in these cases DABSTYLE does not "prevent us from improving or maintaining Wikipedia" and so should not be ignored. It seems to me that either:

1. The acronym is common and reputably sourced, in which case it should be added to the article page, and the article page included on the DAB page (because readers may use the acronym to search for the topic); or
2. The acronym is not common and/or notable and/or well sourced, in which case it should not be on the article page and the article page should not be on the DAB page.

I would appreciate some feedback from other independent editors on this. Either I am correct, and I need some support to convince editors of the above-mentioned DABs/articles; or I am wrong, and I need someone to explain to me where the flaw in my argument is.

Note that the bone of contention for the above-mentioned articles is not about whether BSB and MJ are valid (common, reliably-sourced) abbreviations - it's about whether BSB and MJ comply with DABSTYLE.

In addition to the specific above-mentioned articles, I have noticed a tendency for disambiguation pages that are acronyms to acquire links to articles that do not mention the acronym (example) or that mention that acronym with no further qualification or reference source (example). I wonder whether we should explicitly add something to WP:DABNOT that DAB pages are not lists of every possible thing that some acronym might stand for. Mitch Ames (talk) 11:12, 17 April 2011 (UTC)

• I go along with your view: either the abbreviation appears in the article and the dab page, or it isn't in the article and isn't in the dab page. End of story. Possibly neither of the Michaels should be listed at MJ - but WP:OTHERSTUFF surely applies to elements within articles as well as to whole articles for deletion. PamD (talk) 13:37, 17 April 2011 (UTC)
  • We don't include references (or external links) on disambig pages precisely because those belong in the article. If the acronym is notable, it should be in the article and cited there. If not, then it has no place in the disambig page. bd2412 T 16:44, 17 April 2011 (UTC)

• I agree with the above. If an abbreviation does not appear in an article, the article should not appear in the dab page. In principle, that would require a reliable source. --Muhandes (talk) 17:32, 17 April 2011 (UTC)

• Well, I completely disagree. First, a DAB is nothing but a glorified redirect anyway, and the standard for redirects is whether they are plausible search terms and helpful to users. There's certainly no rule that they must be mentioned at the target article (e.g., Michael Jordon → Michael Jordan, or, you know, Bi-winning → Charlie Sheen and uncountable others). As with redirects, "[k]eep in mind that the primary purpose of the disambiguation page is to help people find the specific article they want quickly and easily." (MOS:DABENTRIES.)

Second, this complaint is based on a thorough misreading of WP:DABSTYLE. DABSTYLE does not say that no entry should be given unless the exact term appears in the article; it says, "Include related subject articles only if the term in question is actually described in the target article." "Related subjects" are things like the example given: Flag terminology is related to Canton. Michael Jackson is not "related" to MJ; he is an example of MJ. This paragraph simply does not apply here; it and Mitch Ames are talking about two very different things. And I notice that the guide Mitch Ames is relying on is the brief overview DABSTYLE, not the full treatment MOS:DAB, where you would expect to find this important principle spelled out -- if in fact it existed.

Third, if a person comes across easily sourceable DAB links that are not currently reflected in the articles, and is struck with apoplexy at the situation, then WP:SOFIXIT. -- Glenfarclas  (talk) 22:21, 18 April 2011 (UTC)

Glenfarclas, you raise some interesting points, which I'll comment on:

1. There's certainly no rule that [redirect pages] must be mentioned at the target article.
   This isn't necessarily true. After following a redirect, Wikipedia automatically puts "Redirected from [...]" at the top of the page - so the name of the redirect is always shown to the reader. And (contrary to your assertion) WP:R#PLA says (with my emphasis):
   Normally, we try to make sure that all "inbound redirects" other than misspellings or other obvious close variants of the article title are mentioned in the first couple of paragraphs of the article or section to which the redirect goes.
2. ... based on a thorough misreading of WP:DABSTYLE [which] does not say that no entry should be given unless the exact term appears in the article; it says, "Include related subject articles only ...
   This is a fair comment, and I admit that I had not registered the significance of "related subject". Your comment re DABSTYLE vs MOS:DAB is interesting. I suggest that the former (summary) should be a subset of the latter (full rules) - but that appears not to be the case. Perhaps a review of the relative contents of WP:DABSTYLE, MOS:DAB and WP:DDD is required.
3. ...WP:SOFIXIT
   I tried "fixing" both the DAB and the target, and kept getting reverted. Discussions on the talk pages didn't help, which is why I brought it up here.
   Mitch Ames (talk) 12:14, 20 April 2011 (UTC)

• Random articles that happen to have the same initials as an acronym do not belong on a disambiguation page and should be removed. This is different than pointing common misspellings or unique buzzwords at a target. If an editor is struck with apoplexy at that situation and can easily source the acronym, he should also WP:SOFIXIT by adding the information to the article. -- JHunterJ (talk) 01:26, 19 April 2011 (UTC)

In this case, there is nothing to fix. The DAB in question fully conforms to MOSDAB.  Glenfarclas  (talk) 04:27, 19 April 2011 (UTC)

Glenfarclas, I really don't understand what you are trying to achieve or what you are suggesting. Should Mickey Jones also be added to MJ? Should all people be added to the disambiguation articles of their initials? Clearly, something should be added to a DAB page only if it is likely to be used as a target. If something is likely to be used as a target, it should probably be noted in the article, and a source should be provided. How else are you suggesting to determine which initials should be listed and which not? --Muhandes (talk) 22:41, 19 April 2011 (UTC)

Yes, this is true, at least probably in most cases. But it does not amount to what Mitch Ames seems to want, which is a rule stating: "No matter how useful to users, it is strictly prohibited to link to an article from an acronym DAB if the acronym is not explicitly mentioned in the article." That is not presently the rule, and it probably would not be a very helpful rule. So how to determine which people should be listed under their initials? The same way virtually everything else is determined: by seeking and developing consensus among editors, keeping in mind MOSDAB and that "the primary purpose of the disambiguation page is to help people find the specific article they want quickly and easily"! --  Glenfarclas  (talk) 05:30, 20 April 2011 (UTC)

Glenfarclas, I think that your statement regarding ".. what Mitch Ames seems to want ..." is a little unfair. What I wanted - and what I specifically asked for - was feedback on my interpretation of the existing rule; either to confirm it or to point out the flaw in my reasoning if I was wrong. Although most of the comments here seem generally to agree with my interpretation, you have rightly pointed out that I was ignoring the words "related subject" in DABSTYLE, and that the rule in question isn't mentioned in MOS:DAB at all. Mitch Ames (talk) 12:37, 20 April 2011 (UTC)

Fair enough. This whole conversation has somehow seemed a bit snippier than I customarily engage in, so my apologies if I came across a bit abruptly. That said, you did indeed refer to a "DABSTYLE violation" in creating "a link on DAB page to an article that doesn't mention the acronym," and suggested that "other parties are not interested" in fixing it, so I interpreted that more as making a criticism than as offering a suggestion and asking for feedback. --  Glenfarclas  (talk) 20:47, 20 April 2011 (UTC)

• Editors working on dab pages need a set of rules against which to assess which items should be included. The rule stated above (plus an element of WP:IAR occasionally) is reasonable: if the abbreviation is sufficiently well known that a consensus of the editors who work on the article consider it worth mentioning, then it is worth including in the dab page. If not, then not. If you think a particular abbreviation should appear in the dab page, then add it to the article text if it is not present. If it gets challenged and removed by consensus there, then it has no place in the dab page. PamD (talk) 06:51, 20 April 2011 (UTC)

• Disambiguation pages are not guides to slang usage and are not glossaries for all possible manners in which people may refer to subjects. Either there are verifiable indications that a subject is commonly referenced by an ambiguous term in the article on the subject or there is nothing to disambiguate in the context of Wikipedia. older ≠ wiser 12:47, 20 April 2011 (UTC)

I've moved the following comment by PamD|talk]]) 07:50, 24 April 2011 (UTC), from #OMGWTFBBQ below, to avoid unnecessarily splitting the discussion. Mitch Ames (talk) 09:22, 24 April 2011 (UTC)

• This is quite thought-provoking, in the light of the above discussion. I wonder whether the criteria for inclusion in a dab page should include anything from which a redirect could have a valid rationale. Otherwise we could get the situation where:

• An article is created
• A redirect is created, perhaps from an abbreviation not mentioned in the text, because someone considers it a plausible searchterm (see WP:R#KEEP item 5 "Someone finds them useful" as a reason for not deleting redirects).
• A handful of new articles are created (perhaps years later) on totally different topics which are ambiguous with the abbreviation
• A dab page is created
• It is ruled that the redirect to the first article can't be included in the dab page
• So the long-established pathway from abbreviation to article is lost because of the creation of new articles. The redirect has been deleted.

Does this seem reasonable?

• And then perhaps consider the same scenario, without the period where the abbreviation was the only use of that string of letters? Does this argue us into having a more inclusionist approach to abbreviations etc in dab pages? Perhaps "If it could be justified as a redirect, then it can be justified as a dab page entry"? PamD (talk) 07:50, 24 April 2011 (UTC)

I suggest that "someone finds them useful" needs to be qualified or justified. If the original redirected acronym isn't mentioned in the target article, either:

• No other article uses the redirected acronym, ie it is not common, and it's not really "useful" at all - so we should delete it. (Perhaps the onus is on whoever thinks it is useful to justify that claim.) or
• Other articles use the acronym (presumably as a link) and/or it is common and useful, in which case it ought to be mentioned in the target article.

Mitch Ames (talk) 10:01, 24 April 2011 (UTC)

I would use "If it is a redirect, then it can be justified as a dab page entry (even if the ambiguous title does not appear in the article)." Redirects can be vetted through WP:RFD; skirting that to open the door to things that could be redirects but aren't won't reduce the arguments, since the sides will claim that it could or could not be justified as a redirect either. But it is simple enough to create the redirect and dab page entry, and if the redirect is later deleted the dab page entry can be as well. -- JHunterJ (talk) 11:31, 24 April 2011 (UTC)

But it's not so simple: if there's already a dab page at the time that someone feels the need to create a redirect from a term ambiguous with that dab page, then they can't create a redirect but would add it to the dab page, and may find that their addition to the dab page gets deleted. If there wasn't an existing dab page, or any articles at the term in question (the abbreviation or other alternative title), then the editor would be able to create a redirect and defend it at WP:RfD. It just seems a bit inconsistent! PamD (talk) 12:21, 24 April 2011 (UTC)

Well, they could create a redirect and add it to the dab page. The redirect just needs to have the ambiguous title and whatever disambiguator would be appropriate if that title were the title of the article. I have done so in the past. -- JHunterJ (talk) 18:50, 24 April 2011 (UTC)

Further to my comment of 10:01, 24 April 2011 (UTC), we need to balance (reason to keep) "Someone finds them useful" against WP:R#DELETE item 1 "The redirect page makes it unreasonably difficult for users to locate similarly named articles via the search engine." For PamD's scenario, assume the article is Michael Jackson and someone had in the past created a redirect page "MJ". Now there are other possible MJ's so someone has created a disambiguation page "MJ (disambiguation", and Jackson is not the WP:PRIMARYTOPIC. Someone typing "MJ" (looking for something other than the pop star) into the search box would be sent (via the MJ redirect) to "Michael Jackson" and not find what they wanted. Putting {{about}} at the top of that article is one way to solve the problem, but a better solution would be to 'delete the redirect page "MJ", rename "MJ (disambiguation)" to "MJ" and then debate whether Michael Jackson belongs there. If other articles link to "MJ", that's a strong case for adding Jackson to the DAB page - and adding the acronym to Jackson's article. The change from a redirect to a disambiguation page doesn't hurt readers of any articles that link to MJ (intending Michael Jackson), because they can still easily find the pop singer. The diligent reader can improve Wikipedia for all future readers by disambiguating, as suggested by {tl|disambig}} on all DAB pages. Mitch Ames (talk) 13:30, 24 April 2011 (UTC)

I'm sorry I missed this discussion up here; I wrongly assumed that my inquiry at #OMGWTFBBQ was unique and nobody else would have a similar problem. I was obviously as wrong as I could be, based on this discussion. I've skimmed this section, and it seems to be hitting the same points I did, but I'd like to bring a new twist in.

I wanted to look up the computer game Command & Conquer: Red Alert; I eschewed those 28 characters, instead shortening my search by ~86% to "CCRA". I found this redirected to Canada Customs and Revenue Agency, so I created a DAB page to list both articles. Mwtoews (talk · contribs) reverted this saying, "article does not assert that it is also known as CCRA". Okay, so CCRA can be redirected to Canada Customs and Revenue Agency without the article reliably citing that initialism, but a CCRA dab page cannot exist since both articles lack reliable sources for the initialism? Which is the proper SOP? — Fourthords | =/\= | 18:33, 24 April 2011 (UTC)

Right, absent any indication that "Command & Conquer: Red Alert" is known as "CCRA", it's not "Wikipedia ambiguous" with "CCRA". OTOH, if you think it is ambiguous with "CCRA", you can add that info to the article, and (unless or until it's challenged & removed) create or expand the dab page; alternatively, you could create a redirect CCRA (video game) and target the video game article, and link to that in the new or expanded dab page (unless and until an RfD resulted in its deletion). -- JHunterJ (talk) 18:50, 24 April 2011 (UTC)

Canada Customs and Revenue Agency actually mentions "CCRA" in the article (albeit unsourced), whereas Command & Conquer: Red Alert does not, so I would also have deleted the latter from a disambiguation page because I believe that it violates the spirit of the rule that triggered this whole section. I would not have deleted the acronym CCRA from the Canada Customs and Revenue Agency article just because it is unsourced; because I am not knowledgeable on the topic, so basically I'll take the article on faith. Likewise I wouldn't delete "CCRA" from the game article if someone were to add it without a citation. And if someone did add it to the game article I'd then happily accept it on a DAB page. My own personal "rule" - or implementation of the stated rules and spirit of Wikipedia (acknowledging Glenfarclas' disagreement with my interpretation of the DAB rule, but see #Do we need to update the guidelines? below) - is that the DAB page and the target article(s) should be consistent, ie an article linked from a DAB page should include the term/acronym of the DAB page (eg "CCRA") explicitly, but (for topics that I'm not knowledgeable on) I'll take it on faith that the article is "correct", even if unsourced.

So - if you do believe that CCRA is common abbreviation for the game, add it to the article (ideally with a ref), and then turn the CCRA page back into a DAB. If you believe that CCRA is/was not a common acronym for Canada Customs and Revenue Agency, remove the acronym from the article. Not that no other article links to CCRA.

However, I do strongly believe that adding acronyms indiscriminately (ie without reliably sources indicating common usage) runs the risk of violating WP:NOT#DICTIONARY (of acronyms). Mitch Ames (talk) 01:52, 25 April 2011 (UTC)

I don't approve of indiscriminate lists. However, I think the mindless rule of "remove if the initialism is not explicitly named at the top of the current version of the article" is a very bad one. It doesn't even take temporary vandalism into account!

My bottom line is this: anyone who doesn't have enough editorial judgment and common sense to sort out this issue without relying on this very bad rule should find something else to do on Wikipedia, and leave dab page weeding to someone who is competent. WhatamIdoing (talk) 03:02, 25 April 2011 (UTC)

Some editors appear to agree with me. [1][2][3][4][5] Although, I acknowledge that others do not. [6] The fact that we are having this debate - rather than everyone telling me how wrong I am - suggests that "editorial judgment and common sense" are not sufficient; some specific guidelines and/or clarification to the existing guidelines are required. Hence #Do we need to update the guidelines? Mitch Ames (talk) 03:28, 25 April 2011 (UTC)

Yes, I saw those comments.

IMO this always requires judgment, and it may require knowledge of the specific subjects at hand. If it didn't require judgment, we could have a bot do it. A bot is perfectly capable of checking to see whether the named initialism is present in the linked article. If that was even a significant fraction of the goal, we would be doing this automatically, instead of hoping that people will bring intelligence and discretion to the task.

I'm not convinced that it is possible to provide written guidance that will cover enough of the actual issues to be worth writing down, without insulting the intelligence of the people reading it. WhatamIdoing (talk) 05:32, 25 April 2011 (UTC)

• Re CCRA: there are actually 2 other entries which make a dab page reasonable whether or not the game is included, so I've created it. PamD (talk) 08:32, 25 April 2011 (UTC)

Do we need to update the guidelines?

Based on all of the comments so far, I suggest that our disambiguation page guidelines need some work. In particular:

• Should WP:DAB (which contains WP:DABSTYLE), MOS:DAB and WP:DDD be merged?
• Perhaps there should be a clearer separation between content (what can be included) and style (bullet points, sentence fragments, one blue link etc). Most of my references have been to WP:DABSTYLE, but the dispute is actually about content. In addition to content and style, we also need to cover layout, eg as per the current MOS:DAB sections 2, 4, 5, 7.
• WP:DABSTYLE includes the rule Include related subject articles only if the term in question is actually described in the target article. However DABSTYLE refers to the "Main page: MOS:DAB", and that main page does not include the rule. Surely the "main page" should contain all of the rules?
• Should the rule Include related subject articles only if ... be Include entries only if the target article use the term, ie as I had (mis)interpreted it.
• Should we mention acronyms explicitly? Ie something along the lines of: don't include acronyms unless the target article includes the acronym?

Any thoughts Mitch Ames (talk) 13:15, 20 April 2011 (UTC)

• In my view:

(1) DABSTYLE, DDD, and MOSDAB serve different functions and should not be merged. It is an interesting question, though, exactly what is the function of each, or at least of DABSTYLE (which is just a subsection of Wikipedia:Disambiguation) and MOSDAB. Now that I have reflected on it, let me revise the view I intimated above: I would think that WP:DAB is the principal guideline controlling the content of DABs, while MOSDAB should just cover issues of style -- not issues of content, propriety, and the like. The question of which pages should be linked from a DAB is not within the scope of "style." Thus:

(2) There is no need for this rule (about related-subject articles) to be in MOSDAB.

(3) I do not see any need for the new rule that is proposed here. It would be highly discontinuous with our practice in redirects, where the standard for retention amounts to being potentially useful and not causing undue confusion. The probable result of instituting such a rule would just be the deletion of a bunch of useful DAB links. Who benefits? Is there any big issue here that can't be resolved on a case-by-case basis through consensus among editors?

(4) That said, it would not be out of the question to strengthen somewhat the sparse language on this topic in WP:DAB by pointing out that links like these should not be created unless they will be useful, in that the subject is commonly known by its initials or acronym. MJ certainly doesn't need to be filled out with the likes of Mitchell Jenkins and Moira Jones. Isn't this more or less common sense, though? I have not seen signs of some epidemic that needs curing, still less with a straightjacket rule.

(5) P.S., Mitch Ames: if you're looking for a DAB-related style question to go on a crusade about, could you look into getting rid of those utterly pointless and moronic navboxes that we now have on every two-letter acronym? ;) I mean, is there anyone in the world who looks up DL and would conceivably have any reason to "navigate" over to CL, DK, DM, or EL?

--  Glenfarclas  (talk) 20:47, 20 April 2011 (UTC)

I agree that we need to keep "content" and "style" separate. What about "layout" (WP:MOSDAB sections 2, 4, 5, 7)? Is that "content" or "style"? I think style, which is fine, because it's in WP:MOSDAB. But currently WP:DABSTYLE "Page style" (part of WP:DAB) includes a guideline about content, ie "Include related subject articles only if ...". We may disagree about the interpretation of that guideline, or how rigidly we should follow it, but it unambiguously about content not style. Consequently I propose that it be moved (unchanged) out of the Page Style section. Details below in #Proposal to move (but not change the meaning of) "Include related subject articles only if the term in question is actually described in the target article". Please continue any discussion on the topic of that move (with no meaning change) in the separate section below. Mitch Ames (talk) 14:53, 24 April 2011 (UTC)

When you say "I do not see any need for the new rule that is proposed here" I'm not sure whether your referring to "Include entries only if ..." or "don't include acronyms ..." or both. Could clarify please, so that I can address your objections meaningfully. Mitch Ames (talk) 14:57, 24 April 2011 (UTC)

Glenfarclas asks (about not creating acronym links unless they will be useful, ie subject is commonly known by its initials) "Isn't this ... common sense ... I have not seen signs of some epidemic ..."
Sadly not everyone's idea of "common sense" is the same, which is why we have guidelines so that we don't have to have a debate every time the matter comes up. Even if we agree - and include in a guideline - "target should be commonly known by its initials", how do we verify that? Answer: by explicitly requesting that the target article state the fact (with the implied usual criteria for verifiability and notability) - which is basically how I have been interpreting (perhaps incorrectly) the existing "Include related subject articles only if..." and why I'm suggesting that it be changed.

And yes, there is an epidemic. Exampled I've edited recently: MJ, BSB, Pa, NV, Ive, AC. Mitch Ames (talk) 15:18, 24 April 2011 (UTC)

Glenfarclas, you said: ... still less with a straightjacket rule. and ... go on a crusade about ...

Please stop with these negative connotations. I did not request a "straight-jacket rule" - I suggested changes to a rule to clarify it. So far as I can see, those changes are consistent with what most of the contributors to this discussion already think. I'm trying to get some consensus, so we can make some improvements to the guidelines so as to help editors create a better encyclopaedia. Feel free to disagree with my proposals, or point out specific flaws in my reasoning, but please try to be civil about it. Mitch Ames (talk) 15:47, 24 April 2011 (UTC)

I can say I am having the same problem of having to explain over and over again why something should not be on a dab page, and I also see the necessity of a clear rule, in the lines of what Mitch Ames suggests. I'm afraid common sense leaves too much for interpretation in this case. --Muhandes (talk) 18:23, 25 April 2011 (UTC)

Person names dab pages

I'd like to combine the Tommy Smith and Tom Smith dab pages into the existing Thomas Smith page, as by listing these variant names together it is easier for people to find the particular person they want. A number of other dab pages already combine names (e.g. Andrew Williams includes Andy's), and I note from [7] that combining similar names seems to be allowed. Whilst the individual pages do contain 'see also' links for the alternative names, this is not as easy to view compared with if the names are integrated in the main list. Is the overall view that it is reasonable to combine such name dab pages? Eldumpo (talk) 10:10, 20 April 2011 (UTC)

If combining them helps readers find the article they want faster, they should be combined. If separating a disambig would help readers find the article they want faster, they should be separated. Probably people named Tommy or Tom or Thomas are not commonly referred to by the other names, so a full merger may not help -- each reader would have a longer list to wade through. Individual Thomases who are sometimes referred to as Tommy could be included on each page, etc. -- JHunterJ (talk) 11:31, 20 April 2011 (UTC)

Thanks for your response. My query was a general one on merging name dab pages as well as a specific one about Thomas Smith. I don't know the full background as to why separate pages exist for Tom/Tommy/Thomas, but I suspect it may be just down to individual editor preferences at the time they were created, rather than someone specifically deciding that these three names warrant separate pages. I don't really see any difference to Andrew/Andy. If someone is generally known as Tom/Tommy there may still be instances when they are referred to as Thomas as that will generally be their formal name, and some people/sources may refer to them that way. I do appreciate that size of overall article is something to bear in mind, although don't think in this instance it would be too long, and additional sub-divisions can be created if necessary.Eldumpo (talk) 11:47, 20 April 2011 (UTC)

I would think that people referred to as "Tommy Smith" would actually be named "Thomas Smith" but referred to by a nickname. If that is more commonly the case than not, I would recommend merging the pages. bd2412 T 02:31, 21 April 2011 (UTC)

I think JHunterJ is right. If an article is titled with "Tom", "Tommy" or "Thomas", we should assume that is how the person is most commonly known (absent evidence to the contrary) and therefore is how readers are most likely to search for that person. So if readers search for "Tom Smith" they are more likely to find who they are looking for more easily among other Tom Smiths on the "Tom Smith" dab page than on the other two. For those relatively few cases where readers search under the wrong variation, the other two dab pages are referenced. These three dab pages are already long enough that merging them would make things more difficult for most readers. Station1 (talk) 05:33, 21 April 2011 (UTC)

I appreciate the responses to my question, and based on the consensus, I do not propose to merge the pages. However, I am not convinced by some of the arguments. I often get involved in writing/expanding sports biographies, and people are often referred to by different names. Yes, the title of the article should generally follow what they are most frequently called, but there will often be instances of other names (e.g. some publications may use their full birth name). The advantage of listing all the variant names together is that people can see at a glance the various people they may be looking for, e.g. at David Evans people named Dave and Dai are integrated into the list, and people within different sports are listed together, separated by birth year. I would suggest most people searching for a Tom Smith or whatever will already know something about him (i.e. that he is a footballer) rather than just be searching for a Tom Smith who they know nothing about. Anyway, in this instance, the length that the proposed merged article would be, helps to form an argument against. Thanks. Eldumpo (talk) 15:30, 21 April 2011 (UTC)

There are two issues are work: what should an article be titled and what articles might be sought by searching for a particular title. If "Dai Evans" is ambiguous, a short list of the ambiguous entries is useful to a reader searching for "Dai Evans". If some of the Dais are also sometimes referred to as "David Evans", they should also be listed on a disambiguation page for "David Evans". But the disambiguation pages might still be left unmerged, unless the merged page is very short or all of the entries are all sometimes referred to by each name. -- JHunterJ (talk) 17:25, 21 April 2011 (UTC)

Two proposed rule clarifications

I propose the following two points as additions to our current rules governing the designation of disambiguation pages.

1. Compound disambiguation titles should be avoided
2. Lists of lists are not disambiguation pages

Complete explanations are below. bd2412 T 02:25, 21 April 2011 (UTC)

1. Compound disambiguation titles should be avoided

There are a number of compound disambiguation titles unnecessarily combining two distinguishable concepts without actually disambiguating multiple uses of the title phrase.

For example a hypothetical page presented as a disambiguation page at the title, Army boots and pants:

Army boots and pants may refer to:

• Army boots
• Army pants

A list of currently existing pages that present this problem can be found at Wikipedia:Disambiguation/Compound disambiguation titles.

Such a disambiguation page is generally improper because neither individual term on the page would be likely to be referred to by the title term. Furthermore, the title in this case is an unlikely search term, and in most cases is a less likely search term than the individual items listed. Some of the pages presenting this problem are pages that list members of collaborations like Lunt and Fontanne. Unless there are two separate collaborations by this name, it is not ambiguous at all, but is an unambiguous reference to the collaboration itself. It is unlikely that anyone would refer to either individual author by the name of the collaboration, and it is therefore incorrect to say, for example, that "Lunt and Fontane" may refer to Alfred Lunt, individually. There are also many instances of pages like Google and censorship or Jesus and history, containing links to pages broadly relating to the topics but generally not to pages with similar names. Such titles should instead be treated as broad concept articles containing links to the related articles in the context of outlining the concept.

Note that this rule does not apply to titles containing common "Foo and Bar" combinations where that is ambiguously used as the actual name of several different things, as in North and South or Cash and carry.

2. Lists of lists are not disambiguation pages

A number of pages at "List of" or "Lists of" titles are currently tagged as disambiguation pages (I have not yet asked for a list of "list of" disambiguation pages to be generated). In some instances, the contents of these pages are merely lists of lists, and are therefore not truly ambiguous.

For example, a page at the title, List of historic aviaries in Georgia, disambiguating between lists of historic aviaries in the country of Georgia and the state of Georgia, would be a proper disambiguation page. However, a page at the title, List of historic aviaries in Alabama, containing links to separate lists divided into northern, middle, and southern Alabama, or divided into chronological periods in Alabama history, or divided by different types of aviaries, would not actually be distinguishing any ambiguous concepts. Such a page is merely a list of lists, and should be moved to a "Lists of" title and categorized as such, and should not be tagged as a disambiguation page.

responses

• With respect to (1), I do not see much of a problem. Let's look at Lunt and Fontanne. Apparently they did perform together under this moniker, so it is a very plausible search term. But the collaboration does not appear to have been so important to history that it deserves its own standalone article. Why proliferate needlessly when we have perfectly good and not overly long articles on Lunt and Fontanne already? So then the question is what to do with the link Lunt and Fontanne. Deleting it would be counterproductive and decrease the helpfulness of the project. Scrounging up an article to take up space there would be unnecessary cruft. A redirect is the type of thing that would make sense, but there is no "right" target, since Lunt and Fontanne are equally important halves of the duo. As a result, we settle on DABs of this type, and I really don't see that adopting any other option would make this better and not worse.

It is certainly true that DABs of this type which were created as the result of moves to clean up bad page titling (e.g. Ben and Mena Trott) or done to soft-delete an unnecessary article by turning it into a DAB (e.g., Rob and Amber) can probably be removed from the project with no harm. But, actually, I see very few of instances of this type in the list you posted. --  Glenfarclas  (talk) 03:44, 21 April 2011 (UTC)

• It is still incorrect to say that "Lunt and Fontaine" may refer to Alfred Lunt, the individual, because the phrase simply doesn't. However, if the collaboration is notable enough to be "a very plausible search term" then it is likely notable enough to be a short stand-alone article. Compare for example the collaboration between Steven Spielberg and John Williams (who has scored the majority of Spielberg films). It would be absurd to have a disambiguation page titled Steven Spielberg and John Williams (or just Spielberg and Williams directing readers to one or the other article, because despite their many collaborations, they have not been billed as a duo under that name. Conversely, we have a great many articles on short-lived bands having a few notable members, which could under a loose enough standard be framed as disambiguation pages pointing to only the names of the members. For a collaboration such as Lunt and Fontaine, an article would lay out the fact that the phrase refers to the collaboration of the two individuals, and would indicate how and when this collaboration originated, what sort of works they produced, how well received it was, how long it lasted, and how and when it came to an end. That way, we would avoid the problem of people linking to the disambiguation page when what they actually intended to do was to link to a page on the collaboration.
  I agree, there are not many of these pages; the majority of the compound disambig pages are of the utterly useless Army boots and pants type: Cooking and eating utensils; Coptic art and architecture; Dukes in Spain and Portugal; Exergonic and endergonic reaction; North Carolina Cabinet and Council of State; List of motorways and highways of Pakistan (which is also basically a list of lists). If these are allowed, there would be no basis to avoid having a literally infinite number of combinations. Why not Dukes in France and Spain; Coptic pottery and textiles; Wine glasses and coffee mugs (stating that it "may refer to wine glasses or coffee mugs") or Flags of Brazil and Mongolia (stating that it "may refer to Flags of Brazil or Flags of Mongolia). Right now we have zero standards for explaining why such pages should not exist. Let's have some. bd2412 T 11:12, 21 April 2011 (UTC)

• With respect to your first point, we do not let DAB policy dictate what articles should exist and what their content should be. That is the tail trying to wag the dog. DABs exist in service of articles, not vice versa. If someone to believe, "DAB policy requires the creation of X article," he or she has developed an unhealthy view of the importance of DABs.
  
  And with regard to the second point, some judgment has to be exercised. In certain cases, DAB titles are the result of historical accidents. For instance, Dukes in Spain and Portugal was spun off from Duke in 2006. Spain and Portugal are Iberian countries with similar aristocracies and feudal systems, so somebody thought it would be a good idea to put them together. A few months later somebody thought it would make more sense to split the list by country -- so now what to do with the link Dukes in Spain and Portugal? Merely deleting it might break incoming links, and could potentially have adverse GFDL consequences and obscure page history. Normally we create a simple redirect in those cases, but, as with Lunt and Fontanne, there is no single appropriate target. The result is a DAB of this type. Sorry! But this, too, explains why your worry of a proliferation of Dukes of Spain and France or Dukes and Earls of Spain, and whatnot, is unfounded. It is not that somebody got up one morning and said, "Hey, it would be a good idea to create a brand new DAB for these two topics which already have separate articles."
  
  Looking back at page histories you will see more of these situations than you might think. Even though the North Carolina Cabinet and Council of State are two different things, it was spun off from North Carolina under the title North Carolina Cabinet, and eventually renamed North Carolina Cabinet and Council of State because somebody realized that, although half of the people listed in the article, including the Lieutenant Governor, "may be colloquially called 'Cabinet members'," in fact "as elected officials are actually members of the Council of State" and thus technically are not part of the Cabinet at all.
  
  So a little sensitivity to unique page histories will, I think, take you a long way to understanding why we have DABs under some of the titles we do. --  Glenfarclas  (talk) 19:18, 21 April 2011 (UTC)

• Nevertheless, the articles linked on these pages are not ambiguous, and therefore nothing is being "disambiguated". However these pages came to be, and whether they should exist at all, they are not actually disambiguation pages. As for the objection that incoming links may be affected, we're one click away from finding out what links exist, and a few more clicks away from fixing them. Accidents of editing history are no reason to inaccurately characterize pages as serving a disambiguation function. bd2412 T 20:58, 21 April 2011 (UTC)

• Agree completely. --Auntof6 (talk) 18:59, 27 April 2011 (UTC)

• On point 1: What we seem to need in some cases is something mid-way between a disambiguation page and a redirect! I had a look at the list and spotted Dunkeld and Birnam: two separate adjacent places, each with their own WP article, but referred to as a duo in the article Niel Gow and also on their own tourism website. It isn't a standard dab page; it isn't a redirect; if it was called an article it would be a stub and gain a ragbag of tags. It seems a useful entity - not least to stop someone from creating an article for "Dunkeld and Birnam" in future. Is it time to invent something new, or to recognise a different category of disambiguation page: a page which leads from a collective term to the two or more existing articles. A sort of forked redirect. The wording perhaps needs to be changed to "... refers to ... and ...", on one line, rather than the bulleted dab page format? But it needs some sort of tag to indicate that it is an acceptable entity and shouldn't be tagged as unsourced, orphaned, stub, etc. Treating it as a category of dab page would achieve this. PamD (talk) 07:32, 21 April 2011 (UTC)
  • I don't see a problem in having very short articles explain the relationship between conflated places. Some subjects merit encyclopedic coverage, even though there is little to say about them, and having an "article" at Dunkeld and Birnam would at least provide a place to explain why these two things are identified collectively (as opposed to, say, Dunkeld and Iverness), without incorrectly stating that "Dunkeld and Birnam" may be used to refer to Dunkeld alone (or worse yet, to Perth). bd2412 T 13:09, 21 April 2011 (UTC)
    • If you think it was referring to Perth, you may have been misunderstanding the Birnam article's title: Birnam, Perth and Kinross is not a list of 3 places, but a placename with a comma disambiguation to the unitary council area of "Perth and Kinross". I did suggest above that there needed to be a revised wording for these things, to say "refers to A & B" rather than "may refer to A or B". Be that as may, I've converted Dunkeld and Birnam to a brief article but fully expect to see it plastered with stub tags, labelled as an orphan, tagged for deletion as a dicdef, and similar fates from which dab status would have protected it! We'll see.PamD (talk)
• Some of these compound terms seem to have been much requested - see Special:WhatLinksHere/Asset_and_Content_Management. If we removed the "dab page" at that title, sometime someone is likely to "helpfully" create an article at that title. There are a huge number of ghits for the compound phrase. OK, I've re-worded Asset and Content Management so it's correct. But do we leave the dab template? I suggest we should have a new template for cases like this: "This disambiguation page indicates that a compound topic is covered in two or more distinct articles.", keeping the same "if an internal link..." text or modifying it slightly. PamD (talk) 14:26, 21 April 2011 (UTC) On second thoughts I'm not sure whether or not the "Content" part of "Asset and Content Management" is the same as our Content Management, so have undone my changes! PamD (talk) 14:35, 21 April 2011 (UTC)

• • I agree that we may need some kind of intermediate status for articles on unambiguous terms that are commonly used together. Dunkeld and Birnam might be helped with an explanation of why they happen to be referenced in this way, while most adjacent towns are not. We might also use a hidden comment to explain why this is neither a stub nor a proper disambiguation page denoting multiple uses of the same term or phrase. However, this sort of page is less of a concern to me than the Cooking and eating utensils type of page, which probably should not exist at all and should have nothing pointing to it. As for Asset and Content Management, is it possible to write an article covering the concept as a whole? This goes back to the question, why are people, in their searches, thinking of this as a phrase that will take them to a single concept? bd2412 T 14:31, 21 April 2011 (UTC)

• For Cooking and eating utensils, it may be because it's a Mesh heading - see List_of_MeSH_codes_(J01) which is the one real incoming link. PamD (talk) 14:41, 21 April 2011 (UTC)

• ... and it was horribly incomplete even for what it tries to be, so I've added another link to it for now. Seems an area in need of more work - see strange stub at Kitchenware. Hmm, is this something to do with lack of overlap between Wikipedia editors and cookery enthusiasts (though I know one very serious cook who is also a serious Wikipedian...). PamD (talk) 14:51, 21 April 2011 (UTC)

• Why don't we just redirect Cooking and eating utensils to Kitchenware, which would at least encompass all of the topics that should be covered on a page about "Cooking and eating utensils", and then plump the latter one up a bit? bd2412 T 15:40, 21 April 2011 (UTC)

• • • • I suppose every such collocation has some reason for how it originated. Still, the phrase is not an ambiguous phrase in need of "disambiguation". I'd also like to point out that the list as originally generated had several hundred more entries, and I went through all of them to narrow it down to the 100 remaining problematic entries. bd2412 T 14:47, 21 April 2011 (UTC)

Fascinating. While Cooking and eating utensils might not be ambiguous in a typical sense -- there are several articles that might have relevant information for a reader interested in that topic. Perhaps we need a message with text similar to the soft redirect for {{wi}} such as Wikipedia does not have an encyclopedia article for <title>. The following articles may contain related information:". older ≠ wiser 15:27, 21 April 2011 (UTC)

---

I have noticed that all of the discussion on the two points I raised is directed at the first point, relating to compound disambiguation terms. Does anyone have anything to say against the second point, that lists of lists are not properly labeled as disambiguation pages? Cheers! bd2412 T 14:47, 27 April 2011 (UTC)

Dubious "disambig" page

Please see Talk:Disney XD (disambiguation)#Why is this a "disambiguation" page? for discussion and proposed resolution. Note that this page has well over 200 incoming links. --R'n'B (call me Russ) 22:16, 21 April 2011 (UTC)

OMGWTFBBQ

Moved to #Acronyms, as I believe this is an extension of that section. Please continue the discussion under Acronyms. Mitch Ames (talk) 09:22, 24 April 2011 (UTC)

What is the policy, guideline, or SOP regarding disambiguation pages at all-capitalized initialisms (e.g. JOR or UO)? Do entries only belong on such a dab page only if their article has reliably sourced evidence that it is known by that initialism, or can it be listed simply by virtue of having the words that constitute that initialism in the title? For example, the album Murder by Pride has no evidence it is ever referred to as "MBP". Should that article either: not be listed since it's not known by/as "MBP", or remain listed because "MBP" is a plausible search term for the entrant I know the latter is a valid rationale for redirections, but does it hold water for entry on a disambiguation page? Lastly, if I'm asking a stupid question that's found somewhere else, please feel free to WP:TROUT me, and point me in the right direction. Thanks! — Fourthords | =/\= | 05:43, 24 April 2011 (UTC) Short answer: I believe that there should be cited evidence (on the target page) that the acronym is being used. It's not sufficient that it might be used by someone. A much longer "answer" is the discussion in #Acronyms above. Mitch Ames (talk) 06:01, 24 April 2011 (UTC) This is quite thought-provoking, in the light of the above discussion. I wonder whether the criteria for inclusion in a dab page should include anything from which a redirect could have a valid rationale. Otherwise we could get the situation where: An article is created A redirect is created, perhaps from an abbreviation not mentioned in the text, because someone considers it a plausible searchterm (see WP:R#KEEP item 5 "Someone finds them useful" as a reason for not deleting redirects). A handful of new articles are created (perhaps years later) on totally different topics which are ambiguous with the abbreviation A dab page is created It is ruled that the redirect to the first article can't be included in the dab page So the long-established pathway from abbreviation to article is lost because of the creation of new articles. The redirect has been deleted. Does this seem reasonable? And then perhaps consider the same scenario, without the period where the abbreviation was the only use of that string of letters? Does this argue us into having a more inclusionist approach to abbreviations etc in dab pages? Perhaps "If it could be justified as a redirect, then it can be justified as a dab page entry"? PamD (talk) 07:50, 24 April 2011 (UTC)

Proposal to move (but not change the meaning of) "Include related subject articles only if the term in question is actually described in the target article"

 Done

Something that has come out of the #Acronyms discussion above is that currently WP:DABSTYLE "Page style" (part of WP:DAB) includes a guideline about content, not style. Specifically:

Include related subject articles only if the term in question is actually described in the target article. (For example, the Canton disambiguation page legitimately has an entry for Flag terminology.)

I propose moving this guideline out of the "Page style" section. Does anyone have any suggestions where it should be? One option is to invert the logic and include it under "What not to include", probably under a separate sub-heading (as is everything else under "What not to include"), eg (emphasis here is only to show change, it would not be in the article):

Do not include related subject articles unless the term in question is actually described in the target article.

Does anyone have any objections, or suggestions as to exactly where it should be.

Note that the interpretation of this guideline may be the subject of some debate in #Acronyms - my intention here is to move the guideline, but not to change its meaning in any way. Thus I would either keep the wording exactly if possible, or invert it exactly as shown above. Mitch Ames (talk) 14:44, 24 April 2011 (UTC)

In the absence of any objections, I've made the change described above. Mitch Ames (talk) 09:48, 30 April 2011 (UTC)