Wikipedia:Move review/Log/2012 June 18

Big (talk|edit|history|logs|links|archive|watch) (RM)

This request occurred almost two months ago. Complaints were made at Talk:Big and then WT:Disambiguation/Archive 36#the Big mess about the closure. Attempts to discuss this with a closer did not lead to anything. According to one user, there were eight supports and five opposes. Opposers used statistics as proof of primacy. Supporters prove that the film has no long-term significance. I see that WP:PRIMARYTOPIC is used as part of a closure rationale; unfortunately, for me, the guideline (not a policy) is too flimsy to cite, as both sides of rationales have very good reasons to believe one side or another and follow the guideline logically. George Ho (talk) 14:19, 18 June 2012 (UTC)

• Comment by closer. At the risk of being accused of badgering since I'm going to repeat stuff from other RMs, WP:RMCI#Determining consensus: "Consensus is determined not just by considering the preferences of the participants in a given discussion, but also by evaluating their arguments, assigning due weight accordingly, and giving due consideration to the relevant consensus of the Wikipedia community in general as reflected in applicable policy, guidelines and naming conventions." (emphasis added). Applicable guidelines raised in the discussion: WP:PRIMARYTOPIC. If you believe the guideline is too flimsy to cite, please get consensus to have the guideline deleted or de-guidelined. The rationales on both sides were not equally strong. -- JHunterJ (talk) 14:39, 18 June 2012 (UTC)
  • I'll rephrase: citing a guideline as part of a reason to close... It's as strong as both sides, which you consider not strong. Well, to be honest, some people cited the guideline there. Nevertheless, if votes do not count to you, at least I don't think closure rationale is greater than any other. I don't see favoritism in that guideline; just tools to determine what is primary or not, and a rule that there is "no criterion" actually. --George Ho (talk) 15:05, 18 June 2012 (UTC)
    • I'm guessing English isn't your first language, George? I'm having trouble understanding what you wrote here. Powers ^T 14:47, 20 June 2012 (UTC)
      • I'll rephrase for you: Why citing WP:PRIMARYTOPIC? According to current version of the guideline, there is no true criterion of primary topic. Only "usage" and "long-term significance" are discussed more often and used to determine whether either or both are useful for one or two topics (like "My Sister's Keeper"), but there are other criteria not mentioned. Tools can help you decide which of topics with the same name is the most primary, such as Google, statistics, and encyclopedia content. There is nothing else in the guideline that can help supporters or opposers, who merely cited the guideline as an argument, strengthen their arguments, unless they are themselves not good at arguing. I hope my English is very good to you. --George Ho (talk) 15:25, 20 June 2012 (UTC)
• "Supporters prove that the film has no long-term significance." Objection: nothing of the sort was in fact proven. Powers ^T 14:47, 20 June 2012 (UTC)
• Overturn (from consensus to not move to no consensus). Kauffner's analysis, while good, does not overwhelm Dicklyon's, and Diego's input I found persuasive. Many others said sensisble things on both sides. "No consensus" means that WP:RETAIN applies, and that the title Big should cover the film, as it did at 05:49, 13 January 2003, and thus the status quo is correct.

  I think the discussion did show that editors think that the film is the primary topic for the word "big", but it did not show that editors agreed that the section found at WP:PRIMARYTOPIC is a final decider of the question. --SmokeyJoe (talk) 13:36, 21 June 2012 (UTC)

  I forgot to say also, the discussion as a whole seemed to be dominated by a group of regulars divided and at loggerheads, neither side coming close to convincing the other, and with insufficient outsiders commenting in the long threads, I think a "no consensus" close is better justified. It is a fuzzy line though. --SmokeyJoe (talk) 02:16, 22 June 2012 (UTC)

  You see consensus for keeping the primary topic, but not for using WP:PRIMARYTOPIC to identify the primary topic? What should be used as the "final decider" of the question? Whatever that it, we should update WP:PRIMARYTOPIC to reflect (which I've proposed before, most recently at WT:D#The problem with the language here again, and Perth). -- JHunterJ (talk) 15:17, 21 June 2012 (UTC)

  No, I didn't see WP:PRIMARYTOPIC come through as the paramount consideration, though perhaps it was assumed by many and just unclear to non-RM regulars. The "final-decider" should be the formal discussion, explicitly informed by existing guidelines such as WP:PRIMARYTOPIC. I read in the discussion that the PRIMARYTOPIC for "Big" is the film, but there were also others who seemed to say that "Big" is too generic for a title, being a commonword, which maybe could be read as saying that WP:PRIMARYTOPIC might not apply in cases of commonwords. Dunno for sure. --SmokeyJoe (talk) 02:16, 22 June 2012 (UTC)

  Well, the final decider should not be merely "WP:PRIMARYTOPIC" without any reason to cite it because, while the guideline is currently accurate and precise, it might result an inadequate closing rationale. If you want to cite WP:PRIMARYTOPIC, explain how and why this guideline applies. It cannot be just itself; there must be also reasons. --George Ho (talk) 15:26, 21 June 2012 (UTC)

  Well, the final decider has never been WP:PRIMARYTOPIC without any reason to cite it. If you want to have WP:PRIMARYTOPIC not cited in some move discussions, please explain when and why the guideline wouldn't apply. We can then add those caveats to the guideline. -- JHunterJ (talk) 15:29, 21 June 2012 (UTC)

  There must be a counterpart of essay section WP:JUSTAPOLICY, unless we must change the title of an essay and its layout to reflect arguments of all kinds, including deletion and move discussions. --George Ho (talk) 15:37, 21 June 2012 (UTC)

  I don't understand what you're saying. If you're saying that WP:PRIMARYTOPIC doesn't apply to some set of discussions about what article or redirect should be at the name the naming conventions would give to multiple topics, which set of discussions (or what kind of discussions) is that? If it's up to each discussion to determine if WP:PRIMARYTOPIC applies, then again, we have no need for WP:PRIMARYTOPIC as written at all, and it should be rewritten to note that every discussion gets to use its own consensual criteria (or other mechanism) for determining how the readers will navigate from that title to the possible topics they intended. We can keep the list of possible criteria, but we should get rid of any indication that they are guidelines if they're not. Can you give me an example of your suggested counterpart? -- JHunterJ (talk) 16:47, 21 June 2012 (UTC)

  Another thing: SmokeyJoe has a point there. Both sides are more convincing than policy-rationales. There must be an essay counterpart of WP:AADD, which is also an essay, that explains how to avoid such rationales. --George Ho (talk) 15:28, 21 June 2012 (UTC)

  Policy rationales should be avoided? That is your argument? If rationales are not to be based on policy, what do you believe they should be based on? --Born2cycle (talk) 17:25, 21 June 2012 (UTC)

  Well... if policy is cited, policy must be explained in a closing rationale, not just "WP:PRIMARYTOPIC". Get what I'm saying? --George Ho (talk) 17:33, 21 June 2012 (UTC)

  Oh, well, it depends on the context. Where the topic for a given article is clearly primary for the title in question, which is also clearly the most common name for that topic, and that's the sole rationale for someone's support or opposition to a proposed move, I see nothing wrong with just citing WP:PRIMARYTOPIC. It's done all the time, and it is the rationale for probably the majority of our article's titles. Just click on SPECIAL:RANDOM a dozen times and I'll bet you'll find at least six examples of it. --Born2cycle (talk) 04:14, 26 June 2012 (UTC)

• Endorse. I see nothing wrong with closer's decision; certainly nothing cited in this review. Maybe there were a few more supports than opposes, but WP is not a democracy. The lack of long-term significance is true of most movie titles, but it's only relevant to the question of primary topic if there is another use that is more commonly used or more significant. None such was suggested. Even if the film has no long-term significance - which we don't know and it was not proven - it can always be moved in the future. Good decision and kudos the closer for sticking with policy supported by broad community consensus rather than capitulating to a few !vote counts. --Born2cycle (talk) 17:25, 21 June 2012 (UTC)

  The problem with closer's decision is that it was based on Kauffner's opinion that pageview statistics proved the film as a primary topic, but (as was stated in the dicussion) mere logic dictates that pageviews for an article located at the base no-disambiguated title cannot be counted as support for the topic located there (since there's no way to know what article readers intended to visit), but only for the name.

This is not a flaw in the PRIMARYTOPIC guideline (which is too flimsy to neither support nor oppose either side), but in JHunterJ's (and other editor's) way to interpret it at various closures. The only way to show that a topic is primary based on pageviews only, is to move it away from the base title and see if then it's still the one with the most pageviews; but this didn't happen here, so there was no valid rationale for that closing decision. Diego (talk) 20:44, 21 June 2012 (UTC)

• Cannot simply be counted as support, agreed. However, you can take them along with the numbers for the dab and any pages hatnoted from the base name and find that logic might still dictate that even if the visitors to the dab and to the ambiguous topics all cam through the base name, there are still enough left over to support the topic located there. In this case, for example, 39000 (film) - 1000 (album) and 591 (dab) = more than 37000 of 39000 reaching the intended page when they land at Big, or better than 95%. (Probably more than that, since probably some readers are being double-counted in the dab and album counts). So rather than inconvenience 19 readers for the sake of 1, we inconvenience 1 reader for the sake of 19. This is not a flaw in the closure, but a perceived flaw by those who wish fewer titles had primary topics. -- JHunterJ (talk) 21:25, 21 June 2012 (UTC)

  Which brings us again to the argument that The Notorious B.I.G. has three times more visits that Big the film. If the topic is not the one with the most visits at the DAB page, and we don't know what percentage of those visitors were looking for the film, how can visits be used as evidence for primacy of the topic? You're once and again assuming that the 39000 visitors were arriving at the intended page and not simply that they read whatever was presented to them and then left Wikipedia without further navigation, which is a very strong assumption to make.

  Which brings us again to the response to that argument: Is the topic of The Notorious B.I.G. actually ambiguous with "Big"? Absent name collisions, could the article have been titled "Big", or is the topic commonly referred to by the term "Big"? -- JHunterJ (talk) 14:17, 22 June 2012 (UTC)

  And that response was WP:PRIMARYTOPIC-based, how? Diego (talk) 18:32, 22 June 2012 (UTC)

I'm one wishing that primary topics are determined with evidence that they are the topics more looked after, not just the ones that were first created at Wikipedia and happened to occupy the string more looked after. Your argument is effectively using "history of the page at Wikipedia" as the main criterion for determining primary topics. Diego (talk) 05:56, 22 June 2012 (UTC)

Diego argues that, mere logic dictates that pageviews for an article located at the base no-disambiguated title cannot be counted as support for the topic located there (since there's no way to know what article readers intended to visit), but only for the name. Indeed, I'm pretty sure I've made arguments along those lines in the past. But here's the counter-argument: Google. I think we all know that most people get to most WP articles via Google, not via WP search. So page view counts are an accurate measure of how often articles are viewed relative to other articles, and, for the sake of the minority who does use WP search, we should serve them accordingly. --Born2cycle (talk) 23:00, 21 June 2012 (UTC)

Google points to whatever is placed in Wikipedia, so that is a WP:Circular reasoning, and still favoring whatever topic was created at Wikipedia first instead of what is more looked after.

In this particular case, a search for 'big' at Google is returning the Bjarke Ingels Group (BIG) link as the first result. Want to review your argument? Diego (talk) 05:56, 22 June 2012 (UTC)

No, Google points at whatever in WP gets the most hits for the given search term. For example, when I search for "Corvette" on Google, the first WP result is Chevrolet Corvette, not Corvette (the warship), which doesn't show up in the search results until the bottom of the second page. This one example completely refutes your theory, by the way.

If the WP page Bjarke Ingels Group showed up before Big in the Google results, then you might have a point. Their actual website showing up first simply reflects payments they've made to Google.

What we care about in terms of determining relative popularity of WP articles for a given term is the order those WP pages show up in the Google results when Googling for that term, without regard to other results. --Born2cycle (talk) 19:44, 22 June 2012 (UTC)

Another way to test this is to limit Google to WP with "site:en.wikipedia.org". Some examples:

• site:en.wikipedia.org Corvette
• site:en.wikipedia.org big
• site:en.wikipedia.org
• site:en.wikipedia.org apple
• site:en.wikipedia.org peach
• site:en.wikipedia.org banana

Note that for peach and banana the first article in the results is the fruit, but for "apple" it's the corporation. Why is that? Because when people search for "apple" they click on the link to the WP article about the corporation more often then they click on the link to the WP article about the fruit. So this is an excellent tool for determining relative article popularity for a given search term without regard to actual current title of the articles in question.

Same with "big". The movie comes up first because the WP article most often clicked on by people searching for "big" who click on a WP article is the one about the movie. It has nothing to do with that article being the one with title "Big". --Born2cycle (talk) 19:57, 22 June 2012 (UTC)

Using the order of Wikipedia pages at Google? Now that doesn't make any sense. We're supposed to assess the primacy of topics, not of pages. When using Google, we should care which topics Google considers more important, not what Wikipedia pages Google considers more important; of course Google will show first pages that Wikipedia editors place at a prominent place; but again that's circular reasoning. You're seriously suggesting to use decisions by Wikipedia editors to say that a Wikipedia page is important? You don't need Google to do that, but then again don't say it's an objective measure. The Bjarke Ingels Group website shows as the first result, so it should be considered the primary topic according to Google results. But when you have contradicting information about what topics are important, per WP:PRIMARYTOPIC none of them should be primary. Diego (talk) 08:22, 23 June 2012 (UTC)

There is a 1:1 correspondence between a given topic and the WP page about that topic. If you Google search for apple and the WP page for Apple Inc. comes up before Apple, that means users searching for apple click on Apple Inc. more often than they click on Apple. The page hit counts on those articles confirm this. Apple has been viewed 149821 times in 201205. Apple Inc. has been viewed 390961 times in 201205.

You write: "Google will show first pages that Wikipedia editors place at a prominent place". You're not paying attention. Google shows first WP pages that users click on the most - which WP pages Google shows first have nothing to do with which ones editors place at a prominent place. In fact Apple is more prominent than Apple Inc., yet when we search Google with apple, it shows Apple Inc. in the results first. How does your theory account for that?

The relevant search at Google for determining which topic (if any) is primary for "big" is site:en.wikipedia.org big, and that shows that the article about the movie is what people are searching for most often.

You also say: "The Bjarke Ingels Group website shows as the first result, so it should be considered the primary topic according to Google results." Utter nonsense. Considering Bjarke Ingels Group for primary topic is baseless since Bjarke Ingels Group barely gets 1000 views per month, compared to 46734 views for the movie Big. This demonstrates how pointless the suggestion of looking at raw Google results is; results for non-WP sites are skewed by payments corporations make to Google. That's why we need to limit our Google results with the site:en.wikpedia.org key. --Born2cycle (talk) 17:26, 25 June 2012 (UTC)

But there is not a 1:1 correspondence between a given topic and an article title, which is why we have disambiguation at all. I'm paying attention, I simply don't agree with you; please avoid the condescending remarks. If the string "apple" lead to the current Apple Inc article, by your criterion it would be an adequate primary topic, but we have the fruit instead even if it doesn't have more page views; which means that other criteria are also important.

Of course the title without disambiguation will get more hits, but this doesn't say anything about the topic located in it. By favoring pages that historically are ranked higher at Wikipedia relative to other Wikipedia pages, you're using "time of creation and degree of development of the article at Wikipedia" (i.e. what articles were first placed at a prominent location by Wikipedians) as the effective criterion for primacy, not which topic is more likely to be sought after by readers. This is unacceptable navel gazing, in which the criteria for which articles are primary is "those articles that have always been primary".

Your argument relies on page counts as the sole arbiter of primacy, but that's not what we have in the guideline. If a source that's generally counted as valid for determining primacy (Google searches) ranks the Bjarke Ingels Group above the Big movie, that's an argument against the movie being the primary topic - may be a definite argument or maybe not, but it's definitely something to consider; we usually rely decisions on third parties around here. We don't go second-guessing the quality of their search results; we simply take them as one factor among several to consider.

Back on topic to the RM review: there were several arguments in the discussion against using pageviews as support for primacy, and it's easy to find several more. This means that the close was premature since the arguments had not been exhausted, and that there was no consensus to have a primary topic. Diego (talk) 06:10, 26 June 2012 (UTC)

You say you're paying attention and yet your commentary is full of misconceptions, the most glaring of which (this time) is: "Of course the title without disambiguation will get more hits, but this doesn't say anything about the topic located in it". You say this despite the elephant in the room: The title Apple IS without disambiguation and yet does NOT get more hits than Apple Inc.. So it's not true at all that "the title without disambiguation will get more hits" - and, so, the number of hits an article gets has nothing to do with the title, and does say a lot about the topic located at it, especially with respect to other uses of its most common name. Your whole argument -- "This is unacceptable navel gazing, in which the criteria for which articles are primary is "those articles that have always been primary" -- is based on this faulty premise.

There aren't very many examples like apple because we do tend to put the article with the most hits at the undisambiguated title. This is because of WP:PRIMARYTOPIC, but in the case of apple we have the unusual but illustrative situation of a term with two extremely well known uses, and we've decided (at least so far) that the traditional meaning takes precedence in this case, even though the article about the company gets more hits. What's illustrative about this example is that it totally refutes the argument that a high hit count for an article at an undisambiguated title needs to be dismissed or discounted because' it is at an undisambiguated title.

Imagine if one of the previous proposals to move Apple to Apple (fruit) and Apple Inc. to Apple had succeeded - you probably would have argued that its higher page view count and being first in the Google results is because it is at the undisambiguated title. Yet Apple Inc. is not at the undisambiguated title, and it has a much higher page view count, and shows up first in Google results, than the article at the undisambiguated Apple. How do you explain that?

Along the same lines, you're first declaration, "But there is not a 1:1 correspondence between a given topic and an article title, which is why we have disambiguation at all", also ignores that there is a 1:1 correspondence between Apple and the topic that is the common fruit, and between Apple Inc. and the topic that is the corporation. That's a 1:1 correspondence between each of those topics and their respective titles. There is a similar 1:1 correspondence between Big and the movie topic of that article, as well as between every other title on WP and the topic of its article. That is, for every article title (dab page titles and redirects are not article titles), there is exactly one topic associated with it.

You also reveal a lack of paying attention when you make this claim: "Your argument relies on page counts as the sole arbiter of primacy". That's not true. I'm saying it's the primary topic because it has the highest page view count, is the highest in the site:en.wikipedia.org big Google search results, and there is no other candidate use of "big" for primary topic, much less a contender! I underlined that clause because it is fundamental to my argument, which you conveniently ignore. Now, if we had an article about the common adjective "big" like we have one about the common noun "apple", that might be a different story. But of course the adjective is not an encyclopedic topic, so that does not apply here. That addresses your last paragraph too.

I'll just add that the closer understood and appreciated all these points, correctly discounted the arguments made by those who did not understand them, and that's why he properly judged the decision to be no move rather than merely no consensus. --Born2cycle (talk) 21:14, 26 June 2012 (UTC)

• So, my not-vote is to overturn the closure without prejudice to creating a new move request. Diego (talk) 20:46, 21 June 2012 (UTC)
• I agree with SmokeyJoe that it would be more accurate to characterize the result as no consensus to move rather than consensus to not move. But either way, the page should not be moved unless a stronger consensus develops. older ≠ wiser 23:08, 21 June 2012 (UTC)

• While it won't make any practical difference to the final result in the article, it's still in the interest of the new review process to provide a clear answer as to whether the initial closure should be endorsed or overturned. Diego (talk) 08:29, 25 June 2012 (UTC)

• As the distinction makes no difference whatoever in the outcome, I endorse the close. older ≠ wiser 00:45, 26 June 2012 (UTC)