Archives

1, 2, 3, 4, 5, 6, 7, 8, 9, 10 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 21, 22, 23, 24, 25, 26, 27, 28, 29, 30 31, 32, 33, 34, 35, 36, 37, 38, 39, 40 41, 42, 43, 44, 45, 46, 47, 48, 49, 50 51, 52, 53, 54, 55, 56, 57, 58, 59, 60 61

Archives by topic: Common names 1, 2, 3 Naming conflict 1, 2 Precision and accuracy

This page has archives. Sections older than 14 days may be automatically archived by Lowercase sigmabot III when more than 4 sections are present.

There is a move request at Talk:China that could use the opinions of editors involved with Wikipedia naming issues and not invested in intra-China politics. SchmuckyTheCat (talk)

The WP:POVTITLE section of this page currently states:

When a significant majority of English-language reliable sources all refer to the topic or subject of an article by a given name, Wikipedia should follow the sources and use that name as our article title (subject to the other naming criteria). Sometimes that common name will include non-neutral words that Wikipedia normally avoids (Examples include Boston Massacre, Rape of Belgium, and Teapot Dome scandal). In such cases, the commonality of the name overrides our desire to avoid passing judgment (see below). This is acceptable because the non-neutrality and judgment is that of the sources, and not that of Wikipedia editors. True neutrality means we do not impose our opinions over that of the sources, even when our opinion is that the name used by the sources is judgmental. Further, even when a neutral title is possible, creating redirects to it using documented but non-neutral terms is sometimes acceptable; see WP:RNEUTRAL.

My concern is with the first sentence. My reading of "When a significant majority of English-language reliable sources all refer to the topic or subject of an article by a given name" is that it is intended to mean the same as WP:COMMONNAME, but others interpret it more literally, arguing that it means POVTITLE applies only if the name in question is used by "a significant majority" ("like 9 out of 10") of sources, allowing them to disregard this section altogether in any case that doesn't involve a term that is practically the only name used to refer to the topic in question. Surely that cannot be the intent here.

What is a more accurate reflection of how our titles actually use POV terms? How best to say this? Any suggestions?

My interpretation of this section overall is that in order to be neutral when deciding titles, we (ironically) simply do not give neutrality any direct consideration, and only look objectively at usage and commonality (and the other criteria), and follow that. If it's the term indicated by that process, then we use it; if another term is indicated, then we use that, without regard to how "neutral" these terms are... that's how we remain neutral. So, I propose we be more clear about it:

Titles should be decided without regard to the "neutrality" of a given name or term. Simply follow usage and commonality in the sources and use that name as our article title (subject to the other naming criteria). Sometimes a name commonly used in sources will include non-neutral words that Wikipedia normally avoids (Examples include Boston Massacre, Rape of Belgium, and Teapot Dome scandal). In such cases, the commonality of the name overrides our desire to avoid passing judgment (see below). This is acceptable because the non-neutrality and judgment is that of the sources, and not that of Wikipedia editors. True neutrality means we do not impose our opinions over that of the sources, even when our opinion is that the name used by the sources is judgmental. Further, even when a neutral title is possible, creating redirects to it using documented but non-neutral terms is acceptable; see WP:RNEUTRAL.

--Born2cycle (talk) 18:54, 4 August 2011 (UTC)[reply]

The section should reflect what actually happens in the field. Is it the case that titles are decided without regard to neutrality, or do people take neutrality into consideration? Whichever one really happens should be the one documented in policy. -GTBacchus^(talk) 19:19, 4 August 2011 (UTC)[reply]

Let's not fail to be authoritative in our effort to avoid being authoritarian.

According to the reasoning of this section, as I understand it, taking neutrality into account when deciding titles is, ironically, not being neutral. I think explaining this (authoritatively) is the point of this section, not to reflect what really happens, or dictate what should happen. I mean, if people take neutrality into consideration because of mistakenly thinking that they need to do so to comply with our neutrality pillar, isn't WP most improved if we explain why the opposite is true? But perhaps we explain this by being more authoritative and less authoritarian, like this?

Titles are decided neutrally by disregarding the "neutrality" of any given names or terms under consideration. We are being neutral when we simply follow usage and commonality in the sources and use that name as our article title (subject to the other naming criteria). Sometimes a name commonly used in sources will include non-neutral words that Wikipedia normally avoids (Examples include Boston Massacre, Rape of Belgium, and Teapot Dome scandal). In such cases, the commonality of the name overrides our desire to avoid passing judgment (see below). This is acceptable because the non-neutrality and judgment is that of the sources, and not that of Wikipedia editors. True neutrality means we do not impose our opinions over that of the sources, even when our opinion is that the name used by the sources is judgmental. Further, even when a neutral title is possible, redirects to it using documented but non-neutral terms are often created; see WP:RNEUTRAL.

--Born2cycle (talk) 19:38, 4 August 2011 (UTC)[reply]

Your first sentence needs some work ... but I like the rest. Blueboar (talk) 20:28, 4 August 2011 (UTC)[reply]

I disagree with the premise; and with the tactic. If there's a neutral title with significant usage, we should choose that over a non-neutral title with majority usage. That's at least within the range of the present guideline, and within what we actually do, at least sometimes. And proposing to change the guideline, in the middle of an argument that it pertains to, to gain an advantage in that argument, without mentioning that argument, is probably not a good thing to do. Dicklyon (talk) 01:11, 5 August 2011 (UTC)[reply]

Thanks, but I was hoping to have an unbiased discussion about this guideline. I don't see how you see a preference for a less commonly used neutral title over a most commonly used non-neutral title in the current wording. In fact, it says the opposite: "In such cases [common name will include non-neutral words], the commonality of the name overrides our desire to avoid passing judgment". --Born2cycle (talk) 01:35, 5 August 2011 (UTC)[reply]

I agree with Dicklyon, and I really don't like some of the assumptions implicit in the proposed text ("desire to pass judgment" is just one). Tony (talk) 01:51, 5 August 2011 (UTC)[reply]

That "desire to pass judgment" wording is in the current version and is unrelated to what I'm proposing, though of course that could be improved as well. What do you suggestion?

Since you agree with Dicklyon, perhaps you could explain how you see a preference for a less commonly used neutral title over a most commonly used non-neutral title in the current wording, when it seems to say the opposite (as I explained above). --Born2cycle (talk) 02:06, 5 August 2011 (UTC)[reply]

The problem with saying that we should choose "a neutral title with significant usage" in preference to "a non-neutral title with majority usage" is figuring out what "neutral" means. Majority usage is neutral—in the sense of NPOV. An article is neutral when it reflects the balance of high-quality sources, not when it is inoffensive. I think what Dick actually meant is that it would be better to choose an inoffensive title (assuming one exists that has significant usage) over an offensive title that happens to be more common. WhatamIdoing (talk) 03:53, 5 August 2011 (UTC)[reply]

No, that's not what I was thinking, though it could apply to offensiveness as well as to neutrality. B2C wants to restore Pro-life, in preference to the more precise Opposition to the legalization of abortion (not an open RM, just a discusssion on the talk page there). I'm not saying that the latter is a great title, but Pro-life is certainly not a "neutral" description of the topic, which is about the groups that oppose legalization of abortion, under a banner of "pro-life". They are very commonly called "pro-life", perhaps in the US even a majority of the time. But I think that to try to change the guideline to support his position in this argument, without even saying that that's what he's doing, is unacceptable. An alternative more neutral title might put "Pro-life" in quotes, and call them "Pro-life" opposition to abortion, or something like that. As to his twisting my words above, it would be silly to try to explain "how you see a preference for a less commonly used neutral title over a most commonly used non-neutral title in the current wording", since that's not what this section is about; other sections are about neutrality, and this one is about a sort of exception to it, and if he weren't trying to make a stronger exception we wouldn't be having this discussion. Dicklyon (talk) 04:15, 5 August 2011 (UTC)[reply]

Apparently I misunderstood. In regard to what is "within the range of the present guideline", you wrote: "If there's a neutral title with significant usage, we should choose that over a non-neutral title with majority usage. " Doesn't that mean "you see a preference for a less commonly used neutral title over a most commonly used non-neutral title in the current wording"? If your words in green (blue) do not match the meaning of my words in green (blue), that's my mistake. Please interpret them to mean whatever you meant in the respective color. --Born2cycle (talk) 05:07, 5 August 2011 (UTC)[reply]

Yes, you misunderstood me. I took your query about "in the current wording" to refer to the section whose wording you are changing. The preference for neutral is not in that section. Dicklyon (talk) 06:16, 5 August 2011 (UTC)[reply]

So you misunderstood me (you misunderstood "in the current wording" to mean "in the proposed wording" when I meant "in the present guideline"). What did I misunderstand?

Regardless, let's try again. In regard to what is "within the range of the present guideline", you wrote: "If there's a neutral title with significant usage, we should choose that over a non-neutral title with majority usage. " Doesn't that mean "you see a preference for a less commonly used neutral title over a most commonly used non-neutral title in the present guideline? --Born2cycle (talk) 07:29, 5 August 2011 (UTC)[reply]

changes backed out

I have just backed out Born2Cycle's changes of today and yesterday, since they bear specifically on this argument that he is in. I'm in it a bit, too, having expressed my opinion against his, but I'm pretty sure it's not an OK tactic to rewrite applicable guidelines during such a dispute without even letting it be known that that's what you're doing. If there's a consensus to accept his changes, that will be OK, but probably this particular dispute doesn't need to leave it's tracks in the MOS this way. Dicklyon (talk) 04:25, 5 August 2011 (UTC)[reply]

I'm not so stupid as to rely on new language I just added to a policy in a discussion I'm in. But of course being in a discussion often means taking a close look at policy, and taking a close look is when you're likely to find opportunities for improvements. Had I seen this 6 months ago or 6 months from now, I would have made the same edits. I didn't think these clarifications (not changes0 would be controversial, so I did not bring them up here first, like I did this more significant rewording. What specific substantive objections do you have to these edits? --Born2cycle (talk) 05:07, 5 August 2011 (UTC)[reply]

Observation based?

I've got a question for Born2cycle. Have you observed that Wikipedians truly don't use neutrality as a naming criterion? How can you observe something like that? I feel that I've observed that Wikipedians do use neutrality as a naming criterion, which gives it completely equal status with the other naming criteria that are already written down at WP:CRITERIA, which have never been anything more than observations - not rules. What's the basis for your claim that neutrality is not a criterion that Wikipedians use?

I like your use of "authoritative" versus "authoritarian". What makes your claim that neutrality is not a naming criterion authoritative? -GTBacchus^(talk) 06:08, 5 August 2011 (UTC)[reply]

I hoped you would like the "authoritative" versus "authoritarian" approach... I got that from a parenting coach....

The issue of neutrality in deciding titles is confusing because in order to be neutral we need to engage our opinions about the neutrality of the various terms we are considering when we are inventing descriptive titles, but we need to avoid considering our opinions about the neutrality of names and terms when choosing names based on commonality and the other principal criteria. But yes, I've observed Wikipedians use neutrality as a naming criteria even when common names are involved, unfortunately. I've also observed Wikipedians engaged in acts of vandalism, incivility, supporting or opposing articles moves and deletions based on WP:JDLI and countless other unfortunate activities. That doesn't mean we endorse these behaviors in policy, does it?

So, I'm not claiming that neutrality is not a criterion that Wikipedians use. The authoritative claim I'm making is that WP:POVTITLE is correct in essentially saying that Wikipedians who use neutrality as a criterion, usually to avoid some particular name despite it being most common, when deciding titles based on commonality are in conflict with NPOV (though they probably don't realize it). That claim is authoritative because the judgement that a given term should not be used because it's not "neutral" obviously stems from a non-neutral opinion about whether that term should not be used due to neutrality concerns. --Born2cycle (talk) 06:29, 5 August 2011 (UTC)[reply]

In other words, when choosing a title based on usage in sources, neutrality requires us to not think about neutrality, because we're relying on the neutrality judgement of the sources, and if we impose our own opinions about whether this or that is more or less neutral than the other, we're no longer being neutral with respect to the neutrality judgment of the sources.

But when we're inventing a descriptive title, we're not relying on sources, so neutrality requires us to think about neutrality and use our own neutrality judgement so that our titles are neutral. --Born2cycle (talk) 06:38, 5 August 2011 (UTC)[reply]

In the current dispute, B2C is arguing that neutrality concerns can play no part in titling. I can see the merit of trying to avoid judgements based purely on editors' opinions. But if there is RS that raises neutrality concerns, that would be relevant, would it not? (The example here is several RS newspapers declaring a potential common name non-neutral (and to be avoided) in their own stylebooks.). VsevolodKrolikov (talk) 06:42, 5 August 2011 (UTC)[reply]

I'm arguing that neutrality concerns can play no part in neutral title decisions when the decision is based on the neutrality judgment of sources: you just look at usage to decide which is most common, natural, etc. (otherwise, if you start imposing your own neutrality judgments, you're not being neutral in terms of relying on the judgment of the sources).

If you're constructing or inventing a title because it's an artificial topic to which RS don't refer directly, then, yes, you need to judge what is more or less neutral, and relying on what RS say about neutrality regarding the terms and names being considered can certainly be part of choosing a neutral title in such a process. --Born2cycle (talk) 07:03, 5 August 2011 (UTC)[reply]

I am inclined to disagree that policy should encourage articles on philosophical positions to have titles which reflect loaded political rhetoric. BigK HeX (talk) 08:10, 5 August 2011 (UTC)[reply]

Huh? What policy, or proposed policy, is encouraging "articles on philosophical positions to have titles which reflect loaded political rhetoric"? This seems like a non sequitur to me. --Born2cycle (talk) 08:28, 5 August 2011 (UTC)[reply]

Born2cycle, I think I understand your argument. Would you say that a consensus of the Wikipedia community has shown that the best way to be neutral is by not trying to be neutral, or is that your argument? Where has the community said that, "when choosing a title based on usage in sources, neutrality requires us to not think about neutrality, because we're relying on the neutrality judgement of the sources, and if we impose our own opinions about whether this or that is more or less neutral than the other, we're no longer being neutral". You say it's "obvious", but is it supported by consensus?

More: "the judgement that a given term should not be used because it's not "neutral" obviously stems from a non-neutral opinion about whether that term should not be used due to neutrality concerns." To whom is this obvious? If it really is "obvious", then why aren't at least 3 or 4 other people agreeing that it's obvious? I think that, on the contrary, editors are expected to use their own judgment, plus consensus, to determine what is or is not neutral. This doctrine that neutrality can only be achieved by not trying to be neutral seems new and original to me. I don't think I've seen it before. (I may be wrong about that.)

A good sign that you're reflecting consensus is the presence of others supporting your claims. Should we be advertising this discussion at, for example, WT:NPOV? -GTBacchus^(talk) 17:33, 5 August 2011 (UTC)[reply]

In this discussion I'm just rephrasing what WP:POVTITLE already says, has said for a considerable time as far as I know, seems like common sense to me, and I had little if anything to do with writing. At essence it simply means that relying on the judgement of sources about neutrality is being neutral. Can "True neutrality means we do not impose our opinions over that of the sources, even when our opinion is that the name used by the sources is judgmental. " mean something else?

Sometimes that's not possible and we do have to rely on our own judgement about what is neutral (particularly when the topic at issue is "artificial", like "List of X", sources don't refer to it, so we can't look to sources for guidance), but ideal neutrality is simply following the sources, and the closer we get to that, the better we're complying with NPOV. My proposal here in general isn't about changing it to say this, but is about saying what it already says more clearly. It's a clarification edit, not a change edit, so while I wouldn't object to advertising this, it seems like a pretty minor thing to advertise like that.

I have to laugh about the one objection to the proposal above that specifically objects to a phrase that's in the current wording, not in anything I'm proposing to change. --Born2cycle (talk) 19:06, 5 August 2011 (UTC)[reply]

Well, any time people are objecting to an edit, no matter how funny you find the objection, it indicates a lack of consensus, and it indicates that it might be worth bringing more eyes to the situation. It seems very clear to you that this is just a clarification edit, but if that's not clear to other editors, then it's not clear. -GTBacchus^(talk) 19:58, 5 August 2011 (UTC)[reply]

How do you achieve consensus when those objecting are unwilling to take the time and effort to understand the current wording, much less what is being changed and why? --Born2cycle (talk) 20:31, 5 August 2011 (UTC)[reply]

You get more people. That's true for every single disagreement on Wikipedia. Convincing people that they're wrong is extremely rare. Once you've gone 2 rounds, it's not going to happen, so that's when you go get more people. As long as you advertise the discussion neutrally, there's nothing wrong with it, and it's a really, really good idea. Arguing here is pretty pointless, really. Open an RFC, or post a neutral invitation somewhere relevant. -GTBacchus^(talk) 21:48, 5 August 2011 (UTC)[reply]

History of POVTITLE?

It might be good to also examine the history of the section in question. Who added it, when? What discussion was there about it, that might support an interpretation or intent? Is there evidence of a consensus in favor of it, or did it just get stuck in and seemed innocuous enough that nobody objected? Apparently, strengthening it to support B2C's preferred outcome in a dispute is meeting some pushback, so maybe these are worth looking into. Dicklyon (talk) 20:25, 5 August 2011 (UTC)[reply]

Regardless of the history, I really, really want someone who is part of the pushback to explain his or her objection to the current wording (not anything I'm proposing) substantively. In other words, show understanding of what it means, and an explanation of the objection. --Born2cycle (talk) 20:34, 5 August 2011 (UTC)[reply]

My concern with a "follow the sources" type of statement like this one is that while it sounds reasonable, it often gets invoked as a way to ignore or override a significant minority alternative. If "a significant majority of English-language reliable sources" do something, might there not still be a significant minority that do the opposite? If there are two POV terms for something, do we choose one over the other due to its significant majority? Or do we look for an NPOV term instead? Who decides what majority is "significant"? Without us discussing and resolving these things, the section will be pregnant with bad possibilities. So we should see if they have been discussed, and not change it without knowing what we're dealing with. Dicklyon (talk) 22:44, 5 August 2011 (UTC)[reply]

Now there is a jolly good point. Thank you! Now we have something to mull over. --Born2cycle (talk) 23:07, 5 August 2011 (UTC)[reply]

I was the one who originally added the language that is now POVTITLE (back in May of 2010, here is the diff where I added it). The idea was to bring this policy into line with what was said at WP:NPOV. Since then, we have had a few discussions that have resulted in some subtle, but important tweaks (I remember a big discussion sometime around June of 2010)... but the core idea remains the same... When the subject clearly has an established and accepted name, we should use it... even if that name may seem non-neutral. But if there is a fairly equal choice (in terms of commonality) between two (or more) terms, go with the one that is more neutral.

Two examples: 1) We had to decide what term to use as the title for our article about the American colonists who favored independence in the American Revolution... The sources give us several choices... "Patriots", "American Whigs", "Congress Men", "Rebels"... even "Traitors". However, a significant majority of sources (both contemporary and modern, and both American and British) use the term "Patriot". And thus so do we... see: Patriot (American Revolution), even though this may seem non-neutral to some people, especially in the UK and Canada (who might consider the real patriots to be those who sided with their mother country... ie the UK). 2) We had a similar but less clear choice when it comes to our article on the other side of that historical conflict... what should be the title for those American colonists who supported the UK? The two most commonly used terms in sources are "Loyalists" and "Tories". But (importantly) both terms are relatively equal in their commonality. Since "Tory" has negative connotations, we go with "Loyalists" as the more neutral of the two terms. see: Loyalist (American Revolution).

We intentionally left the term "significant" open to some degree of interpretation here. It isn't just about numbers. A 55-45 split in raw numbers may be significant if the 55 are all high quality academic journals and the 45 are all lower quality websites and media outlets. Blueboar (talk) 13:25, 6 August 2011 (UTC)[reply]

Article neutrality in the WP:NPOV is and has been for many years phrased to refer to descriptive titles (as made up by Wikipedians) rather than common names derived from sources. While I agree with much of what Blueboar has written, I disagree with some of the nuances: "even though this may seem non-neutral to somemost people"!!; "Since 'Tory' has negative connotations" Since when? (see how many times the The Daily TorygrphTelegraph uses the term!) I also disagree with two points.

"significant majority of sources (both contemporary and modern, and both American and British)". It is common practice to look at the usage in sources over the last quarter of a century or so. If we do not do that then we end up with names that can be distorted towards copyright expired, simply because many such publications are now accessible on the web, but their usage is archaic. This can also be true of many terms from relatively contemporary modern usage thanks to spin which makes a term used while it was a contemporary issue but was quickly dropped as a term once the dust has settled and the name that enters history is something else (often the winners term) eg "[American] Civil War".

"55 are all high quality academic journals and the 45 are all lower quality websites and media outlets" it depends if the 55 hight quality academic journals are only read by a small number of experts and written for their peer group. If the "all lower quality" incorporates heavy weight magazines and newspapers (eg the Economist and the NYT), if that advise is followed we my end up with an expert's name in violation of "The term most typically used in reliable sources is preferred to technically correct but rarer forms, whether the official name, the scientific name, ...".

-- PBS (talk) 07:48, 7 August 2011 (UTC)[reply]

Thanks for finding that. I don't find any directly relevant discussion, but some related things in June, where there was some pushback on the concept and no clear consensus on how to reflect NPOV in choosing titles. It would seem to need closer examination. And if you look at the campaigns that Born2cycle is presently involved in, it's clear why he wants to strengthen the provision, to give easy answer based on counting sources. I think a more nuanced approach would make more sense. Dicklyon (talk) 02:23, 7 August 2011 (UTC)[reply]

I don't understand the reasoning of choosing C when we have trouble choosing between A and B, but that aside, in cases where there is no significant debate about the most common name, in most contentious cases a good argument can be made for the alternative being proposed as well as the most common name, and often for a 3rd one as well. In those situation there really is no good way to make an objective assessment, which is what makes them so contentious. Because of that, if the approach remains "nuanced" as you say, the situation does not change even after some kind of decision is reached. It seems to me that unless the article settles on the name most commonly used in sources, it is likely to remain contentious, forever, except for periods of relative stability, until it is moved to the most common name. So why not go with the most common name right away? Does it really matter that much that we have, for example, Opposition to the legalization of abortion instead of Pro-life? How is the encyclopedia improved by that? As Blueboar says, the principle is that we go with the most common name even if it seems non-neutral. Why not accept that? Why fight it? How are we helping our readers when they will encounter the most common name in our references (and probably the article content) anyway? How is the encyclopedia improved by this "nuanced" approach? --Born2cycle (talk) 03:10, 7 August 2011 (UTC)[reply]

Pro-life is an American term. In Britain were abortion is not a politically divisive issue (Ie it is one where the majority of the public are happy with the status quoe). The abortion debate is phrased as pro and anti abortion (as it would be for hundreds of other political issues). I would guess if you asked someone in the street what "pro-life" meant they would guess that pro-life mean support for a life terms in prison -- meaning life with no parole. In 2002 the BBC used the phrase along with pro-choice to frame the euthanasia debate. So Pro-life probably should be avoided along the lines of fixed-wing aircraft rather than POV issues. -- PBS (talk) 07:48, 7 August 2011 (UTC)[reply]

Pro-life is an American topic. Except for some awkward efforts to try to make it seem more generic, the article really is about the American movement. --Born2cycle (talk) 07:54, 7 August 2011 (UTC)[reply]

There are a few things in danger of getting mixed together here.

• The general principles for common name (choosing more common over more rarefied or "official", eg Bill Clinton over William Jefferson Clinton) are not the same as the particular principles for choosing a common name that is also perceived to be non-neutral. Much stricter conditions are typically applied for the use of non-neutral titles, as evidenced by the reaction to B2C's recent move request at Climatic Research Unit email controversy.
• POVTITLE provides for situations where NPOV concerns may be over-ridden. B2C is flipping this to say that NPOV concerns cannot be said to exist because general usage in itself more or less defines NPOV. This is not the case. General usage is only one of the criteria. NPOV concerns may still exist; POVTITLE indicates that they may not carry enough weight in the face of overwhelming usage patterns.
• In terms of the frequency of a potential title, there seems to be confusion between the term which has the highest number of uses in any google search (plurality), the term which enjoys more than 50% usage amongst all reasonable common terms (majority) and a term which has overwhelming use (significant majority). This matters when you're trying to run COMMONNAME in general and POVTITLE in particular together. A phrase like "most common" is far too loose (it could be applied to all three situations here) to be used in debate or policy.
• To judge by usage while largely ignoring expert or authoritative opinions regarding a term's neutrality or suitability is effectively Original Research. Google searches done by editors are a helpful guide, not an authority; there's a danger of giving too much authority to google here. The frames of any search influence the outcome far too much. In the case of Female genital mutilation we go, inter alia, with the World Health Organisation's usage. In the case of Climategate, the term itself has received RS criticism for non-neutrality. On B2C's interpretation as applied in the pro-life discussion, the opinions of weighty organisations such as NYT, CNN, CBS and Associated Press should basically not count if the wikipedia-editor-run google search goes marginally against them. But are a wikipedian's googling skills as authoritative? As a result relying highly on editor generated estimates of usage risks breaking both NPOV and OR - which makes TWO of the three core content policies being treated with not enough care.
• In many cases we rely on authority over usage. For spelling of elements, we follow the International Union of Pure and Applied Chemistry, resulting in page titles like Caesium despite "cesium" being more common in google searches.
• Letting general usage guide us effectively without reference to expert and authoritative opinion on terminology ignores the problem of real world attempts to skew debate. An RS documented example is the attempt to eliminate "global warming" in favour of "climate change" by US Republican party spin doctors, when talking about global warming. (Thank Heavens Death panel never took that much hold, but not for want of trying). The reason why we need a more nuanced policy is that the world is rather more complicated.

Wikipedians in general are passionately attached to NPOV. In the past week or so I've seen reactions to B2C's arguments and I think it's fair to say people see his approach as weakening our commitment to NPOV.VsevolodKrolikov (talk) 09:45, 7 August 2011 (UTC)[reply]

It is not our commitment to NPOV that my approach weakens. To the contrary, my approach strengthens that commitment. What it weakens is our ability to rationalize our non-neutral views about which use is "neutral" as being "neutral". The bottom line is that to comply with NPOV we ought to trust the assessment of reliable sources over our own opinions, and apparently not very many people like that. It's understandable that a person would rather be the judge of what is "neutral", but that's contrary to, not consistent with, NPOV. --Born2cycle (talk) 19:02, 7 August 2011 (UTC)[reply]

However this gets worked out folks, the section was very poorly written (like a great deal of the page, I see). I've amended the section: shortened it, improved and clarified the wording, and especially (please note!) avoided a piped link to a mnemonic that reloads the current page. Sheesh! Let's not do that, OK? Foresee the consequences of your editing. Not everyone has an infinite and fast internet account, or privileged insight into the information structure that you have in mind. Noetica^Tea? 09:52, 7 August 2011 (UTC)[reply]

The reaction to Born2cycle's proposals tells us that a lot of people don't understand that COMMONNAME overrules, in a sense, absolute NPOV; it says use the name most people use even if some think the name is not neutral. A lot of the objections show how passionate Wikipedians are about their personal, politically correct POV which they, probably unconsciously, equate with Wikipedia's NPOV policy. But NPOV does not overrule COMMONNAME. --Kenatipo ^speak! 18:38, 7 August 2011 (UTC)[reply]

If it comes down to one policy overriding another, we should look into how we came to have conflicts in policy. It's clear that POVTITLE was added with little discussion, and no consensus; maybe it doesn't really belong in policy. Instead of stregthening it, maybe we should be fixing it to be sure we're not adding something that conflicts with NPOV. Dicklyon (talk) 18:54, 7 August 2011 (UTC)[reply]

Instead of using words like "over-ride" and "overrule", perhaps I should say that COMMONNAME appears to embody a small exception to absolute NPOV, but that the exception is valid because it's Reliable Sources using the name, not Wikipedia. --Kenatipo ^speak! 19:47, 7 August 2011 (UTC)[reply]

One reason I like COMMONNAME is because it leads to using titles that are "practical" and "natural" and "efficient". If it were being followed we'd have Pro-life movement, Climategate and Obamacare; instead we have "Opposition to the legalization of abortion", "Climate Research Unit email controversy" and "National Patient Protection and Euthanasia Control Act" (or some such). To me, those choices are no-brainers. --Kenatipo ^speak! 19:22, 7 August 2011 (UTC)[reply]

I don't see it as an exception at all. There are controversies in the real world about what terms to use to refer to various topics. NPOV indicates that we remain neutral on these controversies, and not take a stand. COMMONNAME gives us a criteria to use to select a term independent of the controversy. COMMONNAME allows us to remain neutral. COMMONNAME is not an exception to NPOV. To the contrary, COMMONNAME allows us to comply with NPOV. Going against COMMONNAME in these cases is going against NPOV as well. --Born2cycle (talk) 19:56, 7 August 2011 (UTC)[reply]

Born2 is correct here. COMMONNAME is not an exception to NPOV; it is an application of NPOV ... it's how we comply with NPOV when it comes to article titles. To invent a name when one exists, or to choose a less common name in preference over a more common name (even if done in an attempt to appear more "neutral") is actually a non-neutral act on our part. Blueboar (talk) 21:31, 7 August 2011 (UTC)[reply]

By the way, I'm starting to think the underlying ideas here have greater application than just titles. I mean, the underlying idea of COMMONNAME -- to follow most common usage in reliable sources -- is useful as a neutral arbiter when trying to decide which of several controversial terms to use to refer to topics in article content as well. The alternative is to use our own judgment based on whatever personal biases each of us brings to the table who-knows-what, which by definition is not being neutral.

That said, this is not currently consensus at WP. To the contrary, people use NPOV all the time to rationalize imposing their own opinion on what they believe to be "neutral" in a given situation. --Born2cycle (talk) 21:53, 7 August 2011 (UTC)[reply]

So you're suggesting that we abdicate our responsibility for NPOV, and just adopt the non-neutral language that is common in the polarized sources on controversial subjects, giving the win to the majority? Dicklyon (talk) 22:03, 7 August 2011 (UTC)[reply]

No, it's the majority of the reliable sources that are giving the win to one side in a given controversy about which term to use (we're neutral), not us giving the win to the majority of the RS. By following COMMONNAME, we're neutrally reflecting the winner selected by the majority of the RS, which is the best way for us to meet our responsibility for NPOV. The alternative is for us to step in and take a side... that would be abdicating our responsibility to NPOV. --Born2cycle (talk) 22:26, 7 August 2011 (UTC)[reply]

Sounds like it's time for an RFC on POVTITLE; given all the arguments it's involved in, I can't see how anyone could argue that it has consensus; I suppose we'll find out. Dicklyon (talk) 22:40, 7 August 2011 (UTC)[reply]

Blueboar, when you say "it's how we comply with NPOV", I'd say it's your opinion of how we should comply (since you wrote it); my point is that it hasn't had much discussion, so can hardly just be accepted as consensus for how we should comply. Dicklyon (talk) 22:03, 7 August 2011 (UTC)[reply]

I'm not arguing that the policy should be changed. I wish it had been followed in the case of Pro-life movement, Climategate and Obamacare. However, it shows that NPOV is not absolute, because it allows non-neutral article titles. A non-neutral title doesn't become neutral just because RS's use it. --Kenatipo ^speak! 22:23, 7 August 2011 (UTC)[reply]

If NPOV were absolute, we would be required to use something like "Opposition to the legalization of abortion" instead of the less neutral but more user-friendly "Pro-life movement". But COMMONNAME shows it's not absolute, and there are good reasons for that. --Kenatipo ^speak! 22:50, 7 August 2011 (UTC)[reply]

Blueboar, if you're arguing that in the absence of bias concerns, COMMONNAME is the best way of following NPOV, I would totally agree. No one here is arguing against that. NPOV is defined as being "fair, proportionate and as far as possible without bias", and clearly COMMONNAME is the proportionate part of that. On the other hand, COMMONNAME (measured by low-threshold google hits) is not the be all and end all of NPOV in article titling - but B2C is effectively arguing it is. It is simply not the case that a term that receives one more google hit than any other individual term is automatically considered to be the only NPOV candidate. For that to be the case we would have to completely ignore anything to the contrary regarding:

• Authoritative naming organisations
• Official usage by international organisations
• Major RS statements on usage and neutrality (such as manuals of style)
• Problems of inflated google hits overestimating accepted usage (verbatim quotes, names of groups, scarequotes), because google doesn't distinguish between a medium's own use and their reporting on other people's uses.
• RS analysis indicating the biased, focussed promotion of one particular term by partisan groups
• Issues of geographical applicability and variability, eschewing internationally acceptable titles that would keep all people happy.
• The quality of the RS using the term: the NYT, UN, and BBC would all count for as much as the Accrington Observer, Neepawa Banner and the Minden Press-Herald - because google doesn't distinguish between them, just as books brought out by OUP would only count for as much as books by Master Books.

And this is clearly not what common practice is, right up to the most experienced and respected editors. We *do* care who uses the term, not just how many people use it (that's why Obamacare is not a title, obviously). This has been clear in every debate I've ever witnessed about naming of articles. It is simply not true that unless we use google hits we have to fall back on our own judgement - something that unfortunately keeps getting repeated here. If we have a lot of RS stating that a term is not neutral, then it's not relying on our own judgement at all. I would be very happy for a sentence stating that editors with neutrality concerns must produce evidence that the term's neutrality is disputed in reliable sourcing, just to make it clear that vague bellyaching or blogosphere chatter isn't admissible. POVTITLE as it stands basically means when usage is clearly very high across all mainstream media and scholarly work, we can basically ignore neutrality concerns; we presume it is the best neutral term from the extensive use, if you want to phrase it like that. What I find interesting is that people don't seem to produce many examples of POVTITLES that we have, and certainly none for recent events. Jack the Ripper doesn't impress me as a particularly controversial choice. (As an aside WP:NPOV#Naming reads a little more generously about the occasional use of descriptive titles to avoid taking sides (a practice that ARBCOM has also recognised as permissible) than the absolutist position one or two seem to take here. I wonder if there has been an unnecessary conflation of neologism and description on this talkpage.)

A final point: Following a basic form of google hits without any right to question also opens up wikipedia to manipulation from political groups. We are a massively important information source, whether we like it or not. if we decide to rely on general google hits, we'll just influence the rhetoric more, not less. Being "on message" with your group's key phrase to get the media to frame stories according to your agenda is part and parcel of politics now. We could just make that worse. VsevolodKrolikov (talk)

I agree with Blueboar's last contribution to this thread. VsevolodKrolikov, it seems to me that you are over emphasising some sections of NPOV and understating the importance of the NPOV#Naming section (should we change that section heading to to "Article titling"?). I think your point about Google hits is a strawman argument, this policy does not encourage the use of Google hits it encourages the use of a survey (and analysis) of reliable sources. As several people has told you this, can you please stop flogging that dead horse?

Reliable sources as added to the policy some years ago (2008?) and was introduced after discussions which brought up the points you are raising about unreliable sources being used to justify article titles in naming debates. Its introduction into this policy page has allowed us to simplify naming convention guidelines that previously were rule based to simulate usage by reliable sources. You brought up climategate as an example. I added a * Comment to the debate at Talk:Climatic Research Unit email controversy#Requested Move that in my opinion nearly all those who opposed the move (up to the time I posted my comment), did not do what this policy and NPOV#Naming requires -- discuss the requested move in terms of what the reliable sources use. Instead they based the justification on rejection of what they thought was a biased name (without backing up their opinion through the use of reliable sources) and in my opinion opinions base on "I do not like it" should be discounted in favour of opinions based on reliable sources and surveys of reliable sources. -- PBS (talk) 01:58, 10 August 2011 (UTC)[reply]

The section WP:POVTITLE has been invoked in several current and recent disputes, and is undergoing discussion above. Is it appropriate policy? Has it been misapplied? Does it conflict with NPOV? Or support NPOV? Does its history suggest that more discussion is needed before treating it as policy? Should it be clarified? strengthened? weakened? Dicklyon (talk) 22:50, 7 August 2011 (UTC)[reply]

• Comment The basic principle seems plenty good to me. However, given the nature of some of the disputes over how best to apply the principle, I would suggest a tweak of some sort. Right now, it starts off with When a significant majority of English-language reliable sources all refer to the topic or subject of an article by a given name… I would suggest something like the following, which has had the word “most” added: When a significant majority of English-language most-reliable sources all refer to the topic or subject of an article by a given name… Something along those lines might avoid editors suggesting Wikipedia adopt the names of what are initially whimsically titled topics that wouldn’t be referred to by those names years later.

  Let’s say, for instance, that the next Lorena Bobbitt-like incident involves her attorney hiding evidence. The whole incident might be widely known for a summer—and capture the public’s imagination during that time—as “Wienergate”; an encyclopedia ought to consider what name things would be remembered as years later, after the “schtick” moniker has worn off.

  In fact, that last sentence (An encyclopedia ought to consider what name things would be remembered as years later, after the “schtick” moniker has worn off) isn’t bad. Maybe someone can put some lipstick on that pig and pass it off as a prom date (candidate for addition to the guideline). Greg L (talk) 00:02, 8 August 2011 (UTC)[reply]

Because Wikipedia is not paper based we can have out cake and eat it. If in the short term the incident is known as "Wienergate" then we can use that name and if over time the common usage moves to a different name then we can change to that. For example if this encyclopaedia had existed in 1964 we could have had an article titled Cassius Clay when the majority of reliable sources started to refer to the boxer as Muhammad Ali the article title could have been changed (WP:NOTCRYSTALBALL). -- PBS (talk) 06:03, 20 August 2011 (UTC)[reply]

In the Request Move discussion initiated by B2C at Climatic Research Unit email controversy, a few editors have mentioned WP:RECENT (and also WP:NOTNEWS, although I don't see how the latter applies to titles). A reference to WP:RECENT might be a nice shade of lipstick. That is, when neutrality concerns arise, we should take care backing a name which is contemporaneous journalistic shorthand. POVTITLE should preferably be backed with longer term usage in the most reliable literature.VsevolodKrolikov (talk) 02:29, 8 August 2011 (UTC)[reply]

The distinction between Reliable Sources and Most Reliable Sources is novel and interesting, but it has no basis in actual practice, so far as I know.

WP:RECENT can apply, but it should not be an excuse to not use the only common name used by reliable sources to refer to the topic, in favor of an invented/contrived one. --Born2cycle (talk) 02:56, 8 August 2011 (UTC)[reply]

There certainly is a distinction often made in practice between, for example, local newspapers, major nationals, and the scholarly press, and between minor academic journals and journals like Nature. There is better and worse RS.VsevolodKrolikov (talk) 03:41, 8 August 2011 (UTC)[reply]

In the dozens hundreds of RM discussions I have been involved in over the years, I don't recall anyone arguing a point based on a distinction in the quality of the sources. That doesn't mean it's never happened and I forgot, but if it does happen, it's very rarely expressed by even one person, much less by enough to establish consensus that should be reflected in policy. --Born2cycle (talk) 03:53, 8 August 2011 (UTC)[reply]

I don't have as much experience in RM discussions, it's true. However, Mass killings under Communist regimes's very painful move from Communist genocide did involve issues of source quality as far as I recall (it was mixed up with an AfD too, admittedly). But then again, in content the quality of RSs certainly is compared. WP:RSN discussions will often make reference to whether a source is better or worse than others. The concept is certainly not alien on wikipedia.VsevolodKrolikov (talk) 04:07, 8 August 2011 (UTC)[reply]

In addition, the policy page here itself says In determining which of several alternative names is most frequently used, it is useful to observe the usage of major international organizations, major English-language media outlets, quality encyclopedias, geographic name servers, major scientific bodies and scientific journals. Which to me is saying that NYT is more important in usage considerations than the South Townsvilleton Gazette and Advertiser. This is what I presume is meant in this case by "most reliable".VsevolodKrolikov (talk) 04:26, 8 August 2011 (UTC)[reply]

Born2cycle, I have to disagree with your claim that quality of sources is an uncommon consideration. Every time we prefer Google Books over ordinary Google, it's because Google Books tends to return higher quality sources, e.g., scholarly works versus blogs. In fact, Born2cycle, in a discussion that you participated in not even a week ago, I read: "NPOV [...] means going with the best sources about the subject, sources like Science." That's Viriditas addressing you, and you replied to him, so I know you read it.

Source quality arises all the time in move discussions, and I find the suggestion that it does not to be very strange. Should I start a list? Obviously the quality of sources matters, and some are more reliable than others. -GTBacchus^(talk) 23:24, 11 August 2011 (UTC)[reply]

I don't think that Google Books is preferred over a general Google search because of the difference in the quality of reliable sources. We choose Google books because it is far more likely to return fewer non reliable sources. A raw Google search is usually next to useless because so much chaff is returned with the wheat that winnowing it out is difficult. Also Google Books has the huge advantage of allowing a date range that allows one to exclude Books older than a certain date, or to do comparisons to see if the ratio of different names is changing over time. -- PBS (talk) 06:03, 20 August 2011 (UTC)[reply]

• Comment It is always possible to be in breach of WP:NPOV even without referring to the practices of the popular press, but the proclivity of the press to sensationalise and to use simplistically the "-gate" suffix clearly does not help us in trying to describe any given event in an objective and encyclopaedic manner. Of course, one understands that such use is driven by the need to be catchy, and the simplistic need to instantly impart to the reader the idea of a scandal of some sort. WP is an encyclopaedia, and its role is to ensure that information is imparted to the reader in an objective manner. The balance to be struck is not to seek to ban occurrences of all such names in the name of NPOV, but to ensure historical events with such "POV" words in their title are undisturbed.

  The problem is that we usually mostly categorise the mainstream press as "reliable sources" although we know their shortcomings – they are owned by a phone-hacker (I jest, but only a little), don't always get their facts right, aren't objective, and almost all drop diacritics (but I digress). Thus seeking to apply that notion of "reliability" to change article titles is so fraught with NPOV issues. For me, as often seen at the NPOV/N, the issue should be the extent to which these can be relied upon to fulfil our objectives; the popular press must seem like the "great unwashed", and ought to be kept at arm's length. When editors attempt to move articles to obviously 'POV' titles claiming a numerically superior popular sources are in support, we must tread carefully. I don't believe that the current wording was meant to be lawyered in the way we are currently witnessing, but I wouldn't want articles necessarily to be moved to euphemistic names, such that "Tiananmen Massacre", as it is almost universally known, ending up at "Tiananmen Square protests of 1989". Nor do I wish to disturb other articles where the event or incident is known by no other name (viz Boston Tea Party, or Nanking Massacre) --Ohconfucius ^¡digame! 03:09, 8 August 2011 (UTC)[reply]

The article is titled the Tiananmen Square protests because the events covered in the article are more than just the massacre. If an article was written just about the mass killings and the usual name for those mass killing was "Tiananmen Massacre" then there would be a justification for that name. -- PBS (talk) 06:03, 20 August 2011 (UTC)[reply]

• Comment. As currently written, I think that POVTITLE is slightly problematic. However, I don't believe that it should be removed altogether; rather, I think it should be rewritten in a more cautious style to discourage its interpretation in absolute terms.

  It does not mesh particularly well with NPOV. The closest justification for it is WP:NPOV#Naming, but that is substantially more nuanced. While this policy (erroneously, in my opinion) justifies itself by asserting that neutrality and opinions asserted by >50% of the sources are one and the same, NPOV describes things as a trade-off between neutrality and clarity ("While neutral terms are generally preferable, this must be balanced against clarity"), implying that editors must judge whether there is a sufficient boost in clarity to justify the loss of neutrality. Moreover, it says that popular but non-neutral names "may" (as opposed to must) be used, and that "The best name to use for something may depend on the context in which it is mentioned; it may be appropriate to mention alternative names and the controversies over their use, particularly when the thing in question is the main topic being discussed."

  It also clashes with WP:RNEUTRAL, which says: "The subject matter of articles may be represented by some sources outside Wikipedia in non-neutral terms. Such terms are generally avoided in Wikipedia article titles, per the words to avoid guidelines and the general neutral point of view policy. For instance the non-neutral expression "Attorneygate" is used to redirect to the neutrally titled Dismissal of U.S. attorneys controversy. The article in question has never used that title, but the redirect was created to provide an alternative means of reaching it because a number of press reports use the term."

  One of the problems, to my mind, is that POVTITLE is often interpreted to mean that the most frequently used name must always be used. I think that it should be rewritten to indicate that it is valid to do so, but commonality is but one of several factors to consider, and editors must use their judgement. Other factors include: what is the quality of the sources that are being considered? What is the relative prominence and neutrality of all candidate titles (if there's basically only one common name, which isn't neutral, then the choice may be more obvious than if there are two common names, one neutral but less frequently used, the other non-neutral but more commonly used)? Have sources themselves have addressed the neutrality of names? How does the choice of name affect clarity? How do redirects and discussion of alternative names affect the outcome of the choice? Jakew (talk) 09:13, 8 August 2011 (UTC)[reply]

• Congratulations, that was very well put! --Ohconfucius ^¡digame! 09:32, 8 August 2011 (UTC)[reply]

At one point (about a year ago), WP:POVTITLE was close to being a word for word copy of what was said at WP:NPOV#Naming. Since then both pages have been edited to the extent that the two policies no longer mesh properly. This is an example of the all too common problem that occurs when multiple policy/guideline pages discuss the same issue... there was a lack of coordination and centralized discussion. May I suggest that we invite the regular contributors to both WP:NPOV and WP:REDIRECT to join this conversation, with the goal of achieving a common understanding and compatible language between policy pages. Blueboar (talk) 14:02, 8 August 2011 (UTC)[reply]

Yes, please do invite the others. And remind of the history; did you add it here to mirror what was there? And did what was there have more of a history of consensus? Dicklyon (talk) 14:46, 8 August 2011 (UTC)[reply]

As I remember, the bulk of discussion took place on this page, but some of that discussion got off topic and did not really factor into the final language, and it was discussed at NPOV as well... It was not really a case of one page mirroring the established language of the other. The topic of common but POV titles was being discussed simultaneously on both pages, but in a semi-coordinated way so they ended up essentially mirroring each other. Most of the discussions occurred around May - June of 2010 if you want to look through the archives.

By the way... I don't think the two policies need to use the same exact language (I don't think they can, since NPOV#Naming talks about more than just Article titles) ... they just need to use compatible language (ie not contradict each other). Blueboar (talk) 15:25, 8 August 2011 (UTC)[reply]

Well, I sure do like the wording better at WP:NPOV#Naming, both now and the May 2010 version; and thanks for fending off Born2cycle's current attempts to rewrite that policy, too. If we could represent that same spirit of balance here, that would be better; that is, moderate the "follow the sources" aspect rather that abdicating NPOV to it as Born2cycle seems to be advocating. Dicklyon (talk) 07:08, 9 August 2011 (UTC)[reply]

That would be a definite improvement. Another important question is, does either policy reflect current practice? Have recent title discussions favoured the POVTITLE model or the more balanced NPOV model, or indeed an alternative model altogether? Are there interesting exceptions from which we can learn? For example, has the community applied additional, currently undocumented criteria in some discussions? My gut feeling (very possibly biased) is that there are enough exceptions to indicate that this policy in particular needs to be updated. Jakew (talk) 10:17, 9 August 2011 (UTC)[reply]

The whole point of moving this page to "Article title" was to make it clear the the title is not a name. The Article title should be the name that most people will search on even if in the opinion of some people it is biased. Who judges whether a name is biased? Who judges when a name move from current to historical? If we have reliable sources that make those judgements for us wonderful, but it not then whether a "name" is biased or non historical is a matter of opinion and as such divisive for the project. Whether the majority of sources use the name "climategate" is quantifiable and not so open to editorial points of view as to whether it is biased. In my opinion this is like the judgement of whether Lech Wałęsa is "correct" or an "eyesore", and is not something we can agree upon, but editors in good faith can agree on whether reliable sources in English usually use Lech Wałęsa or Lech Walensa. Equally we can agree whether a name (biased or not) is the most common usage in reliable English language sources. Better to use a metric that brings a speedy resolution to naming disputes (and integrates with the content policies) than non quantifiable method that leads to disputes based on the opinions of a small number of Wikiepdia editors. -- PBS (talk) 12:03, 9 August 2011 (UTC)[reply]

I don't see that "whole point" on the policy page. If the whole point is to name an article with "the name that most people will search on", why don't we say so? Dicklyon (talk) 14:47, 9 August 2011 (UTC)[reply]

Um... we do... see our criteria of Recognizability and Naturalness. Blueboar (talk) 14:55, 9 August 2011 (UTC)[reply]

Those are some good points to weigh, not "the whole point"; it doesn't say they necessarily outweigh NPOV. Dicklyon (talk) 16:59, 9 August 2011 (UTC)[reply]

All these guides are fine when choosing names that are not subject to NPOV disputes on the grounds that they are biased or disparaging. However, if there are NPOV concerns based on RS, we cannot simply plead google, for three reasons. Firstly, NPOV cannot be wished away, even though some people try to by saying that NPOV is, with various turns of phrase, not really-weally a core policy when it comes to titles. Secondly, we do not, and because this is wikipedia cannot have, a user-generated search-engine based metric that categorically determines anything about real world usage and neutrality, because that would become original research. Who decides the parameters, and who decides what we count? (hits? publications? unique occurrences? books (how much) more than newspapers? All publishers?). These kinds of questions are the subjects of disputes in peer-reviewed research; let's not pretend it's a simple, uncontroversial process. Thirdly, NPOV is clear that being proportionate is not the only consideration. The only way that NPOV is applied differently to titles compared to content is that we cannot fudge by including two (or more) different points of view. POVTITLE is a recognition that NPOV is complicated - sometimes there is a trade-off between being proportionate, fair and as far as possible unbiased.

Anyway, if you want a metric that brings a speedy resolution to disputes, everyone just agree with me. I've just done a search on both my real and user names and the terms "wrong" and "right", and usage suggests that I'm right, in a very real sense.VsevolodKrolikov (talk) 15:40, 9 August 2011 (UTC)[reply]

Comment I would agree with PBS above on this one-it’s really about what the sources say. Too often we use words in our policies and guidelines whose real meanings are inconsistent with the intent they are trying to convey. Three of those words come to mind out of WP:POVTITLE and WP:NPOV#NAMING (from dictionary.com)

• Bias: A particular tendency or inclination, especially one that prevents unprejudiced consideration of a question
• Neutral: Not aligned or supporting any position in a controversy
• Point of view: an opinion, attitude or judgment

Indeed we should strive to make our article titles as neutral as practical and use words that don’t blatantly favor one POV over another or infer a bias that is not fully supported by sources. All sources display some bias and some POV. Neutrality (especially our NPOV policy) can only be approximated based on what the sources say. The Marias Massacre or Baker Massacre was indeed a massacre by the very definition of the term: massacre. It is an unbiased and neutral title because the preponderance of sources calls the event by those names. There is no controversy. There certainly is POV, a POV that most all sources agree that the event was indeed a massacre. Having read POVTITLE and NPOV Naming, I find no issues with them as long as they emphasize that we use article titles that are consistent with the spirit and verbiage of the preponderance of reliable sources. (along with the other naming criteria) I think it says that now.--Mike Cline (talk) 15:55, 9 August 2011 (UTC)[reply]

If the preponderance of sources go with a title containing Words To Avoid without much fuss then I would tend to agree. However, the subtext to this discussion is that there are one or two editors campaigning for a much lower threshold - easily low enough that "one side" would win in a googlefight simply by generating little more than the bare plurality of usage. Perhaps policy should state reasons for concern about neutrality to establish when we look at POVTITLE rather than COMMONNAME.VsevolodKrolikov (talk) 16:04, 9 August 2011 (UTC)[reply]

I just read WP:Words to Avoid and it cannot be interpreted without context. Everyone of the sub-sections talk about words used in a specific context, not words used period. Take the word fundamentalist. WP:LABEL. Its use should be avoided when it arbitrarily characterizes someone or some organization as fundamentalist. However, when it is part and part of the name of an organization (Association of Fundamentalist Evangelizing Catholics) it would be perfectly fine in a title. An alternate title would be inappropriate. Bias, POV and neutrality of words is contingent on context. Whether that is clear or not in the policies is a legitimate question.--Mike Cline (talk) 17:27, 9 August 2011 (UTC)[reply]

In the end, the NPOV title is the title that is used by (high-quality) reliable sources, even if that title includes what some individuals believe are "biased" or "disparaging" terms. We want non-judgmental terms when editors are making up the name, but we can, do, and should choose "biased" and "disparaging" terms whenever good sources agree on that. So: Boston Massacre is the NPOV/neutral title, even though "massacre" is biased; Teapot Dome scandal is the NPOV/neutral title, even though "scandal" is disparaging; and so forth. To name a controversial article that Dicklyon has been involved in, Homosexual transsexual is indisputably the NPOV/neutral title, even though nearly all sources, regardless of POV, agree that the term is highly offensive to many of the people it labels. WhatamIdoing (talk) 16:42, 9 August 2011 (UTC)[reply]

That's an interesting example, but I don't recall discussing it as an article title; it's about the only term used for that category in Blanchard's transsexualism typology, and we shouldn't make up another name for that category. But we don't have an article typically on a category in a guy's controversial typology, so it wouldn't be an appropriate article topic or title. As a redirect, it's OK, but one has to be careful how that redirect shows up in links, since it would obviously be offensive to apply it to people that someone judges to be in that category. Dicklyon (talk) 16:57, 9 August 2011 (UTC)[reply]

In reviewing that, I see that the redirect Homosexual transsexual was previously an article that talked about the controversial term itself: "Homosexual transsexual is a controversial term used by some sexologists...". It doesn't appear that I ever edited that article or its talk page that WhatamIdoing warred on a lot, but I was aware of the controversy (perhaps not that particular article) as reflected on some related pages (many are aware that my friend Lynn Conway was a leading critic of that terminology, and had her WP bio smeared by some of those same sexologists). Dicklyon (talk) 17:44, 9 August 2011 (UTC)[reply]

• Mostly Retain. Using majority (high-quality) sourcing seems to be perfectly in keeping with NPOV. I think the problem is that a number of editors are involved in either articles based on recent news events and probably lack high-quality sorucing or articles that really should only be subsections of existing articles and thus have a dubious appropriateness as standalone topics in an encyclopedia (e.g., Comparison of the health care systems in Canada and the United States, Libertarianism and Objectivism, List of wars between democracies, etc). The first comment (from User:Greg L above) seems quite reasonable to me and his tweaks would probably be helpful IMO. BigK HeX (talk) 02:19, 11 August 2011 (UTC)[reply]

• Comment. I think the current text can indeed contradict NPOV, at least in some cases. Neutrality should not be based only on the judgement of editors, but the current text seems to be claiming that common names are always neutral (and this claim is actually based on editor judgement). If there are reliable sources explicitly stating that a specific name is in some way non-neutral (like pejorative terms), this should also be considered when naming an article. I think these guidelines should clarify, that especially when there are multiple common names (and one of them is identified by sources as being non-neutral), we should try to choose the less controversial common name (even if it is not the most common name). And regarding descriptive neutral titles, I don't think there are major problems using them in some very controversial cases.Cody7777777 (talk) 13:17, 11 August 2011 (UTC)[reply]

See my response below at #Are common names necessarily neutral?. --Born2cycle (talk) 14:37, 11 August 2011 (UTC)[reply]

Comment - Remember that part of WP:NPOV is the concept of due weight. This applies to choosing a title as well. If one name stands out as being used by sources significantly more often than others, then it is WP:UNDUE to use something else.

The problem comes when the sources use several names more or less equally... when there is no clear WP:COMMONNAME. In such situations we are faced with a bit of a dilemma: We can only have one title... and we are supposed to follow the sources, which means choosing one of the names used by the sources ... but the very act of choosing one of them will give the chosen name UNDUE weight over the other names. Woops... Huston, we have a conflict.

Thankfully, WP:Article titles gives us a way to resolve this dilemma... it tells us to fall back on consensus to resolve conflicts. Reaching a consensus is often a messy and contentious process... but despite its flaws, it is still the best (and perhaps the only) way to deal with conflicts. Blueboar (talk) 14:33, 11 August 2011 (UTC)[reply]

• I think the basic advice given is sound, but in light of recent discussions I think it could be made clearer (or maybe it already is elsewhere in the policy) that article titles are ultimately determined by editors and in some (indeterminate) circumstances editors might decide that there are good reasons to prefer a neutral name even if it is not necessarily the most common name. Determining "consensus" in difficult cases might involve one or more RFCs. Policy follows practice and with few exceptions should not be formulated as a club with which to beat horses that have already ran their last race. older ≠ wiser 14:53, 11 August 2011 (UTC)[reply]

Summary

I interpret the remarks above as general support for the concept of POVTITLE, but a desire to move toward a more nuanced and less absolute interpretation, perhaps rewriting it to be more in alignment with careful statements at WP:NPOV#Naming. Anyone disagree? Dicklyon (talk) 17:54, 9 August 2011 (UTC)[reply]

A "more nuanced and less absolute interpretation" effectively means "more vague and indeterminate", thus paving the way for everyone to feel justified regarding their JDLI rationalization, whatever it may be. Exactly how is the encyclopedia improved by having more editors wasting more time arguing pointlessly about titles? PBS's main points were not addressed:

1. The Article title should be the name that most people will search on even if in the opinion of some people it is biased. A "more nuanced and less absolute interpretation" means using a name that is not "the name that most people will search on even if in the opinion of some people it is biased.". Why?
2. Who judges whether a name is biased? Who judges when a name moves from current to historical? Great questions - and I have not seen answers from the "more nuanced" proponents. The answer of course is either a) the predominant used by reliable sources, or b) some group of WP editors. Choosing b) over a) is fundamentally non-neutral, and a violation of NPOV.
3. Whether the majority of sources use the name "climategate" [or any other term] is quantifiable and not so open to editorial points of view as to whether it is biased. Exactly. That's what makes the non-nuanced approached neutral and compliance with NPOV.
4. Equally we can agree whether a name (biased or not) is the most common usage in reliable English language sources. Better to use a metric that brings a speedy resolution to naming disputes (and integrates with the content policies) than non quantifiable method that leads to disputes based on the opinions of a small number of Wikiepdia editors. Right. Asking us to decide, neutrally, whether a given term is used most often is reasonable. Asking us to decide whether a given term is "neutral" is inherently a biased quest fraught with conflict and indecision.

The bottom line is not what the rules say, but what makes the encyclopedia better. Now, one can argue that for every single naming case we should debate which name makes the encyclopedia better, but I suggest the law of diminishing returns applies very quickly in these cases, and the encyclopedia becomes worse to a greater degree from that debate than it could from having that article be at any of the titles being seriously considered.

If we could quantify WP "goodness" in some kind of units - I'd say that in the vast majority of cases the most common name is clearly better. In some cases another name might be better, but, if so, not by much. What we're talking about is whether the encyclopedia is improved by debating the issue in all those cases, and in the marginal ones where the more common name is slightly better (but close enough to debate). I'd really like to see a good argument for how the encyclopedia is improved by all that time and effort (multiplied countless times indefinitely). --Born2cycle (talk) 06:04, 10 August 2011 (UTC)[reply]

Let me address these points in turn:

• A "more nuanced and less absolute interpretation" means using a name that is not "the name that most people will search on even if in the opinion of some people it is biased." More correctly, a more nuanced and less absolute interpretation means that there will be times (probably fairly uncommon) when a name is not that which most people will search for. However, they'll be able find it through searching as a result of a redirect.
• Why? Because NPOV is one of the more important policies at Wikipedia, usually taking precedence over other policies in the event of conflicts. As a rule Wikipedians feel a very strong commitment to it.
• [Re who judges whether a name is biased] The answer of course is either a) the predominant used by reliable sources, or b) some group of WP editors. Choosing b) over a) is fundamentally non-neutral, and a violation of NPOV. Wrong. The predominant term used by reliable sources is not necessarily the most neutral term, especially when controversial issues are involved. It may simply reflect or encapsulate the more popular viewpoint, which is not the same thing at all. NPOV applies to us but not to our sources, and it is unreasonable and unrealistic to expect them to adhere to it, even collectively. As with other issues involved with writing an encyclopaedia that conforms to our policies, the only people able to determine if our policies are met are Wikipedia's editors.
• [Regarding "quantifiable and not so open to editorial points of view as to whether it is biased"] That's what makes the non-nuanced approached neutral and compliance with NPOV. No, it's what makes it easy (hypothetically speaking, if editors were willing to follow it). It almost reduces the article title selection process to an algorithm. I understand that this is what some editors want, and I sympathise with that to some extent. But writing an NPOV encyclopaedia isn't easy. It's hard, and often involves long, tedious, and frequently exasperating debates. And even if you have the very best of intentions, it doesn't work, in Wikipedia's system, to say "here's the formula to use, so stop arguing now", because people will then argue about whether to use that formula or whether to derive their own from first principles.
• [Re "Better to use a metric that brings a speedy resolution to naming disputes..."] Asking us to decide, neutrally, whether a given term is used most often is reasonable. Asking us to decide whether a given term is "neutral" is inherently a biased quest fraught with conflict and indecision. Again, the problem here is that the former is easy, but let's try reductio ad absurdum. Assigning a unique sequential number to each article is even easier (the database already has one). And not writing an encyclopaedia in the first place is a lot easier than that. Neither of these outcomes are remotely desirable, so clearly "easy" isn't enough. We also need "usable" (which the current policy encapsulates fairly well) and "balance with neutrality" (which it doesn't). Jakew (talk) 08:30, 10 August 2011 (UTC)[reply]

• Of course they'll be able to find it via redirect. So what? Making articles easier to find is not why we put articles at "the name that which most people will search for." Among the reasons we do that is so that readers will be more likely to be reassured that they've arrived at the article they're searching for once they get there (principle of least surprise), so that they're more likely to be recognized when viewed in categories, so that editors are more likely to get think links right, etc.
• Your answer is a logical fallacy because it presupposes an answer to perhaps the most fundamental question here: whether following the name chosen predominately by reliable sources can be a violation of NPOV.
• The predominant term used by reliable sources is not necessarily the most neutral term". Strawman. We're not talking about the neutrality of terms, we're talking about who decides which is more neutral. We're talking about the neutrality of views. Remember what the V in NPOV stands for. The issue for NPOV is not whether the result of the name or wording selection is "neutral", but whether the process itself is neutral, thus reflecting a NPOV. We can't expect any one source to adhere to NPOV - but the aggregate of all sources by definition represents a NPOV. That is, neutrality, per NPOV, is about not favoring or disfavoring views represented in reliable sources disproportionate to their representation in the sources. On the issue of which name to use, we are meeting this requirement as best as we can when we use the term used most predominately in the sources. That term might not be "neutral" from the POV of some minority, but by choosing any other term we are disfavoring a more predominate view, and that's contrary to NPOV.
• No, the NPOV encyclopedia you apparently envision is not only difficult, it's impossible. If there are three terms to use, and A is the name used by the predominate view, B is used by the opposing view, and C is not used by either, hardly by anyone, C is not the NPOV choice. By selecting C instead of A we're favoring the anti-A view favored by B. We're picking sides. That's not NPOV. That's not how you write an NPOV encyclopedia - that's how you write an encyclopedia from the anti-predominate-view view.
• Again, it's not about easy (although easy is a benefit of following NPOV). It's about following NPOV - which is about fairly representing the views that they exist, whether those views are "neutral" or not (they rarely are). --Born2cycle (talk) 09:07, 10 August 2011 (UTC)[reply]

Let me respond again:

Firstly, I agree that there are other benefits to using the most common name. Similarly, there are sometimes drawbacks (such as biasing articles towards one point of view). That's why we need to balance the two.

Secondly, I think it's an understatement to say that not everyone shares your belief that the predominant term is, by definition, NPOV. We can't have a serious discussion about policy unless you're at least willing to entertain the possibility that views other than your own might have some validity.

Thirdly, re "We're not talking about the neutrality of terms, we're talking about who decides which is more neutral." Actually, no. The process by which we choose a term is internal to Wikipedia. The outcome is what matters, and that is how the adherence to the "contract" of neutrality between Wikipedia and our readers will be judged.

It's important to understand that we're talking about different kinds of neutrality here. You're talking about viewpoints about which term should be used to describe a subject, and you're inferring those viewpoints from the fact that sources use a term. In a very limited sense, terms can be viewpoints, but often they're used out of convenience or habit rather than conviction. So it's misleading to consider only the frequency with which a term is used. We might get a more accurate idea of those viewpoints by looking at what RS actually say about terminology; that is, do they argue for terms or against them, and do they do so on grounds that are meaningful in terms of NPOV?

There is a more important sense of neutrality, one that I'm talking about, and that's neutrality regarding the subject of the article; that is, whether its use predisposes towards or otherwise favours one viewpoint about the subject or not. Ultimately the purpose of an article is to convey information about a topic to the reader, in a way that adheres to our policies. And it follows from that this is the more important kind of neutrality.

So, should a non-neutral but common term be used? We may decide to use it, on grounds of clarity, but we should think carefully about it and discuss the ramifications of such a decision. We might decide to use a less common name if the cost to neutrality is judged too much. That is consistent with WP:NPOV#Naming.

In terms of your argument about "picking sides", unless we use a purely descriptive title (which is always an option), we have to pick a "side". That's unavoidable and, as such, it means that any title is non-neutral in terms of your viewpoints-about-terminology sense of neutrality. But (as pointed out) in terms of the more important neutrality about the subject, some names are more neutral than others, and so sometimes by choosing one of these names the outcome is more neutral. Jakew (talk) 10:10, 10 August 2011 (UTC)[reply]

1) "Similarly, there are sometimes drawbacks (such as biasing articles towards one point of view)" Choosing a title that is not the most common name for POV reasons is " biasing articles towards one point of view" (the point of view that disagrees with the view that favors the term in predominant use). To defend biasing articles towards one point of view by citing the drawback of biasing articles towards on point of view is the epitome of absurdity.

2) "your belief that the predominant term is, by definition, NPOV". Strawman. I never said that. I don't believe it. That language is not even meaningful. Terms can't be NPOV because terms don't have views. Only people can have views, and NPOV requires us to make decisions from a NPOV - which often results in terms that appear to favor some POV (if that's what you mean by an NPOV term - but I never claimed that follow common name, or NPOV, will result in terms that are not that). Maybe others disagree with that, and I'm willing to discuss that. But I'm not going to waste my time defending a nonsensical position that I don't hold.

3) Outcome-based assessment are doomed to fail - because it's the outcomes about which there is the controversy we're supposed to be neutral about.

4) "terms can be viewpoints" Limited sense or not, absurd! Where's the Mad Hatter? Oranges can be animals! Cockroaches can be furniture! Numbers can smell! Terms can be viewpoints! Meaningful discourse is not possible with such a semantic lottery. Looking at RS regarding what they say about the appropriateness of using terms is only useful if the predominant view is that a given term should not be used, but if that's the case it's highly unlikely, practically impossible, that that term is in predominant use among RS if that's the predominant view about its use.

5) "whether its use predisposes towards or otherwise favours one viewpoint about the subject or not.". Okay, we have some common ground here. Yes, it's possible for a term to predispose towards, or favor, one viewpoint about the subject. We agree there. Where we disagree is whether that means we should not use the term. I suggest that choosing not to use the term because it favors one viewpoint IS favoring another (most likely the opposing/minority view). Yes, the new "neutral" term we come up with might not inherently favor that other view or any other view, but by not using that term that is in predominant use, we are disfavoring the predominant view, and that's a violation of NPOV.

6) Clarity is not the only reason to use the more common term - I do disagree with that implication of WP:NPOV#Naming - and don't believe anyone ever intended it to mean that (clarity is the only reason). The main reason to choose the more common but non-neutral term over any other term is compliance with NPOV - for us to be neutral about which term we use... by following usage in sources, whatever that may be.

7) By choosing a "more neutral" term over the term used predominately in RS you are reflecting a POV that is not "representing fairly, proportionately, and as far as possible without bias, all significant views that have been published by RS", directly in contradiction to the defining first sentence of NPOV! Instead, you are unfairly favoring any view that disagrees with the predominate view, you are reflecting those contrary views disproportionately and thus with a bias disfavoring the predominate view.

According to your interpretation of NPOV and reasoning we should move Gringo to Foreigner, Redneck to Uneducated poor farmer, and Spic to redirect to [{Latino]]. It makes no sense. --Born2cycle (talk) 04:45, 11 August 2011 (UTC)[reply]

I'll keep this brief, because we've argued enough in a "summary" section already. You've contradicted yourself in points 1 and 5. Re point 2, I don't intend to argue; certainly the impression that you give is that you believe the two to be one and the same. Point 3 is the important one, and you're wrong, because the outcome is the visible product of the encyclopaedia: it's where NPOV matters most. Furthermore, your assumption that there's a "controversy" about the title requires some justification: it may not necessarily be the case. Nor is it necessarily the case that we favour an opposing viewpoint by using a less common term: we might (as in "Climatic Research Unit email controversy") favour a descriptive title that doesn't favour either "side". Regarding terms being viewpoints, I'm just trying to understand and acknowledge the validity of your point of view; if you respond with ridicule I think I'll decline to bother. Regarding point 6, NPOV itself makes it very clear that you're wrong: we're choosing a non-neutral title because we're trading neutrality against clarity. Finally, in your point 7 you seem to contradict your point 4. You argued in your point 4 that terms aren't viewpoints, but now you say "you are unfairly favoring any view that disagrees with the predominate view". If by "predominant view" you mean "predominant term", then aren't you contradicting yourself? If, on the other hand, you're making a distinction between terms and views, then how do you propose to tell the difference given that you seem to reject using explicitly stated views? Can you propose a test by which we can tell the difference? Jakew (talk) 08:45, 11 August 2011 (UTC)[reply]

The reason I started the RFC was to see if Born2cycle's interpretation had much support among other editors. My impression is that it does not. My summary was intended to draw out more clarity about that. So instead of more dead-horse beating by him, are there others who take a similar position? Or should we just move forward on writing a more moderated version, and consider his move toward a more strict version closed? Dicklyon (talk) 14:28, 10 August 2011 (UTC)[reply]

I think Born2cycle's basic interpretation is essentially correct, but he takes it a step too far in application. Blueboar (talk) 14:42, 10 August 2011 (UTC)[reply]

Since you wrote the section in question, would you like to take the lead on a modest rephrasing that will make it clear to him that balance is required in applying the various provisions related to NPOV? Dicklyon (talk) 02:36, 11 August 2011 (UTC)[reply]

Dicklyon - you keep saying this but none of your statements (or those of anyone else who makes similar statements about me) indicate an accurate understanding of my interpretation and position. When you say your impression is that there is not much support for my interpretation, please explain very clearly what exactly you mean by that. If you would use my actual words from here or from any other discussion about this, that would probably make it less likely to be a strawman. Thanks. --Born2cycle (talk) 04:45, 11 August 2011 (UTC)[reply]

Your interpretation, as I understand it, is that using the "most common" term for a topic is necessarily neutral; do you need diffs to where you said so? An example is in your proposal at WT:Article titles#POVTITLE, in which you want to rewrite POVTITLE to say "...Simply follow usage and commonality in the sources and use that name as our article title..."; where considerations of NPOV are involved, reality should not be simplified that far. Futhermore, you misapply even that suspect concept on RM such as those at Crepe by asserting what is most common without credible evidence (your words and my comments on them can be found there; let me know if it's worth the trouble of me finding diffs). Dicklyon (talk) 17:52, 11 August 2011 (UTC)[reply]

Blueboar please clarify what you mean by me taking my essentially correct interpretation too far in application. Again, actual words of mine would be helpful. Thanks. --Born2cycle (talk) 04:45, 11 August 2011 (UTC)[reply]

Are common names necessarily neutral?

I think some people are conflating the following two issues, thinking an answer of "yes" to the first means the answer to the second is also "yes":

1. Is the name, term or phrase neutral?
2. Is using the name, term or phrase in conflict with WP:NPOV?

For example, above, Cody7777777 (talk · contribs) says, "the current text [of WP:POVTITLE] seems to be claiming that common names are always neutral, ", which is the basis for the claim that "the current text can indeed contradict NPOV, at least in some cases. ". I don't see anything in the text that means or even implies that. In fact, the text contains a number of statements that are clearly premised on the idea that common names often are not neutral (though using them is still fully in compliance with WP:NPOV):

• "Sometimes that common name will include non-neutral words that Wikipedia normally avoids".
• "The commonality of the name overrides our desire to avoid passing judgment".
• "True neutrality means we do not impose our opinions over that of the sources, even when our opinion is that the name used by the sources is judgmental." [my emphasis].

I don't believe the text, or anyone here discussing it, is claiming that common names necessarily do not violate WP:NPOV because the names are neutral. Using them does not violate NPOV not because the names neutral (which they're not or we wouldn't be talking about not using them), but because editors are being neutral when deciding to use them, and in fact are in compliance with NPOV because by selecting the name most commonly used in relevant reliable sources they are "representing fairly, proportionately, and as far as possible without bias, all significant views..." In fact, choosing any other name would be not "representing fairly and proportionately all significant views as far as possible", because using the most common name would be going further towards doing that.

In short, we're talking about cases where the most common name is not neutral. "In such cases, the commonality of the name overrides our desire to avoid passing judgment." That is, per WP:NPOV the commonality of the name overrides our desire to use some other name that is more "neutral". Well, it overrides that desire in most of us. Others seem to have some difficulty with this subduing that desire. --Born2cycle (talk) 14:37, 11 August 2011 (UTC)[reply]

Thank you for the clarifications, but nonetheless there are still problems. The claim that using non-neutral common names as titles, does not violate NPOV, is based on editor judgment (it is your opinion, and other editors, and also readers, can see things differently). It can also be argued, that when we choose a common name, which has been explicitly described (or judged) as non-neutral by sources, we adopt its (non-neutral) point of view (and to me this looks like common sense, and it is possible that many readers will also see it this way). If the most common name does not have any problems, it would obviously be undue weight to not use it. However, commonality is not the only criterion (the other criteria should also have weight), and since NPOV is considered a fundamental principle and pillar of Wikipedia, I think the neutrality of the name should have more weight, if it conflicts with commonality. Cody7777777 (talk) 18:17, 11 August 2011 (UTC)[reply]

That helps; thanks.

Yes, it all comes down to the question of whether using non-neutral common names as titles violates NPOV, and that is a matter of opinion. And since that is the matter of opinion at issue, you can't say giving NPOV the considerable weight it deserves as a pillar favors one side more than the other. Both sides are fully compliant and consistent with NPOV given their respective answers to this key question.

So we need to discuss on why and how using non-neutral common names as titles does or does not violate NPOV, because that's where the disagreement lies. --Born2cycle (talk) 18:34, 11 August 2011 (UTC)[reply]

Regrettably in many cases, this will remain just a matter of editor opinion, and I do not really expect a full agreement will be reached about this. However, in cases where there are reliable sources which explicitly claim that the use of a specific name is non-neutral, then I think its usage by Wikipedia can also be considered non-neutral (at least, unless there are contradicting sources, claiming the use of the specific name is neutral). Cody7777777 (talk) 13:07, 12 August 2011 (UTC)[reply]

An initial rewrite

The current text of the "Non-neutral but common names" reads:

• When a significant majority of English-language reliable sources all refer to the topic or subject of an article by a given name, Wikipedia should follow the sources and use that name as our article title (subject to the other naming criteria). Sometimes that common name will include non-neutral words that Wikipedia normally avoids (Examples include Boston Massacre, Rape of Belgium, and Teapot Dome scandal). In such cases, the commonality of the name overrides our desire to avoid passing judgment (see below). This is acceptable because the non-neutrality and judgment is that of the sources, and not that of Wikipedia editors. True neutrality means we do not impose our opinions over that of the sources, even when our opinion is that the name used by the sources is judgmental. Further, even when a neutral title is possible, creating redirects to it using documented but non-neutral terms is sometimes acceptable; see WP:RNEUTRAL.

My first attempt to rewrite this to be consistent with WP:NPOV#Naming is as follows:

• Non-neutral titles are sometimes acceptable, as clarity must be weighed against neutrality. When a significant majority of English-language reliable sources all refer to the topic or subject of an article by a given name, Wikipedia should may follow the sources and use that name as our article title (subject to the other naming criteria). Sometimes that common name will include non-neutral words that Wikipedia normally avoids (Examples include Boston Massacre, Rape of Belgium, and Teapot Dome scandal). In such cases, the commonality of the name overrides our desire to avoid passing judgment (see below). This is acceptable because the non-neutrality and judgment is that of the sources, and not that of Wikipedia editors. True neutrality means we do not impose our opinions over that of the sources, even when our opinion is that the name used by the sources is judgmental. The best name to use for something may depend on the context in which it is mentioned; it may be appropriate to mention alternative names and the controversies over their use, particularly when the thing in question is the main topic being discussed. Further, even when a neutral title is possible, creating redirects to it using documented but non-neutral terms is sometimes acceptable; see WP:RNEUTRAL. See also WP:NPOV#Naming.

(All changes are shown in red.)

These are just my initial thoughts, and I'd be surprised if they're right the first time. I've taken some of the new language straight from NPOV, as it seems least likely to conflict that way. So ... comments? Jakew (talk) 16:12, 11 August 2011 (UTC)[reply]

Something to consider... We have to factor WP:UNDUE in here... I see WP:COMMONNAME as being an application of the concept of due weight - as applied to the unique constraints of article titles (we can only have one title, which means we are forced to give more weight to one name over all others). When a given name is used by an overwhelming majority of sources (be it biased or not), we obviously use that name as our article title to give it due weight (and to substitute some other name would give that other name undue weight). The situation becomes murkier when the choice is less obvious, when the sources are more mixed in usage. Since we physically can't give equal weight to multiple names in a title, we do the next best thing... We give the more common name more weight by using it as the title, and the less common names their due weight by listing them prominently and in bold text (as alternatives) in the lede... and by using them as redirects. I think this applies to arguably biased (non-neutral) names as as much as any other name. To choose a less common, but seemingly more neutral name over a more common but potentially biased name gives that less commonly used name undue weight, and denies giving the more commonly used name its due weight.

Of course, when the sources indicate that no single name is more commonly used than any other, then weight is not an issue... and we can look to other factors (such as the appearance of neutrality) in reaching a consensus.

In other words... applying WP:DUE to titles ... first we look to see if there is a WP:COMMONNAME (and if so we use it, biased or not). If there is no WP:COMMONNAME, then we are free to use other criteria and factors to reach a consensus title. Does this make sense? Blueboar (talk) 17:00, 11 August 2011 (UTC)[reply]

I see what you're saying, Blueboar, but I'm not sure it's the right way around. Trouble is, there are two types of neutrality to consider. Your due weight analysis is concerned with neutrality with respect to chosen terminology — treating sources' use of terms as a viewpoint about terminology and trying to apportion due weight by choosing the most common. But there is another kind of neutrality as well: neutrality with respect to the subject. The question is what should be done when there is a conflict between the two: that is, when the most common term predisposes towards or otherwise favours a particular point of view about the subject, or which seems to assert a contested opinion about it. I'm not comfortable with your proposed approach, because I feel it overemphasises the first NPOV concern and overlooks the second. I guess that, depending on the context, one or the other may be more important. Sometimes there's more controversy about the subject than about terminology; while I can't think of any examples, it seems perfectly plausible that the reverse is also true. Jakew (talk) 18:13, 11 August 2011 (UTC)[reply]

When we have a situation where a term is not neutral with respect to the subject that's a possible reason to not use it as the title. We agree so far, yes?

I presume we also agree that it's not true that we never use such terms as titles. So, the issue is about where to the draw line between using some non-neutral terms (with respect to subject) as titles, and not others. Now, the only fair and neutral way I can fathom to draw that line which is in accordance with WP:NPOV is by looking at commonality of usage in sources. Can you suggest another fair and neutral method? --Born2cycle (talk) 18:40, 11 August 2011 (UTC)[reply]

Yes, it's called consensus among editors. It's not perfect, but if it's incapable of arriving at neutrality then article titles are the least of our problems. Jakew (talk) 19:19, 11 August 2011 (UTC)[reply]

That's a dodge circular reasoning because consensus is how we resolve disagreement among us, not how we individually base our decisions independent of what other editors are saying, which is what policy is supposed to be guiding. I'm talking about each individual editor - if not by looking at commonality of usage, what is a fair and neutral way for a given editor to draw that line between using some non-neutral terms (with respect to subject) as titles of some articles, and not others, in accordance with NPOV? --Born2cycle (talk) 20:48, 11 August 2011 (UTC) some rewording/clarifying --Born2cycle (talk) 21:22, 11 August 2011 (UTC)[reply]

I'd suggest that editors use their judgement, as they would when applying NPOV (and other policies) elsewhere. It's a hard problem, and I'm not convinced that a formulaic solution that would produce satisfactory results in corner cases is even possible. I'm perfectly happy to use the most common term as a baseline, as the vast majority of the time it's the right thing to do. But when exceptions arise, there's no substitute for individual editorial judgement, extensive (and probably tedious) discussion, and (hopefully) consensus. And I think we need some recognition in the policy that this sometimes happens — "wiggle room" if you like. Jakew (talk) 21:34, 11 August 2011 (UTC)[reply]

Ultimately of course it depends on judgement. But judging what? Based on what? It doesn't have to be formulaic, but if there are to be exceptions, they have to be based on something that applies in some cases but not in others. What is that? You don't have to quantify it, but if you can't even suggest what some of the factors and considerations might be, you're talking about pure WP:JDLI, which is generally eschewed. --Born2cycle (talk) 21:49, 11 August 2011 (UTC)[reply]

No, Born2cycle, I'm not talking about editors' personal whim. I'm talking about editors using their judgement to test whether a title meets with NPOV (and, of course, explaining their reasoning clearly in the process of discussion). The only way to describe the provisions of that policy, while fully capturing every nuance, is to quote it in its entirety, and I'm not going to do that. I'm also not even going to try to present an algorithm for "how to think like a human"; smarter people than I have been trying that for a long time, and haven't managed it yet. Most WP policies deliberately leave details unstated, as it is understood that it is impossible to think of all applications in advance, and it is expected that editors will apply their intelligence when considering a particular application. I think that's the right approach for this policy, too. Jakew (talk) 09:59, 12 August 2011 (UTC)[reply]

I think mixing Non-neutral title and Non-neutral words in this section is a mistake. The mere presence of a Non-neutral word or term as the community has defined them in a title doesn't by any measure make the title Non-neutral. What makes a title Non-neutral is when the title clearly demonstrates bias and POV when there is a controversy about what something is called. As cited many times above, the Boston Massacre is a neutral title because that's what the incident was and is called by the preponderance of secondary sources. If we were to call the Battle of the Little Bighorn the Custer Massacre as a minority of sources have referred to it, Custer Massacre would clearly be a non-neutral title. This section should be talking about the Neutrality of titles, not the presence or absence of non-neutral words.--Mike Cline (talk) 17:12, 11 August 2011 (UTC)[reply]

I agree that's a useful distinction - non-neutral titles vs. non-neutral words. We need to be clear about what we mean by a non-neutral title, however, and, specifically, that we don't mean something as simplistic as "a title comprised of non-neutral words", which is apparently the meaning intended by the proposed wording above (Jake, please correct me if that's not correct).

A more nuanced definition of neutral title is: a title that is selected from a neutral point of view.

With that in mind the proposed text is incorrect in saying that "non-neutral titles are sometimes acceptable", as they aren't. What is acceptable is the use of non-neutral words as titles and as part of titles when that reflects common usage in reliable sources. --Born2cycle (talk) 17:40, 11 August 2011 (UTC)[reply]

Personally, I'd define a non-neutral title as follows. Suppose that no established names existed for a subject, and suppose that Wikipedia editors came up with a word or phrase to describe it in another article. If that word or phrase would be likely to be challenged on non-frivolous NPOV grounds, it's non-neutral. I know that's a slightly circular definition, and I apologise for that, but it's the best I can come up with. Non-neutral words are a potential reason for such a challenge, but it may be a more subtle issue such as an implied one-sided perspective (eg., "pro-life").

I think that trying to define a neutral title as the most common name will be counterproductive. It will inevitably lead to arguments (such as we've seen above) about what "neutral" means. It's much better to understand POVTITLE (and NPOV#Naming) as an exception to neutrality; that way we can bypass discussions about the definition of neutrality (which are ultimately unproductive) and instead concentrate on whether the exception is warranted or not. Jakew (talk) 18:36, 11 August 2011 (UTC)[reply]

I think you're on the right track when you talk about the two meanings of "neutral"—the "complies with NPOV (especially DUE)" and the "non-judgmental or non-offensive, according to the personal opinions of individual editors" idea. Wikipedia really only requires that article titles demonstrate the first type of neutrality. NPOV specifically disclaims the idea that a name that is very commonly used in reliable sources should be rejected merely because editors think it sounds biased.

I don't want to re-write this policy to ratify one controversial page move. There are good reasons to avoid "pro-life" that have nothing to do with an appearance of endorsing the self-description: it is an imprecise label used to represent a wide variety of sometimes contradictory views. For example, one can be "pro-life" meaning "opposed to artificial reproductive technology" or "in favor of ART for anyone who wants to have a baby". I've met more than one person who self-describes as "pro-life" and strongly opposes abortion, but definitely wants abortion to remain legal for various other reasons (suicide prevention seems to figure prominently in the list). It would not be entirely unreasonable for us to make Pro-life into a set index or WP:DABCONCEPT article. I believe it would be unreasonable to extend the fight over that page to re-writing this policy in the (doomed) effort to somehow stop the grumbling over that decision. WhatamIdoing (talk) 19:34, 11 August 2011 (UTC)[reply]

I don't know of any topic that qualifies as WP:PRIMARYTOPIC better than does the anti-abortion meaning of "pro-life". The other uses exist, but they're rarely used, and I doubt anyone would expect to find them covered at an article named Pro-life. It seems to me the "imprecise label" argument is a non-starter. As it stands, not being at Pro-life is an inexplicable exception to our policies, conventions and widespread conventions. --Born2cycle (talk) 20:45, 11 August 2011 (UTC)[reply]

I would offer this characterization of a neutral title: A neutral title reflects the most common and most neutral point on view as what to name the subject of any given article based on the preponderance of reliable secondary sources.

The single most important aspect of the above characterization is having an understanding of what the word neutral means. The dictionary definition is very clear: not aligned with or supporting any side or position in a controversy. I think we must be very careful not to try and apply any other definition to the word. (offensive for instance has nothing to do with neutrality) 1000s of editors who will create articles in the future will see the word Neutral and know exactly what it means: not aligned with or supporting any side or position in a controversy. Any other arbitrary or home-grown WP definition won’t mean a thing.

Our policy then should be for editors to avoid, as far as practical, non-neutral titles. Non-neutral titles being titles that reflect a strong bias one way or the other when there are multiple points of view on a subject and the preponderance of reliable secondary sources reflect wording that is well within the middle ground between the opposing points of view. A neutral title reflects the middle ground as defined by the sources, not individual editor’s POVs, likes or dislikes. We cannot expect to establish a black and white policy on neutrality of titles--too many variables. But, I believe we can clearly characterized what a Neutral title is (what we expect) and what a Non-Neutral title is (what we want to avoid). If we do that, clearly and concisely, then individual discussion on any titling controversy will be governed by those characterizations and whatever the preponderance of reliable secondary sources say. --Mike Cline (talk) 20:49, 11 August 2011 (UTC)[reply]

A neutral title is not necessarily common at all. --Born2cycle (talk) 21:23, 11 August 2011 (UTC)[reply]

The dictionary definition of "neutral" is exactly what we don't mean when we are talking about neutrality on Wikipedia. We frequently choose titles that are "aligned with or supporting" controversial events. The policy even gives several examples of such titles. The policy does not require real-world neutrality; it requires wikijargon neutrality (which for clarity ought to be spelled "N-P-O-V" rather than "n-e-u-t-r-a-l"). WhatamIdoing (talk) 22:44, 11 August 2011 (UTC)[reply]

On this I think you are dead wrong. First, you cannot define neutral with itself (NPOV). Second Neutral Point of View implies there are opposing Points of View (that's the controversy, not that the subject of the article is a controversial topic) and that there is a Point of View that's in the approximate middle. When there are opposing Points of View on any subject, its highly likely there is some neutral Point of View in between. If what we want to prescribe is political correctness and culturalless titles, then we should say so. But pretending to say that NPOV is anything but the middle ground between opposing points of view on any given subject will not help editors create neutral titles. --Mike Cline (talk) 23:16, 11 August 2011 (UTC)[reply]

The revision seems to be overly watered down in favor of wiki-revisionism. Wording like Non-neutral titles are sometimes acceptable....Wikipedia may follow the sources (my underlining for emphasis) seems to overly endorse political correctness (“being non-judgmenetal… man!”) despite common usage. If a significant majority of most-reliable English-language reliable sources all refer to something by a given name, Wikipedia should *generally* abide as such. Notable exceptions would be trendy names that A) seem unlikely to be remembered years later after the hubbub has died down; or B) names that are so colloquial that they detract from the encyclopedic nature. Whatever we come up with, it should anticipate what readers will type and what readers will expect to be taken to.

For instance, if I type “Boston Massacre" into the search field, I expect to be taken to an article by that name without 16-year-old all-volunteer wikipedians trying to Change The World©™® by coming up with something less emotionally charged against the Brits. By the same token, if I type “Octomom” into the search field (which is what she is known by today as well as five years from now), I expect to be redirected to “Nadya Suleman”—even though 90% of readers can’t remember her name and how to spell it but can all remember “Octomom.”

I think the best guideline would have unique examples (like “Octomom”) and would directly address the principles each touches upon. So I call for everyone here to put in the below section, examples of article names that teach to this issue. I’ll start it off with six. After we have a list of names of article titles and redirects, then let’s see a thoughtful guideline that sweeps it all up nicely. Greg L (talk) 00:07, 12 August 2011 (UTC)[reply]

Illustrative or unique titles and redirects to consider when crafting a better guideline

I propose we all leave examples of interesting and/or unique article titles and their redirects here that we think speak to crafting a new guideline. Feel free to post examples that you agree with, as well as article titles you think are a mistake. We can discuss the implications. After that, maybe someone can craft a better guideline. In the below list, I agree with five of the six examples. On the one I disagree with, I don’t have strong feelings about, but still think it would have been best had it been titled as it is most commonly known. Greg L (talk) 00:07, 12 August 2011 (UTC)[reply]

• Octomom → Nadya Suleman …by Greg L (talk) 00:07, 12 August 2011 (UTC)[reply]
• Rape of Nanking → Nanking Massacre This one is a mistaken title in my opinion …by Greg L (talk) 00:07, 12 August 2011 (UTC)[reply]
• Boston Massacre → (no redirect) …by Greg L (talk) 00:07, 12 August 2011 (UTC)[reply]
• Pro-Life → Opposition to the legalization of abortion …by Greg L (talk) 00:07, 12 August 2011 (UTC)[reply]
• Pro-Choice → Support for the legalization of abortion …by Greg L (talk) 00:07, 12 August 2011 (UTC)[reply]
• Patriot (American Revolution) → (no redirect). Acknowledgment to Blueboar, posted by Greg L (talk) 00:07, 12 August 2011 (UTC)[reply]
  
• Defense of Marriage Act → (no redirect) …by Greg L (talk) 21:47, 12 August 2011 (UTC)[reply]

This is an open-forum section; additions welcome below the line.

Another proposal

Thinking about the above, I propose the following:

I consider the above to be part of my own post (verboten for others to edit) but also to be a live sandbox that I might update after seeing comments from others. Those who want to suggest something other than above should create their own green‑div. Greg L (talk) 00:40, 12 August 2011 (UTC)[reply]

Looks OK; I'd think that "the commonality of the name overrides" might be better as "the commonality of the name may override", since that's the main point we're working on here, isn't it? Dicklyon (talk) 02:00, 12 August 2011 (UTC)[reply]

This is good — certainly an improvement over my attempt. I agree with Dicklyon's tweak. Also, "siding with one side" is a little awkward; can I suggest "promoting one side" instead? Jakew (talk) 09:03, 12 August 2011 (UTC)[reply]

This proposal looks like an improvement to the current. The suggestions made by Dicklyon and Jakew should also be included. Cody7777777 (talk) 13:07, 12 August 2011 (UTC)[reply]

• Ergo…

How say others? Greg L (talk) 19:05, 12 August 2011 (UTC)[reply]

I'm not as keen on it as I was on the above version, I have to say. Some specific criticisms... First, the effective status of a name as a proper noun is mentioned in the first para and then as an "other factor" in the second, which is slightly confusing. Second, I'm not keen on "otherwise seemingly non-neutral" — don't be afraid to say "non-neutral"; since we've already explained that can be permissible. Third, unless I'm greatly mistaken, "Whazzup" does not belong here as it's not a non-neutral name. Finally, I strongly dislike "in keeping with exception #2" — we absolutely should not present these as numbered exceptions; they're examples of situations where we've deviated from the general rule. Jakew (talk) 08:48, 13 August 2011 (UTC)[reply]

The new version seems to be detailing too much when non-neutral and neutral terms can be used or not, and this might not be necessary. And I think "may override concerns" is better than "generally overrides concerns". Also, instead of "When a significant majority of English-language reliable sources refer to the subject of an article by a given name..." I would have preferred we had "When the subject of an article is referred mainly by a single common name, as evidenced through usage in a significant majority of English-language reliable sources,..." or something similar, to emphasize that when there are multiple common names, we do not necessarily have to choose a name which is clearly known as being non-neutral (and this would actually be similar to the version from february 2011, where it stated, "When a subject or topic has a single common name (as evidenced through usage in a significant proportion of English-language reliable sources)..."). Cody7777777 (talk) 12:57, 13 August 2011 (UTC)[reply]

OK. Then…

I tried to take as much from what each of you wrote and incorporate it all here. That ended up trimming it down to something much closer to my original proposal.

Note that I left …functionally become a proper noun generally overrides concerns… (my italicizing for emphasis). I get my 2¢ in here too. If a signficiant majority of English-language reliable sources refer to something by a given name to such an extent that the name is ubiquitous and/or has effectively or functionally become a proper noun, then mere all-volunteer wikipedians generally have no business trying to change the world—no matter how well intentioned they might be (*sound of audience gasp*). If Wikipedia can have an article on “Fuck”, (not withstanding that our readership has a large proportion of primary school children), we obviously expect readers to be sufficiently sophisticated to understand that we’re not taking sides when they type “9-11 hijacker” and are taken to an article containing the word “hijackers”.

How say ye all? Greg L (talk) 17:53, 13 August 2011 (UTC)[reply]

Good by me. Thanks for working on it and incorporating suggestions. Dicklyon (talk) 18:33, 13 August 2011 (UTC)[reply]

Thank you, this is clearly an improvement. I do not have any serious problem using "generally overrides concerns" instead of "may override concerns", but I think, there should also be a mention that more neutral alternative common names can also be used where possible (and we don't have to invent descriptive titles). Perhaps the following line "But, in cases where more neutral alternative common names are possible, these can also be used instead." (or something similar) could be added after "In such cases, the ubiquity of the name and/or the fact that the name of an event has effectively or functionally become a proper noun generally overrides concerns that Wikipedia might appear as endorsing one side of an issue." (to explain cases where there are multiple common names). (And regarding the article called "Fuck", as far as I see it is used for documenting the negative uses of the term, it is not used as a title or redirect for sexual intercourse.) Cody7777777 (talk) 18:47, 13 August 2011 (UTC)[reply]

• Cody, I mentioned our article on “Fuck” not in the context of what it might or might not redirect to, but to point out that Wikipedia is written assuming a certain level of sophistication in the reader—as I wrote there.

  As for your suggestion (But, in cases where more neutral alternative common names are possible, these can also be used instead), I am not at all supportive of encouraging wikipedian-crafted, more-neutral alternatives just because they are *possible*—not in the clear context where a given topic is referred mainly by a single common name, as evidenced through usage in a significant majority of English-language reliable sources (my emphasis). This proposal pretty much endorses existing practices on Wikipedia (that is, it doesn’t try to make waves, so to speak), and adheres pretty closely to the general principle of “follow the RSs unless there are good encyclopedic reasons not to.”

  This whole thread and its sub-threads have become lengthy enough that each editors’ views are pretty clear now. Compromise text is not easy; let’s see how the others feel now and see if we have a general consensus on this amalgam of compromise wording. Greg L (talk) 19:04, 13 August 2011 (UTC)[reply]

Looks good to me, too, and I echo the above thanks to Greg L for his hard work. Jakew (talk) 18:52, 13 August 2011 (UTC)[reply]

• OK. I think we have a consensus on something that is essentially uncontroversial. I’m putting the text in. Also, I added mention of WP:RNEUTRAL, which was in the original text, the omission of which fell through the cracks during our discussions. My ∆ edit is here. And here is a direct link to the section: Wikipedia:Article titles#Neutrality in article titles. I think this discussion thread serves as a paradigm of collaborative writing and consensus building. Thanks to all. Greg L (talk) 19:33, 13 August 2011 (UTC)[reply]

I don't think this has changed much, but if the examples are useful, good. As a matter of wording: names do not become proper nouns; they are proper nouns. As originally worded, also, this could be read to support use of Octomom as a title; it's a proper noun, now. The problem with Octomom is that it's not ubiquitous, and may be expected to fade with time. Septentrionalis PMAnderson 21:16, 13 August 2011 (UTC)[reply]

• “Octomom” isn’t ubiquitous? How about 3,830,000 Google hits. I’d bet that not even one in every ten people who are familiar with the term “Octomom” know that the lady’s actual name is Nadya Suleman. But “Octomom” is a colloquialism where using her actual name is obviously far more encyclopedic. Greg L (talk) 22:53, 13 August 2011 (UTC)[reply]
  • Why do we take "Octomom" to be a proper name? There are a fair number of sources that write of "the octomom" in lower case, too. Dicklyon (talk) 23:07, 13 August 2011 (UTC)[reply]
    • I assume, because we are talking about a *particular one*. How many are there? Greg L (talk) 18:42, 14 August 2011 (UTC)[reply]

• I'm sorry that I'm a bit late to the party here, but I think these three exceptions here are completely unacceptable, and I have no idea how it has already been implemented in the policy. These exceptions leaves so much open to interpretation and POV pushing. Especially the third one is a problem for me. It's clearly written with the ongoing discussion about naming of the abortion advocacy pages in mind. But I think it completely misses the point of that discussion. The central point was not that the terms "pro-life" and "pro-choice" were selfcreated slogans per se. The main point was that it could be proven with reliable sources that these terms are viewed as partisan and unneutral, and that they as a result of that are avoided by high quality sources. I think we atleast should make it a principle that NPOV claims to stray from WP:AT's principles should be backed up by reliable sources, so that we avoid that NPOV in titling becomes an anarchy where WP:V no longer applies and the personal opinion and taste of editors reign supreme. I might be able to accept 1 and 2 if they were written in prose and if it is made clear that they are only for when there are good common name alternatives, and that the first one is really just an extention of WP:RECENT.TheFreeloader (talk) 23:27, 14 August 2011 (UTC)[reply]
  • Yes, you indeed are a tad late. I don’t understand your astonishment at how anything has “already been implemented” since this change was extensively debated for nearly seven days and enjoyed wide participation from editors working towards a general consensus. In the military, the expression is “So sad – too bad” whenever one arrives late to a party and professes to be crestfallen at discovering who got drunk and got laid. But your objection still caries weight on Wikipedia and is noted; let’s see if that changes the general consensus now.

    As for your observation that The central point was not that the terms "pro-life" and "pro-choice" were selfcreated slogans per se. The main point was that it could be proven with reliable sources that these terms are viewed as partisan and unneutral, and that they as a result of that are avoided by high quality sources. Let’s compare that to the current text now, which declares the following to be inherently non-neutral and worthy of not using in a title: Persuasive names and slogans crafted by partisans on still-active, contentious advocacy issues. They are essentially the same thing. One big difference pertains to cause & effect. You see, persuasive names and slogans crafted by partisans on still-active, contentious advocacy issues are inherently like shit: off of it comes stink. Your point is amounts to saying “avoid the stink, which comes from the shit.” The existing wording says “avoid the shit, from which comes stink.” Same diff. Cause & effect.

    The more significant difference in your point is over what must be “proven” by RSs about something being biased. That works for long-established issues but breaks down when new articles are being created on new issues. It’s about time we had wording that cut to the chase and said, in effect, “Was the title crafted by partisans who are on dueling sides of an advocacy issue? If ‘yes,’ then think again about naming a Wikipedia article after it.” In short, it expects a healthy dose of WP:COMMONSENSE (*sound of audience gasp*) to be applied without having to wait around for an RS to say “These dorks are biased.” If the RSs come around on a new issue and state as much, that will just validate that Wikipedia was right to have avoided naming an article after it in the first place.

    A (very) serious weakness of your argument lies in the last clause of your argument: …and that they as a result of that are avoided by high quality sources. Uhm… Note Encyclopedia Britannica’s article “Pro-life movement”. The New York Times too (here). The EB is unquestionably a reliable source that is not avoiding naming an article after the (obviously biased) slogan. I happen to agree with the E.B. since I find it absurd to think that anyone would believe that the E.B. is endorsing the views of the pro‑life or pro‑choice movements because they have articles by those names; that’s absurd. I was, however, trying to craft better language that reflects current practices on Wikipedia and doesn’t try to change those practices. In my opinion, there is far too much political correctness run amok on Wikipedia by naive youngsters out to change the world and make it a better place—you know, hold hands on a hill and sing about Coca-Cola (YouTube video), but that sort of thing seems to make wikipedians happy.

    Nevertheless, I did my best to help craft, without passion or prejudice, consensus text that didn’t try to change how things work around here. Greg L (talk) 00:31, 15 August 2011 (UTC)[reply]

    • I think it goes against the whole idea of not just WP:AT but all of what Wikipedia is that we should be prescribe what is and what isn't neutral. Something like that will always just be an opinion, and editors' personal opinions do not belong in Wikipedia, WP:FIVEPILLARS says at much. What is neutral to you may not be to me, and there is no way to reconcile those two opinions. We need outside sources to be the arbiters in that kind of situations. Wikipedia is an encyclopedia build on reliable sources, not on the personal opinions of its editors. I think it is completely unprecedented in Wikipedia policy to have this serious a departure from WP:V and reliance on reliable sources.TheFreeloader (talk) 00:58, 15 August 2011 (UTC)[reply]
      • Well, now that you’ve wrapped your position in The Five Pillars (and linked to it in that “If it’s blue, it must be true” manner), it would be swell if you addressed the parts of my above post that showed your argument to have holes in it large enough to drive a semi clean through.

        If it were up to me, we’d lose #3 in its entirety. Instead, what’s there actually withstands scrutiny rather than crumble under the false pretense that the titles we avoid are avoided by high quality sources. As I provided above, there’s more RSs than you can shake a stick at that use the term (and the E.B. has an article by the name “Pro-life movement”). That’s clearly not the reason Wikipedia avoids such titles so please stop asking us to pay no attention to that illogic behind the curtain. Greg L (talk) 01:02, 15 August 2011 (UTC)[reply]

        • I think I addressed pretty clearly that it isn't Wikipedia's job to prescriptive on anything. BTW, here is the evidence I was referring to from the abortion rights case[1]. It's evidence like that, based on outside sources, I like to be norm for if trying to make a case about neutrality in titling of articles, not just whatever one's personal opinion happens to be.TheFreeloader (talk) 01:14, 15 August 2011 (UTC)[reply]
          • Your arguments are going nowhere here. The link you provided, which is two wikipedians arguing with each other, has something in it where one guy says New York Times - covered; NYT does not use "pro life" or "pro choice" as descriptors as they are not neutral. Do you understand what not using those terms as “descriptors” means? I fully agree with that premiss. That doesn’t have anything to do with how Wikipedia and Encyclopedia Britannica name their articles covering a subject. Time for dinner. Bye. Greg L (talk) 01:24, 15 August 2011 (UTC),[reply]
            • Well, even if those arguments don't float, that wasn't my point anyways, it was just an example. My point was that arguments about neutrality in titling have to rely on reliable sources, not personal opinions. You keep avoiding to address that point.TheFreeloader (talk) 01:36, 15 August 2011 (UTC)[reply]

"Sometimes that common name will include non-neutral words", Rape is usually a neutral word. Massacre fits the bill better. -- PBS (talk) 01:20, 15 August 2011 (UTC)[reply]

• I say, “call it like it is if it is a long-term grown-up word” Always. Even if it is “Pro-life movement” — just like Encyclopedia Britannica does. We would have only exceptions #1 and #2 in the current exceptions rules and would lose #3 (partisan slogans). But I know that one would fly like a wet noodle. Greg L (talk) 01:26, 15 August 2011 (UTC)[reply]

Greg, I think I'm missing your point about the Britannica and NYT. The Britannica link you gave goes to what looks like a redirect to "abortion (pregnancy)"; there's no article at "pro-life movement", or none that I can see anyway. And the NYT article you link talks about the "anti-abortion movement", and uses "pro-life" in the article only in quotes to say how people identify. The fact that the headline writers used it shouldn't be over-interpreted; they work by different rules, to sell stuff by catchy headlines. Dicklyon (talk) 01:48, 15 August 2011 (UTC)[reply]

• Yes, you are correct. I saw this “Pro-life movement” Goggle hit on EB and didn’t read the EB page well enough to see if amounted to an actual article. The EB page looked like one of those “subscribe-to-read” things, so I dismissed it. Type too fast — read too little. I think, therefore, I fully support all three points of the current text (as opposed to my earlier opinion, where I helped author it because it was the consensus view, even though I thought point #3 was namby-pamby). Thanks. Greg L (talk) 02:20, 15 August 2011 (UTC)[reply]
  • So, what do you say, do we put in something about NPOV claims having to be supported by reliable sources?TheFreeloader (talk) 02:01, 16 August 2011 (UTC)[reply]
    • F, I didn't understand why you restored the "last stable version" when we seemed to be making such good progress. I've reverted that. Tell us what your thinking is. I also don't understand what you mean by "NPOV claims having to be supported by reliable sources"; maybe you can propose wording that would clarify what you're thinking? Dicklyon (talk) 06:02, 16 August 2011 (UTC)[reply]
      • I'm thinking that it's good to be conservative when it comes to policy. This thing is far from settled as far as I see it, the RFC is still open, so best not change anything until we agree on this. Also, I don't appreciate my comments being ignored. What I mean by NPOV claims being supported by reliable sources is that it should be proven on a case by case basis that the concerns one is having with a common name's neutrality are concerns which also exist outside of Wikipedia. Else we leave titling discussions open to all sorts of WP:OR type arguments for all sorts of things which can be seen, to some, as not neutral. My central point is, we need to find some sort of way for WP:AT to clearly define how to interpret WP:NPOV when it comes to titling. The old version (kinda) defined it as "the common name is always neutral", but now this language is being weakened, and we therefor need a new way to define it, and I suggest it should be "unless it can be proven to be regarded as otherwise by outside sources, the common name is always neutral". That way there will be room for people who thinks Wikipedia should stand up to higher standards of neutrality than most reliable sources to argue their case, while still keeping personal opinions out of the decision making, the way WP:PILLARS intend it.TheFreeloader (talk) 06:25, 16 August 2011 (UTC)[reply]
        • It does seem to me that this was done slowly and deliveratively with lots of involvement, and that nobody has said that the current version is not an improvement on what we started with. If you still have concerns, let's work on that. On your specific proposal, "unless it can be proven to be regarded as otherwise by outside sources, the common name is always neutral" would seem to just shift the argument to what constitutes "proof"; in any case, editors will make their case to each other and try to be convincing; is there precedence for this kind of clause, and has it been found to work? Dicklyon (talk) 15:01, 16 August 2011 (UTC)[reply]
          • Well, I cited the abortion naming discussion above as an example where outside sources were used in argument for not using the common name out of neutrality concerns. I also think there is plenty of precedent for this clause in the way WP:NPOV is treated in normal content related issues, where it isn't what you think is "neutral" which is what matters, but true neutrality is to inform readers what the most respected sources on given subject have to say.TheFreeloader (talk) 15:25, 16 August 2011 (UTC)[reply]
            • What clause at WP:NPOV are you referring to? Or in some place that refers to it? I don't see anything about proof. I also don't see where your notion of "true neutrality" is specified to exclude "what you think is neutral". To me, the nutshell says it well: "Articles mustn't take sides, but should explain the sides, fairly and without bias." That means even in areas where sources are predominantly biased toward one or another side, doesn't it? The provisions at Wikipedia:NPOV#Naming seem about right, and don't say anything about "proof"; that concept would just escalate the problem to a meta level of argument, I think. Dicklyon (talk) 16:00, 16 August 2011 (UTC)[reply]
              • I think WP:V is a lot about proof, and about using reliable sources for determining what should be in Wikipedia and what it should look like. My notion that neutrality comes from following sources, not personal opinions comes from WP:PILLAR. I'm not say that what I am advocating is already Wikipedia policy, I'm saying that it builds on the spirit of the Wikipedia's core content policies. And just to be sure, I'm not one who wants more neutrality concerns making their way into naming debates. If I were to choose myself, the policy would not allow any exceptions from WP:COMMONNAME based in concerns about neutrality. Because who are we to come and say that what reliable sources choose to call something isn't neutral. So all I'm saying is that if we are going to allow exceptions WP:COMMONNAME based on neutrality concerns, then we ought to at least do it in a way consistent with how WP:NPOV is usually implemented in other areas, which is through the use of reliable sources.TheFreeloader (talk) 17:08, 16 August 2011 (UTC)[reply]
                • I agree that things flow from sources; but sometimes the slogan "follow the sources" is used to mean follow the most common POV (or style, or name, or whatever), at the expense of NPOV. I don't quite get your notion of "proof" still; it seems unprecedented to put something like that into policy and guidelines. Dicklyon (talk) 17:14, 16 August 2011 (UTC)[reply]
                  • Yea, I was thinking about that wording when I wrote it too. But I thought it was safe to say that it is "proven" that a reliable source regard something as not neutral if it can be shown where they say so themself. But instead how about then just saying "if it can be shown that reliable sources regard the common name not neutral". Also, I think there is precedent for featuring the most common POV most prominently in WP:DUE. And I would much rather have "follow the sources" as a slogan for Wikipedia than "go with what you think is right".TheFreeloader (talk) 17:33, 16 August 2011 (UTC)[reply]
                  • So what do you say, do you support removing the current exceptions, and instead require that all exceptions from WP:COMMONNAME due to concerns about neutrality should be made based on reliable sources and on a case by case basis?TheFreeloader (talk) 15:56, 17 August 2011 (UTC)[reply]
                    • I think it's true that, in practice, title decisions based on neutrality are often required by Wikipedians to be supported by compelling, source-based arguments. On the other hand, there are certain neutrality considerations that we tend to accept somewhat automatically, such as eschewing the word "terrorist" unless we have compelling reasons to use it.

                      Does that jibe with other people's observations? -GTBacchus^(talk) 16:31, 17 August 2011 (UTC)[reply]

                      • Well, I actually cannot think of any articles where the term "terrorist", or any the other words from WP:LABEL, is in a subject's common name, but Wikipedia has decided not to use it because of neutrality concern. We do have whole bunch of articles including the word terrorism in this template Template:Terrorism. So I don't think the words to watch are that big a concern here.TheFreeloader (talk) 16:59, 17 August 2011 (UTC)[reply]
                        • Can you make us a new version proposal in a box, like Greg made, so we know exactly what we're considering? Or is it as simple as removing "Notable circumstances..." from his latest? Dicklyon (talk) 17:08, 17 August 2011 (UTC)[reply]

Alright, here is my suggested version:

The language needs work, but this is essentially how I think we could comprehensively define how WP:NPOV should be interpreted when it comes to naming of articles.TheFreeloader (talk) 17:27, 17 August 2011 (UTC)[reply]

Yes, it needs at least a good copy-edit. I think I prefer Greg's; the examples are clarifying. And your stuff about "claims" seems to be talking about an adversarial process rather than a guideline; do we really need that here? Dicklyon (talk) 17:55, 17 August 2011 (UTC)[reply]

Well, alright, what do you suggest we call it instead? I think it is quite shallow to judge these proposals based on language and presentation. My proposal is substantially different from Greg's proposal, in that Greg's proposal is mainly a list on some things which may be viewed as unneutral, while mine is an attempt at filling in the gap in guidance on how to apply WP:NPOV to article titling.TheFreeloader (talk) 18:17, 17 August 2011 (UTC)[reply]

What we call what? All arguments always need to be backed up by something better than opinion; calling out a need to back up claims in this particular provision seems odd and unnecessary. Dicklyon (talk) 16:59, 19 August 2011 (UTC)[reply]

No, I don't think it is unnecessary. As it is right now it is possible, as it is often done, to interpret WP:V only to extent to content related matters, and not to naming discussions. I think it is important we make clear WP:V actually does apply in naming discussions. Questions about neutrality in naming is often quite controversial matters with lots of opinions about what is and what isn't neutral. I think we need WP:V for those kinds of discussions to make sure that the neutrality concerns we allow to influence our choice in article titles, are concerns which also exist outside Wikipedia.TheFreeloader (talk) 18:57, 19 August 2011 (UTC)[reply]

Ah, I'm finally digesting your point. You had said earlier that you're worried about "an anarchy where WP:V no longer applies and the personal opinion and taste of editors reign supreme." I see it quite differently. I think WP:V should be left for what it is; it has not generally been invoked in all the places where editors have to make choices based on policy and guidelines, and I don't see why it needs to be invoked as part of naming. I do agree that we don't want naming (or anything else) to be just a situation where "the personal opinion and taste of editors reign supreme." But some editorial judgement is generally required in interpreting and implementing policy and guidelines, and everyone knows that if there are disagreements then mere opinion carries little weight in settling them. I don't see any need to mess with that, nor any reason to think that adding something about WP:V or "proof" or "claims" will make it work better than it has been. If we need to add something to remind people of how to argue based on evidence, we can do that more globally, not in the COMMONNAME or POVTITLE section. Dicklyon (talk) 23:25, 19 August 2011 (UTC)[reply]

Well, we try to base all other parts of article naming on reliable sources, the nutshell and the lead for this policy explains that, so I don't see why neutrality claims shouldn't also be based on reliable sources. After being part of two discussions about neutrality in naming recently/currently, I think people do need to get reminded that basing their case on reliable sources is the best way to show that their concerns exist outside of Wikipedia/their heads. In the abortion article title dispute most of the discussion in the mediation cabal discussion was based on personal opinions on what people viewed as neutral, which lead to the discussion becoming very disorganized, with nothing to base consensus. The ongoing naming discussion about the China article and the People's Republic of China article has been a discussion with one side almost solely basing their position on personal opinions about what constitutes neutrality. When I tried to ask for any references to outside sources showing that their views were shared by anyone out in the real world, I was told by multiple editors that such demands were unreasonable and irrelevant, as outside sources are not generally used in such discussions.TheFreeloader (talk) 00:17, 20 August 2011 (UTC)[reply]

I'm not saying such arguments shouldn't be based on reliable sources; if you repeat "reliable sources" again in an appropriate place, maybe that will help clarify. But it's not a WP:V issue, which is about what to put into articles. Dicklyon (talk) 02:12, 20 August 2011 (UTC)[reply]

The lead of this policy says that it: " should be interpreted in conjunction with other policies, particularly the three core content policies: Verifiability, No original research and Neutral point of view." Also the lead for WP:NPOV say: ""Neutral point of view" is one of Wikipedia's three core content policies. The other two are "Verifiability" and "No original research". These three core policies jointly determine the type and quality of material that is acceptable in Wikipedia articles. Because these policies work in harmony, they should not be interpreted in isolation from one another [...]" So I think it is reasonable to say that if people are going to make claims based on NPOV, that they be based on what can be verifiable by outside sources, and do not include original theories about what is and isn't neutral.TheFreeloader (talk) 16:23, 20 August 2011 (UTC)[reply]

Your proposal still needs improvements. The second part should have started as "When a subject has multiple common names, questions about their neutrality may be taken into account..." (I think it's more clearer this way, and also avoids repetitions). But, still there were cases where even without multiple common names, editors have used neutral descriptive titles instead, so I think it can also be noted that in some very controversial cases, these may also be considered. And the examples provided in Greg's version can be useful. I agree that sources should be used to determine if a common name is non-neutral (especially if these are not very obvious cases). Cody7777777 (talk) 21:17, 17 August 2011 (UTC)[reply]

I've put your improvement into the proposal above, thank you. The only case I can think of where Wikipedia has chosen to against POVTITLE by creating a descriptive name despite having common names available is the abortion movement case. And that is a case which is still being debated (here), so I think it would be wrong of us to try to base policy on an ongoing dispute.TheFreeloader (talk) 21:39, 17 August 2011 (UTC)[reply]

I think that this needs to more clearly reflect the text from NPOV#Naming, which makes a point of accepting a common name "even though some may regard it as biased." The disregard for editors' personal opinions about what sounds biased from their POV is a major point here, and it needs to be made more clearly. I suggest expanding the first sentence in this proposal with something like 'even if some editors believe that title sounds biased or judgmental'. WhatamIdoing (talk) 16:41, 19 August 2011 (UTC)[reply]

We should be careful of simply recommending disregarding some editors' opinions. I would like to emphasise that with very common names, objections to a name's neutrality should be backed up by reliable sources making such criticism, rather than simply reflect editors' personal judgements. What we're talking about here are valid arguments to bring to a discussion, not a formula.VsevolodKrolikov (talk) 16:52, 19 August 2011 (UTC)[reply]

(ec) I wouldn't make it about editors. The phrase "even though some may regard it as biased" is more about how a term is regarded "out there", I think. Opinions of editors aren't what policy should be based around, one way or the other, and we shouldn't have to go to special lengths to say so in this particular provision; editors should always focus more outward (not that they always do). Dicklyon (talk) 16:54, 19 August 2011 (UTC)[reply]

PBS has ignored this concerned restore the "editor's" that I had removed from W's edit on the policy page. I still don't see why either of you wants to start making this section talk about opinions of editors. We don't normally do that; we could say on everything that we do it even though some editors may have contrary opinions, but we don't do that. The quoted line from Naming, "even though some may regard it as biased", is presumably about how some sources, authorities, or whatever regard it, not the opinions of certain editors, no? Dicklyon (talk) 06:48, 20 August 2011 (UTC)[reply]

No its not, as I said in my revision history if we have reliable sources that state a name is biased, we weight that in to the process of deciding the article title. Taking out the word "editors" makes the clause totally different from what I presume WhatamIdoing intended. If we are lucky enough to find reliable sources that talk about an article title either by providing us with a survey of usage, or commenting on it in other ways such as it is a propaganda phrase etc, then that is precisely the sort of information that we want to discuss and use in deciding article titles. It is the uniformed Wikpedia editorial prejudice that we need to discourage because unless people are willing to discuss the problem by researching reliable sources they are unlikely to agree if they hold different opinions in the first place. If editors consider the usage in sources then in good faith they can usually agree on those to find an acceptable article title while agreeing to digress on the "correct name". -- PBS (talk) 15:20, 20 August 2011 (UTC)[reply]

Dicklyon what do you see as the significant difference between WP:POVTITLE and WP:NPOV#Naming? -- PBS (talk) 06:03, 20 August 2011 (UTC)[reply]

I am not sure that the copying of the WP:NPOV#Naming section over to WP:AT was a good idea for the reasons Blueboar has mentioned -- When we copy stuff from one policy to another over time they get out of sync in subtle ways which then get exposed when a dispute arises and cause disputes on article talk pages and on the policy pages (like this one to become longer than they would if there was only one section in one policy). Perhaps the answer is to replace the bulk of WP:POVTITLE with a brief couple of sentence directing the reader to WP:NPOV#Naming and then change the redirect WP:POVTITLE to point to WP:NPOV#Naming.-- PBS (talk) 06:03, 20 August 2011 (UTC)[reply]

Should we eliminate it from one place or the other? My concern originally was that Born2cycle was applying POVTITLE as if it meant that it's alway neutral to just use the most common thing that something is called. The section at Naming was more nuanced, indicating that things need to be weighed. Dicklyon (talk) 06:48, 20 August 2011 (UTC)[reply]

No, it makes sense to discuss the issue on both pages. There is an important difference between each section. WP:POVTITLE focuses exclusively situations where the title of the article might be seen as POV or biased, while WP:NPOV#Naming focuses more on what names are used in the text of the article. These do not necessarily have to be the same. For example... per WP:COMMONNAME we use Saint Petersburg as the title for our main article on the city in Russia... but per WP:NPOV#Naming we use "Leningrad" in some sections of the text (depending on what historical era the text is discussing).

However, (again per WP:COMMONNAME... and POVTITLE) we use the name "Leningrad" in the title of our article on the Siege of Leningrad (even though it is potentially POV). Blueboar (talk) 13:07, 20 August 2011 (UTC)[reply]

WP:NPOV#Naming was never meant to cover text inside an article is was specifically written and maintained for Article Names (hence the section name "Naming" -- it is just that the section header was not changed when this policy page was moved to its new name). What part of the section WP:NPOV#Naming do you think covers names inside an article? -- PBS (talk) 14:56, 20 August 2011 (UTC)[reply]

The fact that the second paragraph starts with "This advice especially applies to article titles" implies that the advice (in the first paragraph) applies to more than just article titles. Blueboar (talk) 20:58, 22 August 2011 (UTC)[reply]

Is it considered acceptable to use Unicode superscripts and subscripts in article titles, like in C² Centauri? Or would they be considered special characters, and thus avoided in article titles?--十八 20:55, 9 August 2011 (UTC)[reply]

I think superscripts like this are fine. We have several articles that use them. The software does not seem glitch over the difference between a "2" and a "²"... for example if you type "E=MC2" you get hits for articles that use E=MC² in the title as well as articles that use E=MC2. However, I am guessing that this was not always the case... since the disambiguation page is entitlted: E=MC2 (disambiguation)... I am guessing that it was created a long time ago, when the software could not handle superscripts. (and we should probably change it to E=MC² (disambiguation) ). Blueboar (talk) 21:09, 9 August 2011 (UTC)[reply]

I'm beginning to regret that I ever added bullet points for naming criteria to this page. They were not intended to be rules that people have to follow, and I think it's becoming abundantly clear that they're not a very accurate description of how we title articles. Therefore, they're broken, as policy. A strict reading of WP:CRITERIA does not enjoy consensus support. The discussion that I've just closed at Talk:Climatic Research Unit email controversy#Requested Move makes that quite apparent.

So what should we do with this page? Burn it for warmth? Rewrite it to accurately describe (never prescribe) how we actually make titling decisions? Base everything on hit-counts at Google books? Stop reading policy pages, because they represent legalistic cancer that should be ignored?

I like 3 of those 4 ideas. :) -GTBacchus^(talk) 18:48, 11 August 2011 (UTC)[reply]

All of our policies are a combination of describing and prescribing. If there was no prescribing, there would be no point to describe. If the criteria are deficient, that's reason to improve them, not ditch them. Don't throw out the baby with the bathwater. --Born2cycle (talk) 19:13, 11 August 2011 (UTC)[reply]

"If there was no prescribing, there would be no point to describe". That's false. The point of describing is to give people a starting point, and some ideas to consider. It's not binding, though, and the less it's treated as prescription, the better. The idea that our policies are binding rules is a cancer that we need to constantly fight against.

You might notice that I haven't thrown any babies, nor any water, anywhere. However, I think that bullet point criteria are waaaay too likely to be taken as rules, and frankly B2c, you're the strongest evidence of that claim. I'm worried that you're on a crusade to make this policy into a set of deterministic rules, and I think that is destructive. I don't think there's any lack of good faith; I just don't think you understand the role of policy here.

As for what to actually do, I'm really stumped. I don't want to do over at Talk:China what I did at Talk:Mohandas Karamchand Gandhi, because I think that gives the wrong idea about how these things should go. I've just been asked to attempt some kind of closure there, and I'm scratching my head over it. I'm not going to remove the criteria from this policy anytime soon, but I am going to think about how to improve the situation.

I don't like the way you've been using the criteria, B2c; I think it's harmful. Are you reading consensus from observations, or are you pushing for something that you think ought to be?

I also don't buy the argument that following sources is automatically neutral, and I don't think the community buys that argument. I could be wrong about any or all of this. I don't know the answers, and I'm suspicious of anyone who says they do. -GTBacchus^(talk) 20:44, 11 August 2011 (UTC)[reply]

"Are you reading consensus from observations, or..."—I think of it as reading consensus from observations and discussions at places like VP or the policy's talk page in order to craft the policy, and then in "local discussions" like a move discussion the policies carry a lot of weight, unless there is consensus that it's an exceptional case/etc. ErikHaugen (talk | contribs) 23:30, 11 August 2011 (UTC)[reply]

Yeah, that's backwards of how I think of it. Abstract discussions here and at places like VP aren't worth very much. Policy is determined in the field, on a case-by-case basis, and then we try to describe the emergent best practices here. Policy is not made on this page, but in thousands of move discussions. Local discussions are precisely where real policy is created.

When I first added a list of criteria to the top of this page, I was trying to describe what people actually do in naming discussions. They don't pull up a list of 5 criteria and start making check-marks, and I hope they don't start. -GTBacchus^(talk) 23:45, 11 August 2011 (UTC)[reply]

(ec)Really? I think they're pretty accurate for subjects that don't elicit a lot of controversy; although I could be off. And see Wikipedia_talk:Article_titles#.22Considering_title_changes.22 above. Then there's always a balance between the discussion at a particular move request vs. the consensus at pages like this one: who wins when they come to different results? ErikHaugen (talk | contribs) 19:15, 11 August 2011 (UTC)[reply]

(ec) They're kind of accurate, yeah. I just think it looks too cut-and-dried.

For non-controversial titling questions, nobody even consults this page, because the answer is generally obvious. The policy is really here to provide some guidance on difficult disputes, and I think we're not doing that too well.

I really don't know what the solution is; I just do my best following community consensus in closing move requests. I don't think there's consensus to approach each titling question with the criteria in hand, expecting them to lead us to the right title. That is, unfortunately in my view, what a couple of people seem to be pushing for. It might be worth a broadly advertized RfC.

I'm doing a lot of thinking about this. -GTBacchus^(talk) 20:44, 11 August 2011 (UTC)[reply]

If you don't like the idea of policies prescribing what to do, then what is the point of an RFC? Wouldn't that be tantamount to a policy prescribing what to do? I'm confused. Or is the RFC about whether the policy should be prescriptive? How would that have any more authority than a policy in the first place? ErikHaugen (talk | contribs) 23:30, 11 August 2011 (UTC)[reply]

An RfC is a "request for comment". It's a conversation. It's where we ask people how to proceed, and people converse. We find out what the community thinks. My hunch is that the community doesn't want - and doesn't in practice follow - a deterministic set of rules for deciding on article titles. How can I find out if my hunch is correct except by asking people?

I don't know why you linked LOCALCONSENSUS there. When I say policies aren't prescriptive, a more appropriate link is WP:IAR -GTBacchus^(talk) 23:41, 11 August 2011 (UTC)[reply]

Exactly, they work for 99.95% (guessing) of our titles. So, because people ignore them for .05% of the titles - let's throw them out? Makes no sense. --Born2cycle (talk) 20:34, 11 August 2011 (UTC)[reply]

No, Born. For 99.5%, they're not needed because it's obvious what to do. The 0.5% of cases where we need guidance are the cases where we're failing to document what the community really wants to do. I don't think the community wants a list of criteria to deterministically resolve seriously difficult titling disputes. Those are the disputes the community wants to evaluate on case-by-case merits. At least that's what I think I've observed. -GTBacchus^(talk) 20:44, 11 August 2011 (UTC)[reply]

I don't see that as a failure. We may want to discuss the difference between "Climategate" and Boston Massacre, if we can agree what it is; but the upshot of that discussion is what most of us have long known: we do not necessarily adopt the common name, when it is not consensus, and not neutral. Septentrionalis PMAnderson 22:50, 11 August 2011 (UTC)[reply]

One take may be whether to use a name or a descriptive title; one of the ways "Climategate" was opposed was on the grounds that it is not now (and may never be) generally accepted as a name. Septentrionalis PMAnderson 23:20, 11 August 2011 (UTC)[reply]

(ec)I don't think the discussion was a failure; I just think it shows that the idea of using WP:CRITERIA as an absolute set of rules does not enjoy consensus support. I'm not sure I expressed myself as well as possible in starting this thread. I just want to discourage people from taking that list of 5 criteria as a final word, or even a particularly definitive word, on article titling. I don't like the way they're being thrown around as if we're supposed to go to them in order to make titling decisions.

"Broken as policy" was unduly strong wording, no doubt. I'm just tired of seeing some junk I wrote down one day elevated to "Principal naming criteria" and quoted left and right. It's obvious to me that we don't decide titles by consulting that list. -GTBacchus^(talk) 23:37, 11 August 2011 (UTC)[reply]

In general that's a good argument to make, because if it's not the name then it won't be natural, recognizable, etc. But in this case usage in reliable sources indicates it does not apply. --Born2cycle (talk) 23:31, 11 August 2011 (UTC)[reply]

Re: because if it's not the name then it won't be natural, recognizable, etc. Huh?... something like US Strategy during World War II isn't a name... yet I would say it meets all of the five criteria. Blueboar (talk) 00:35, 12 August 2011 (UTC)[reply]

I don't think we want to get rid of the concepts of Recognizably, Naturalness, Precision, Conciseness and Consistency... as broad principles to guide us in figuring out the best titles they are excellent... but I agree that they were never intended to be a set of rules to follow. Principles are what underlie "the rules", and what "the rules" are based upon. I suppose you could say that something like WP:COMMONNAME is a policy "rule", one based upon the broad principles of Recognizably and Naturalness.

Perhaps part of the problem was in calling them "Criteria" (a term which can imply a set of "rules")... what if we call them what they are: "principles". Blueboar (talk) 00:25, 12 August 2011 (UTC)[reply]

Oh, great. "The principal principles"! These principles are criteria.. "A principle or standard by which something may be judged or decided". --Born2cycle (talk) 00:36, 12 August 2011 (UTC)[reply]

Born, what I disagree with is carrying them around to move discussions, citing and listing them, and using them to make decisions. They're (an incomplete set of) ideas in the background, not a framework for making decisions. We don't decide titles by going down a checklist, or by applying an algorithm. That's what I'm worried the bullet-list format encourages, and I want to discourage it. -GTBacchus^(talk) 00:40, 12 August 2011 (UTC)[reply]

Hmmm... are you suggesting we drop the bullet list format, but keep the language? Blueboar (talk) 00:44, 12 August 2011 (UTC)[reply]

Maybe. Maybe we just need to add language around the bullet list explaining how it's not intended to be used. We could probably add a criterion (or "principle") or two as well. Neutrality is clearly a criterion that people like to use, and so is accuracy. I know- I know about accuracy, and the problems around it. Still, when it's a simple question of, say, a company that changed its name, accuracy is precisely the criterion that we use. This happens every day, and it generally doesn't even involve the RM process. Also with neutrality, it's a fantasy to say that the community supports the COMMONNAME=NPOV claim with any kind of consensus. We actually think about neutrality in titling sometimes, employing our judgment when it seems appropriate to do so. -GTBacchus^(talk) 02:19, 12 August 2011 (UTC)[reply]

[2] Submitted for consideration. The dots points are now properties that we have found consensus titles often have, rather than principles to be applied in naming. Hesperian 02:34, 12 August 2011 (UTC)[reply]

I find he phrase "Consensus favours" to be cryptic, at best. Also British ;^}. Dicklyon (talk) 04:56, 12 August 2011 (UTC)[reply]

That's cool; it isn't an example of my finest prose, just a possible new angle on what we want to say. Go ahead and polish it, including fixing my accent. Hesperian 05:17, 12 August 2011 (UTC)[reply]

British, really? So should we keep it, per WP:ENGVAR? (That was a joke.) I'd say that's exactly the spirit I'm talking about, Hesperian. I'm not too fussed about the exact wording. -GTBacchus^(talk) 05:40, 12 August 2011 (UTC)[reply]

Being British myself I find "consensus favours" to be natural. But while looking at the edit and trying to think of a better wording I had a thought. Would be useful to rephrase the points as questions to be considered? To use the first two points to illustrate:

• Recognizability – is the candidate title a recognizable name or description of the topic?
• Naturalness – what title(s) are readers are most likely to look for in order to find the article? Which title(s) will editors most naturally use to link from other articles? Such titles usually convey what the subject is actually called in English.

My reasoning is that they are and should be questions that editors use to guide their thinking, not items on a tick-list ("check-list" to our friends across the puddle). Presenting them as questions might serve to guide discussion, rather than create a battleground for competing ticklists. Jakew (talk) 11:57, 12 August 2011 (UTC)[reply]

Rephrasing as questions works for me. Blueboar (talk) 12:34, 12 August 2011 (UTC)[reply]

I like them as questions also. Helps make the heuristic function of the guideline clearer as opposed to the illusion of simplistic black or white criteria. older ≠ wiser 13:23, 12 August 2011 (UTC)[reply]

"Tick list"? Really? That's almost as bad as "whilst"!

Presenting the criteria as questions is better than presenting them as rules, but I most prefer presenting them as descriptions of what generally happens. I won't fight about it, though. -GTBacchus^(talk) 15:18, 12 August 2011 (UTC)[reply]

• Chuckles* Hmm, I see what you're saying, but I wonder if we could do both. Suppose we clarify the function of the questions in the opening paragraph(s). For example, what if we said something like, "In discussions about page titles, consensus has generally formed around answers to the following questions: [...]"? Do you think that would be an accurate thing to say? Jakew (talk) 15:39, 12 August 2011 (UTC)[reply]

Yeah, I think something like that could work. -GTBacchus^(talk) 17:46, 12 August 2011 (UTC)[reply]

(unindenting) Since we seem to have broad agreement on this change, I've implemented it. Jakew (talk) 14:26, 14 August 2011 (UTC)[reply]

Question... since we have changed "criteria" to "questions"... what should we do about the fact that we have shortcuts that still use the term "criteria"? Not saying we need to get rid of the shortcuts, but I am a bit concerned that they might send a mixed message. Blueboar (talk) 15:04, 14 August 2011 (UTC)[reply]

Hmm, that's a good point. It would probably make sense to create some new shortcuts in the near future: how about TITLEQUESTIONS for a start?. I'm not sure whether it's a good idea to delete the old redirects or not: their existence would reduce confusion when reading old discussions, but it might create some confusion if people continue to cite them using the old criteria-based names. Jakew (talk) 15:27, 14 August 2011 (UTC)[reply]

I agree that we probably should not delete them outright (that would cause problems for existing archives and discussions)... I think the best solution is to simply omit listing them on the policy page itself. They can still exist as redirects ... but would do so behind the scenes. This would allow them to slowly fade away as far as usage goes. Blueboar (talk) 15:40, 14 August 2011 (UTC)[reply]

Sorry — I completely misunderstood there! Yes, I agree, that's a good solution. Jakew (talk) 15:46, 14 August 2011 (UTC)[reply]

When did WP:COMMONNAME come to be such a big and powerful provision, apparently written to supersede the other considerations? Consider where it was at the beginning of September, 2009, when it was just one provision parallel to the others, called "Use common names of persons and things" that just gave this sensible advice:

Convention: Except where other accepted Wikipedia naming conventions give a different indication, title an article using the most common name of the person or thing that is the subject of the article (making the title unique when necessary as described in the following section and in the disambiguation guideline. Where articles have descriptive names, the given name must be neutrally worded.

That month of Sept. 2009 was the "inflationary period", driven by high-temperature gases largely from Born2cycle and Pmanderson. It stretched in and out with lots of pushback, to finish the month with a much bigger section, but still much smaller than today's, called "Use Common Names", incorporating Born2cycle's usual call for "normally titled using the most common", saying: [3]

Articles are normally titled using the most common English-language name of a person or thing that is the subject of the article. If the article's subject has no evident name, a concise, recognizable and neutral description is used instead. In determining what this name is, we follow the usage of reliable sources. As part of this, the name chosen for an article, while in common use, should be neither vulgar nor pedantic: readers will not expect such names to be the title of an article in an encyclopedia such as Wikipedia.

Occasionally, specific subject domains may follow a standardised precedent that is not strictly the common name. This practice is often controversial, and should not be adopted unless it produces clear benefits outweighing the use of uncommon names; when it is, the article titles adopted should follow a neutral and common convention specific to that subject domain, and they otherwise adhere to the general principles in naming articles on Wikipedia. The decision to adopt such a convention may be influenced by factors such as:

• Most of the articles on the subject do have ambiguous common names, so that the convention extends a standardized disambiguation to articles which do not need disambiguation
• Many articles deal with subjects with several common names,
• There is no obvious method to determine which names are the most common or otherwise suitable common names are ambiguous.

This includes PMA's favorite follow the sources language.

Born2cycle inserted "most commonly used" in multiple other provisions as well, but some of them didn't survive.

It seems clear that there was little consensus; all this inflation was hard fought; lots on the talk page, but no clear consensus (as noted on 20 Sept 2009), and no apparent attempt to assess opinions. And today, Born2cycle is relying on policy he wrote into here, on multiple fronts, to disempower editors from exercising editorial judgement, turning control over to people who make claims about how common different names are in sources, for example as was done at Talk:Crepe#Crepe.

Maybe it's time for a process to figure out what a consensus version of COMMONNAME would look like? Dicklyon (talk) 06:01, 12 August 2011 (UTC)[reply]

• This is interesting. I've been working in RM since 2006, and I hardly ever read policy pages, so I think my perspective is based on community interaction, and not on edits to this page. I feel like COMMONNAME has been invoked a lot, since well before 2009, but to check my memory, I can always consult User:GTBacchus/RM closings.

  I started closing move requests in September 2006, and the first time I closed one where someone cites COMMONNAME (using that shortcut) appears to be here, in December of that year, but it didn't carry the day. Same here. Then at the end of December is the first time I moved a page per COMMONNAME, here. That was followed by this and this. Here it failed, here it succeeded, and here it was mentioned. Another success, and another. That gets us to the end of January 2007, so in those first 5 months, I closed 284 moves, where COMMONNAME was directly cited 10 times, "winning" about half the time. I know this is statistically meaningless, but I do like Memory Lane.

  I think COMMONNAME is popular because it's simple to understand and apply, and it gives us the "right answer" a high percentage of the time. What do I mean by the "right answer"? I mean a title that sticks, and isn't reversed in the next move request a few months later. It doesn't do this all the time, however. I think it would be accurate to say that we very often give consensus support to titles that are consistent with COMMONNAME, but in contentious cases, all bets are off, and the community does what it does, which is to weigh the individual case on merits in long and complicated discussions.

  I consider these to be a sign of good health for the project, and don't want to make them algorithmic and cut down the length and verbiage. We learn a lot in those discussions, and not all of what we learn needs to be stated in policy.

  Anyway, those are my late-night thoughts on the matter. I guess I'll sleep on it, and see you gentlemen tomorrow. :) -GTBacchus^(talk) 07:06, 12 August 2011 (UTC)[reply]

In response to the original question, I'm not seeing how COMMONNAME has become "more powerful" as a result of the changes made to this page - I would say it's been given less emphasis, not more, since it is now acknowledged that commonness of name is just one of a number of criteria that we use to determine titles, not the one overriding one that must be followed unless some other written convention gives explicit permission to deviate from it. On the other hand, I think the common name principle has always been and remains one of the most often used criteria in practice - and perhaps is slowly becoming even more so (we no longer have that once-standard monstrosity "Victoria of the United Kingdom", for example).--Kotniski (talk) 08:00, 12 August 2011 (UTC)[reply]

My impression is that it has gone from one of many to a top-level section on its own. There are a number of RMs where Born2cycle seems to be saying that you don't need to look beyond commonname (and even there, bad interpretation of Google hits has been allowed to drive moves such as Talk:Crepe#Crepe; the result there doesn't seem defensible, but correcting a bad move is always hard, as you can see there). He invokes the "most common" language especially in his campaign to strengthen and apply WP:POVTITLE (which was added as a part of, or extention of, COMMONNAME), essentially overriding what others would consider to be sensible interpretations of NPOV. It just seems odd, so I'm looking for ways to get better balance into such RMs, based on consensus. From the history and the controversies that it gets involved it, it's hard to say that this section has consensus. Dicklyon (talk) 15:07, 12 August 2011 (UTC)[reply]

Would it help if we clearly stated that ultimately "Article titles are chosen based on discussion and consensus"... EVERYTHING else we say in this policy can be seen as a broad statement of intent, geared towards guiding editors in achieving a consensus ... but there are exceptions and caveats to every rule, and every principle must be applied with a degree of editorial judgment. Blueboar (talk) 16:29, 12 August 2011 (UTC)[reply]

Dicklyon, I realize you don't like my close at crepe, but really this is a complain with how the common name was decided, right? It's not really a complain about COMMONNAME. ErikHaugen (talk | contribs) 16:44, 12 August 2011 (UTC)[reply]

I don't see this as a referendum on your close there, or on you in any way. It's probably worth considering the status of COMMONNAME separate from any particular recent event. -GTBacchus^(talk) 18:23, 12 August 2011 (UTC)[reply]

I didn't mean to imply that anyone thought it was a referendum. I'm just trying to separate disagreements about how to apply COMMONNAME from disagreements about how important COMMONNAME is. Does that make sense? ErikHaugen (talk | contribs) 18:51, 12 August 2011 (UTC)[reply]

Oh, ok. Sorry. I guess I see "how to apply COMMONNAME" and "how important COMMONNAME is" as being difficult to separate. -GTBacchus^(talk) 19:06, 12 August 2011 (UTC)[reply]

Yes, I was commenting on Born2cycle's misapplication of COMMONNAME. But also on the status of COMMONNAME as a big section in parallel to the section "Deciding on an article title," when it used to be just one provision in such a list of principles. It seems that a number of editors including Born2cycle has wanted to make it the key criterion, to keep the process more mechanical, and then when he applies it mechanically he makes errors that make a mockery of it. Dicklyon (talk) 17:22, 16 August 2011 (UTC)[reply]

I think the community tends to be pretty reasonable in applying COMMONNAME. In practice we don't weigh it more heavily than all other guidelines. Whether that's accurately reflected on the policy page is... kind of irrelevant to me, but I know that many people believe in word-magic, and read policy pages. We make exceptions to COMMONNAME for various reasons, including but not limited to: ENGVAR, project-specific naming conventions, NPOV, and MOSTM, to name the ones that spring immediately to mind. -GTBacchus^(talk) 18:35, 16 August 2011 (UTC)[reply]

It wouldn't bother me either, if I hadn't seen it invoked so much recently as if it's the only thing that matters. Like at Talk:Calculator where it is said to trump Precision, and Talk:Female genital mutilation and Talk:Climatic Research Unit email controversy where some argue that it should trump NPOV (or that it should redefine NPOV). Dicklyon (talk) 22:32, 16 August 2011 (UTC)[reply]

The second sentence of COMMONNAME does seem to contradict the previous section: 'Wikipedia does not necessarily use the subject's "official" name as an article title; it instead uses the name that is most frequently used to refer to the subject in English-language reliable sources.' That seems to render the previous section null and void, doesn't it? Should we change that to 'often it uses'? Jakew (talk) 19:06, 16 August 2011 (UTC)[reply]

Sure, make the policy internally consistent, but please be sure that it reflects actual practice. We actually use official names quite often, changing article names when radio stations change their call-letters for example, and only when there is some contention does COMMONNAME supersede. In cases such as those radio stations, we don't wait for common usage to catch up to the official name, because of what I'd call "common sense". -GTBacchus^(talk) 19:15, 16 August 2011 (UTC)[reply]

No... the point that is being made here is that sometimes we use the "official" name, and sometimes we don't use the "official" name (and instead use something else) ... To answer the question of whether to use the official name or something else, we apply WP:COMMONNAME. Blueboar (talk) 23:34, 16 August 2011 (UTC)[reply]

But it's the wording of WP:COMMONNAME that we're discussing... I changed it to "prefers to use" rather than "uses", since clearly we don't always use the commonest name as defined in that sentence, as we have five criteria not one. (I think this is the "contradiction" that was actually being pointed out by Jakew above.)--Kotniski (talk) 06:22, 17 August 2011 (UTC)[reply]

Yes, thank you for putting it so much more eloquently, Kotniski. I think your fix is good, but I wonder if we need to say something to the effect that we try to meet the principles in the section above, and using the commonest name tends to be the best way to do that. That is, emphasising that the principles are the primary goal, and commonality is a means to an end. Thoughts? Jakew (talk) 09:55, 17 August 2011 (UTC)[reply]

Blueboar, what I'm saying is that, in practice, we don't always use COMMONNAME to decide whether or not to use the official name. Very often, we just go straight to the official name, without even considering COMMONNAME, and that's that. It's only when there's contention that we start to think about naming criteria. I think it's fair that policy should reflect this reality. -GTBacchus^(talk) 13:32, 17 August 2011 (UTC)[reply]

@No one in particular -- Although I've grown somewhat blasé about it, I've been complaining about the fetishization of Common Name since at least November 2006, coincidentally in another earlier contentious discussion with Serge, who became Born2Cycle. older ≠ wiser 13:58, 17 August 2011 (UTC)[reply]

It's a symptom of the larger problem of the fetishization of rules. If Wikipedia were a society, with rules that were meant to insure justice and prevent oppression, it would be one thing. Turns out, though, we're an encyclopedia. -GTBacchus^(talk) 16:19, 17 August 2011 (UTC)[reply]

I agree, but would add and emphasize NPOV encyclopedia. The problem I see with WP:COMMONNAME is that it doesn't emphasize that it often ignores NPOV in order to improve the encyclopedia, and that this is not to be done lightly. It also has systemic bias in defining commonality and while this is trivial is say, Caesium (instead of the more "common" Cesium) or Automobile (the more common "Car"), it does becomes an issue of extraordinary importance when the "commonness" itself is verifiably under controversy, such as with "climategate" over at Climatic Research Unit email controversy. Pointing out to editors that suspending NPOV is not to be done lightly, or even to consider descriptive or other alternatives to common names if the common name itself is in controversy, shouldn't be seen as extreme position: its defending a cornerstone of the project! I think the problem is that only a very few editors really belief in NPOV, or believe NPOV means compromise and consensus in which they try to cajole as much as they can for "their side", and hence they do not want to be reminded that NPOV matters and created a bunch of rules to codify exceptions and caveats and loopholes using "consensus". WP:TRIFECTA beautifully illustrates this. We are not supposed to be here to be making or obeying rules but making an NPOV encyclopedia of great quality... and fuck anything that gets in the way of that.--Cerejota (talk) 08:44, 18 August 2011 (UTC)[reply]

GTBacchus - I accept that COMMONNAME is cited a lot at RM. But how often does it ever actually trump the other concerns I listed - ie when the move is seriously contested (and not just by single obsessive POV editors)? (Apologies for not working out a quick way of answering this myself) I think Cerejota raises an important question about the importance of NPOV - how many editors really support it? I think s/he is wrong to say that very few editors believe in NPOV, but that's beside the point. More interestingly, even if NPOV editors are in the minority overall, they are the ones who end up winning arguments across the encyclopedia. My impression is that COMMONNAME by itself is not taken as a powerful argument where other doubts exist. If wikilawyering is defined as choosing ends and finding policy to support those ends, then my (admittedly less well-informed that your) impression is that that is one of the major uses of commonname.VsevolodKrolikov (talk) 09:21, 18 August 2011 (UTC)[reply]

Let me clarify. I think the immense majority of us are attracted to the project precisely because of NPOV. I know I was. So that's not my argument. My argument is that a lot of the rules exist as collection of the exceptions to NPOV people needed to make at some point. I find it interesting that things like "Words to Avoid", which were born out historically directly from NPOV and defend NPOV, get thrown to the side and disrespected and even reduced to meaningless, whereas exceptions to NPOV like COMMONNAME which exist as a codified WP:IAR (ie NPOV could itself keep us from improving quality) get used all the time, even when there is no controversy. This is similar to how people throw WP:CIVIL down the drain against vandals and malicious editors, which often lead them to think that if they label someone a vandal or a malicious editor they can get away with WP:CIVIL violations - which then promotes a negative editing environment. I am a rules minimalist, but if they exist, we do need COMMONNAME as rule because it is useful, but it should also draw a bright line that says "I stopped being a tool for improvement, so I need to be ignored". JUst think the countless hours saved actually improving Climatic Research Unit email controversy, rather than bickering about the title like we did? And all that bickering was a result of a lack of clarity in COMMONNAME that neutrality should only be compromised when doing so doesn't compromise the article neutrality itself - how does a reader approaching the topic feel if the article is named after a term the article shows is in dispute? Who would the reader feel the encyclopedic voice supports? The article that taught me this is Israel and apartheid analogy, which began with the COMMONNAME "Israeli apartheid". Naming it Israel and apartheid analogy has proven to be the single most important step towards improving that article and turning it into an encyclopedia article. It is still a POV magnet, and has many issues to improve, but it used to be nearly perma-tagged with {{NPOV}} and yet it has remained untagged for a while (1RR from arbcom also helped). All of that was made possible by choosing a descriptive title rather than a COMMONNAME - changing the title because of NPOV made the encyclopedia better. We shouldn't ignore that.--Cerejota (talk) 09:49, 18 August 2011 (UTC)[reply]

In that sense I agree with you. I particularly like the elegance of the view that many rules beyond the pillars are basically elaborating IAR exceptions to the pillars. I too believe that descriptive titles are not as heinous a solution to intractable disputes as some seem to think. One problem is that a few editors have studiously conflated neologisms and descriptive titles, and understood the ban on the former to be a ban on the latter. Your point that"neutrality should only be compromised when doing so doesn't compromise the article neutrality itself" is very well put. Perhaps that should go into policy, as I think it expresses one of the main reasons why NPOV trumps COMMONNAME in discussions.VsevolodKrolikov (talk) 11:22, 18 August 2011 (UTC)[reply]

(Unindenting) Just a thought: there's clearly debate over the relative importance of neutrality and frequency of use, but it's also clear that they do matter, at least sometimes. So would it be worth adding bullet points for both of these to the "Deciding on an article title" section (with appropriate caveats)? The effect of this would be to make commonness one of several principles — I think that's consistent with practice, do others agree? Jakew (talk) 09:39, 18 August 2011 (UTC)[reply]

I would not.

Why is use of common names valuable? Becasue they are recognizable and natural; they have no independent virtue.

Again, why, and when, are neutral terms valuable? When we are inventing a descriptive term; when a common name exists, and is universally used, we have it as the article title: Katyn massacre.

So one is superfluous; the other limited to what is already said. Septentrionalis PMAnderson 14:43, 19 August 2011 (UTC)[reply]

Actually, that's a good idea, PMA: just make it clear that COMMONNAME is in support of recognizability and naturalness, and has no other independent principle going for it. Then, any title that is common enough to be widely recognizable and natural is good enough, and we can defuse some of the counting nonsense. Dicklyon (talk) 15:13, 19 August 2011 (UTC)[reply]

I was going to agree with Jakew, but PMA's analysis changed my mind (sorry Jakew). Commonality is a useful way of expressing naturalness and recognisability. And it also raised something that has been bothering me - the way that current policy suggests that there is only one common name for something. Instead, Commonname should allow for more than one potential title to be admitted for consideration on grounds of widespread attestation. An example: I've been involved in a move discussion at East-West schism. The desired move was to Great Schism, which seems to be generally opposed because that title applies to another event too. In looking into it, I realised that in the literature two commonly used names exist that would be appropriate - the current one and Great Schism of 1054. For me, the essential idea of COMMONNAME would mean that both qualify as suitable. The choice then is not a count, but a style preference of some sort. This would mean that if (as an imaginary example) one title has 55% usage and another 45% usage, if the 55% used one raises serious enough concerns on (for example) POV, worldwide or ambiguity grounds, then the 45% one, which is also widely attested but not problematic, also qualifies perfectly well on grounds of commonality. It's attested enough; we lose little or nothing in recognisability and naturalness, but at great gains in other areas.VsevolodKrolikov (talk) 15:50, 19 August 2011 (UTC)[reply]

Okay. I think we need to clarify the relationship between the main principles and commonness, and clearly and explicitly stating that the latter is based on naturalness & recognisability works for me. I do still think, however, that there's room for adding neutrality, as this is a principle that we do consider, always in descriptive titles, and occasionally in the case of common names (eg., the community's rejection of "climategate"). Obviously we'd need caveats and a link to the section giving more info, but I think it should be included in recognition of the fact that does affect the choice of some titles. Thoughts? Jakew (talk) 16:26, 19 August 2011 (UTC)[reply]

I agree neutrality needs to be mentioned, as its absence from COMMONNAME has led some people to imply that NPOV somehow doesn't apply. There are two separate neutrality issues, however. There are titles with "words to avoid" in, and then there are titles which in Cerejota's excellent formulation, "compromise the neutrality of the article". (Katyn massacre is an example of the former (acceptable through near universal usage), Climategate the latter (not acceptable because of the implication that there was a cover-up of climate science discovered).VsevolodKrolikov (talk) 16:39, 19 August 2011 (UTC)[reply]

I have added The most common name for a subject is often used as a title because it is recognizable and natural; one should also ask [#Deciding on an article title|questions] outlined above; ambiguous or inaccurate names for the article subject, as determined by reliable sources, are often avoided even though they may be more frequently used by reliable sources. For a discussion of neutrality in titles, see [#Neutrality in article titles|below]. When there are several names for a subject, all of them fairly common, and the most common has problems, it is perfectly reasonable to choose one of the others. (The brackets are piped links to the sections immediately above and below.) This appears to reflect this discussion without saying anything that will inspire dissent. Septentrionalis PMAnderson 17:12, 22 August 2011 (UTC)[reply]

Pretty good, except I am going to changing some wording, I hope un-controversially: "Neutrality is also to be taken into account, for a discussion of neutrality in titles, see [#Neutrality in article titles|below]." It is more direct. The way you worded it makes neutrality seem as an optional intellectual exercise, when in fact it is part and parcel of article naming and in particular in considering "commoness". --Cerejota (talk) 19:18, 22 August 2011 (UTC)[reply]

That is a non-trivial and potentially controversial change; one reason I only mentioned neutrality as a link is that the next section gives a limited role to neutrality; Boston Massacre is the title because it is the term everybody uses, even though it would not otherwise be neutral; even the sentence The Boston Massacre was not a massacre is the natural phrasing. Thoughts? Septentrionalis PMAnderson 21:50, 22 August 2011 (UTC)[reply]

Would substituting "NPOV" for "neutrality" work? Titles should always be consistent with NPOV, but WP:NPOV#Naming does sometimes permit titles that are not, strictly speaking, neutral. Jakew (talk) 22:08, 22 August 2011 (UTC)[reply]

The point that bares repeating is that our titles (like everything else in Wikipedia) should be based on the sources... and the sources do not need to be neutral... we do. That means we present what the sources say with neutrality. When a significant majority of reliable sources all use the same words or terms as a name for something, to not use that name would mean that we are not editing with neutrality. We are substituting our own personal opinions and POV over that of the sources. In other words... a common name like Boston Massacre actually is a neutral title, even if it may seem non-neutral. Blueboar (talk) 00:03, 23 August 2011 (UTC)[reply]

I disagree: "significant majority" is the wrong standard there. We should accept a title if it is used in "almost all" or an "overwhelming majority" of reliable sources (like your Boston Massacre example), but if it's a significant majority using a term with obvious bias, and there's a significant minority using an obviously more neutral term, as often happens in controversial areas, then it's a no-brainer for us to choose the more neutral one. The problem is how to write it to give us that much (or some appropriate amount of) editorial discretion. We need to stay consistent with WP:NPOV, which says "NPOV means representing fairly, proportionately, and as far as possible without bias, all significant views that have been published by reliable sources;" I don't think that choosing one biased POV term when less biased ones exist could be considered fairly, proportionately, and without bias, even if that term has a majority behind it. In this I support Pmanderson's added sentence, "When there are several names for a subject, all of them fairly common, and the most common has problems, it is perfectly reasonable to choose one of the others." Dicklyon (talk) 00:30, 23 August 2011 (UTC)[reply]

Which is better for an article title: "Death of" or "Murder of"? There has been an edit/move war at Murder of Keith Blakelock. The argument for 'Murder' relies on it being the common name, while proponents of 'Death' cite NPOV and it being a more encyclopaedic title. Is there a possibility of 'one size fits all' for this or do we think it best to be a per article basis? violet/riga _^[talk] 12:02, 14 August 2011 (UTC)[reply]

Appropriate titles are always determined on an article by article basis... guided by the broad principles laid out in this policy. In this case, we can be guided by, and apply the same principle that we do for titles that include the term "allegation" ... While no one is currently charged with the crime, there has been a finding of murder in a court of law, it is therefore appropriate (and not POV) to use the term "Murder" in the title. Blueboar (talk) 12:34, 14 August 2011 (UTC)[reply]

Well yes, but that only covers one part of the argument for 'Death of'. Is "Death of..." a more encyclopaedic and consistent title?

When it is clear someone was murdered, there is nothing POV about using murder, as Blueboar said. When the overwhelming story of the person in the article is their murder. using "death of" is possibly misleading and, while not a euphemism exactly, has a similar effect (if the crime is not the main focus, then normally the article just takes the person's name). I don't know what "encyclopedic" means in this context but I am always suspicious when people say anything is "unencyclopedic" without assigning substance to that charge. We generally use the common name for articles, and we should almost never invent our own name where there is a common name. So if this was unquestionably a murder, and that is the focus of the article, and it is the common name, death of would be inappropriate. Note that we have many articles using murder of in the title. See here.--Fuhghettaboutit (talk) 13:25, 14 August 2011 (UTC)[reply]

Just because you don't quite understand the use of "unencyclopaedic" does not mean that it should be viewed with suspicion. The point is that "Murder of" is more sensationalistic than "Death of". I totally understand all the policies that we currently have - I raised it here for discussion of a common convention rather than ways existing policies fit in. There is sense in having a common naming convention across articles, though there is also sense in using common names. violet/riga _^[talk] 15:54, 14 August 2011 (UTC)[reply]

We can not have a common convention on the use or non-use of a specific word. Sometimes that word will be appropriate and at other times it will not. Blueboar (talk) 15:59, 14 August 2011 (UTC)[reply]

A murder is always a death therefore it is possible in most cases if that were decided. violet/riga _^[talk] 16:30, 14 August 2011 (UTC)[reply]

You misunderstand. "Unencyclopedic" standing alone is tautological. "X doesn't belong in an encyclopedia!" "Why?" Because it's unencyclopedic." Circular reasoning without substance. Here, for titles, you have identified sensationalism as a concern but sensationalism implies there's an element of false exaggeration, and here there is none I see, just a word that aptly describes what occurred. You're right that death is a broader set. All murders involve death, and the reverse is not true. That does not mean using death as a stand in for murder is not deceptive. When a person says someone died, we normally take that to implicitly mean they were not murdered (or assassinated or committed suicide) because that is the way we use language for such matters. We use specifics unless we are hiding what happened. As I said, it smacks of euphemism. This is all beside the issue that we should almost always use the common name, if indeed there is one here.--Fuhghettaboutit (talk) 18:06, 14 August 2011 (UTC)[reply]

I do understand as I've been a party to such discussion for many a year. While some people bandy the term around I assure you that I did not. I was using it to illustrate the the use of professional, mature, and balanced language. Moreover I value consistency and the use of "Death of ..." in some places compared to "Murder of ..." in others can show a lack of coordination. Your point about euphemisms and unintended biases from a "Death of ..." is an interesting one, and such a viewpoint was what I hoped this discussion would raise. violet/riga _^[talk] 19:29, 14 August 2011 (UTC)[reply]

The title will be dependent on the article content. if the article is about a murder case and the background to the case and a murder trial etc. then it should be Murder of e.g. Murder of Sarah Payne or Murder of Danielle Jones. This is also where the subject of the article is only notable through being murdered. Death of is where the subject is already notable and the circumstances surrounding their death are also noteworthy but are not murder such as Death of Michael Jackson or Death of John Lennon. Both subjects are notable and the deaths are also notable due to the subjects and the reasons behind the death are also discussed and there is no focus on criminal aspects.--Lucy-marie (talk) 19:44, 14 August 2011 (UTC)[reply]

I'm liking this thought, though there are some exceptions. Death of Mark Duggan for example is only notable for his death but it is immediately controversial giving any police shooting a "Murder of ..." title especially before any true investigation is complete. However in such cases WP:COMMONNAME would likely apply. violet/riga _^[talk] 20:06, 14 August 2011 (UTC)[reply]

John Lennon's death was not a murder? That is a novel view of the situation, regardless of the article title. Rlendog (talk) 14:59, 15 August 2011 (UTC)[reply]

It certainly depends on the specifics. It's also worth remembering that figuring out the True™ title is often easier when we have some distance from the event, and that WP:There is no deadline. In fact, many articles that people might title as "Murder of..." turn out to be non-notable, and are almost unheard of after the initial, short-lived media frenzy dies down. WhatamIdoing (talk) 02:32, 15 August 2011 (UTC)[reply]

In any event the word "murder" should never be used on WP unless a competent legal authority (court, coroner, etc.) has declared it to be so. Conversely refusing to use "murder" when it is clearly relevant could be inapropriate euphemism. I'm not so sure that notablility only for being murdered or if the victim is already notable for other reasons is a relevant criterion for deciding the use of the word "murder". (Would for example John Lennon's death be more deserving of being called a murder than the death of an otherwise non-notable bank teller killed during a notable robbery?) Roger (talk) 07:27, 15 August 2011 (UTC)[reply]

The non-notable bank tell probably would not rate a stand alone article under WP:ONEEVENT, so we would not have to choose between "Murder of Joe Bankteller" and "Death of Joe Bankteller". We might have an article on the robbery (entitled something like: "2012 Bank of Townsville Robbery") and in that article, it might be appropriate to mention that Joe Bankteller was killed during the robbery. I agree that we would avoid the term "murder" unless someone was convicted of "murder" (as opposed to some other charge such as "Homicide"). What makes the Blakelock article so interesting (and contentious) is that three people were convicted of murder, but the convictions were subsequently overturned. Thus, we have a legal finding of "Murder", but no legal finding as to who committed the murder. Blueboar (talk) 13:57, 15 August 2011 (UTC)[reply]

What is wrong with saying "Joe Bankteller was murdered during the robbery" if the robbers have been convicted of murder? Roger (talk) 14:16, 15 August 2011 (UTC)[reply]

Nothing, assuming that's what happened, and I don't believe that Blueboar meant to imply otherwise. Also, a legal conviction isn't strictly necessary: for example, we could have a murder without a legal conviction if, say, the killer committed suicide before being convicted. WhatamIdoing (talk) 14:41, 15 August 2011 (UTC)[reply]

Or, indeed, there could be consensus that the action was murder, even if nobody was convicted for it, or the conviction were widely regarded as wrongful. (Sir Edmund Berry Godfrey may be an example for this.) But murder is a claim of fact as well as law, about the intent of the killer, among other things. Those seem to be in dispute here. Septentrionalis PMAnderson 22:30, 15 August 2011 (UTC)[reply]

Case-by-case is how we should always approach any content issues. Any bright-line prohibition is basically begging WP:IAR. I agree with Blueboar that what makes the Blakelock case interesting is the circumstances that surround it beyond the actual action (which is what makes it also be an article on itself, otherwise it would be part of the article of the riot in which he was killed. I disagree on his interpretation of legal proceedings, in overturning the conviction, the whole process, including the finding of murder is overturned. Furthermore, the controversy around the accused is a big political issue (according to the RS). I basically had little idea of this until I became involved on 2011 England riots and I renamed and article now under AfD, and people raised WP:OTHERSTUFFEXISTS arguments and I started exploring the articles and found a bunch of mostly British articles with "murder" in the title. It seems there is a media sensationalism around violent deaths in the British press, and even usually reliable sources go apeshit with any violent death. Perhaps we in the US are so jaded that a "murder" can be called a death or killing until there is a guilty verdict, but the reality is that when I went into the articles, some where indeed using "Murder" in the title in an appropriate fashion from my perspective, but some weren't. So I boldly went in, because WP:BRD is nice. My arguments are there, where they belong, the article talk, where they should be discussed.

In general terms, in this discussion what I see is a misunderstanding of what "common" and what "npov" means in the context of titles. The way I see it, the gist of WP:POVTITLE is that we have to make a balance between the need for neutrality and the need for commonality, giving more weight than we do to commonality in the article text itself, but not ignoring NPOV.

It is clear to me that some of the suggested edits to the policy seek to change this call for a nuanced and balanced consideration, into a bright-line allowing of POV titles. I do not believe the community would be comfortable pushing NPOV aside in such a fashion. One thing is in some cases err on the side of commonality because current consensus supports it, another is to diminish WP:CCC and WP:NPOV with a bright-line allow. WP:AT allows for descriptive titles, so using a name like "climategate" is not mandated or even necessarily preferred - it simply wouldn't a policy violation to do it. Most (but not all) of the people that push hard for POV titles are usually people that agree with the POV the title would support, and we need to adjust for that bias (the anti-POV bias is offset by genuine . If on top of that pre-existing bias, we adopt bright-line commonality, there wouldbe hundreds of thousands of articles that would be renamed, everything from Caesium to Automobile, not to mention politico-religious controversial articles. As you see, the petty differences of POV pushers have been wisely chosen to be stopped by not having bright-line policies on anything but BLP and behavior issues. We should continue to do so. I take NPOV over commonality in all but historical matters that are not of contemporary relevance, even if there are contemporary debates around that. Hence we shouldn't have articles titled "pro-choice" or "pro-life" or "climategate" or "murder of contemporary controversial figure for which no one has been found conclusively guilty", but we can have articles for "Peterloo massacre" or some such. --Cerejota (talk) 04:35, 16 August 2011 (UTC)[reply]

What does "contemporary" mean? In Irish politics 1641, 1649 and 1690 are still of contemporary relevance. Have you ever looked at the edit wars that take place over Palestine or eastern Europe there are admins who spend much of their time on Wikipedia trying to keep a lid on those disputes, and what about Liancourt Rocks or Catholic Church? See also Wikipedia:Working group on ethnic and cultural edit wars. If there is one common name then we should use it. That is not to say that when in cases like Liancourt Rocks where there are three common names, weighting should not be given to a neutral one, but that is different from using a descriptive name (If you are not familiar with LR then see its entry in WP:LAME#Liancourt Rocks). As Wikipedians can not agree on Catholic Church or Roman Catholic Church should we use a descriptive name? Take your example "climategate", it is no longer in the news and newspapers that contained it are not even fish and chip paper any more, so what makes it contemporary? Even if it were contemporary not alternative name has been given instead people who did not want it did not propose a different name instead then proposed a descriptive one. I think descriptive names should be discouraged if there is a proper name to be used. -- PBS (talk) 11:02, 16 August 2011 (UTC)[reply]

Contemporary means, in the way I am using it: "sufficiently controversial that not changing the title to a less common, but NPOV, article title, results in the near impossibility of generation a Good Article". So, it can even be - and has been - an issue in very old controversies in which otherwise solid editors have lost their minds. If its controversial today to the point of keeping us from improving the encyclopedia, then a title change is in order. The way I see it, the reason we use common names in titles is to improve the encyclopedia, in the case of articles titles by driving traffic and making them easy to spot. If abandoning NPOV to get a common name doesn't help to improve the encyclopedia, then why do it? --Cerejota (talk) 09:39, 17 August 2011 (UTC)[reply]

BTW, the only alternative to a non-neutral common name is not necessarily a descriptive name, we can also use an NPOV less common name.--Cerejota (talk) 09:45, 17 August 2011 (UTC)[reply]

Then take the example of the Catholic Church and Roman Catholic Church both controversial which third alternative would you recommend? -- PBS (talk) 15:03, 20 August 2011 (UTC)[reply]

There is no reasonable controversy in that one, you will need a better example: Catholic Church is how the Church itself uses, so a self-description overrides, in my view, any other consideration - the controversy is an idiotic debate - Perhaps we should ammend WP:AT say exactly that: self-description is better than neutrality *poof* that "debate". Its how I am a keen defender of WTA (or the weak version that it was re-titled to "word to watch") in titles and in most content, but throw that away for titles of book or for articles on the terms themselves.--Cerejota (talk) 13:57, 21 August 2011 (UTC)[reply]

Actually, there has been a lot of controversy (and endless debate) on that one... In the US at least the Church uses both names "officially". However, looking at English language sources, over all, a significant enough majority use "Catholic Church" (without the additional "Roman")... so WP:COMMONNAME indicates that we should use that as our title. Blueboar (talk) 20:42, 21 August 2011 (UTC)[reply]

I think we should name and document this malady. I'm coming here directly from the second disputed close in as many months at Talk:Crêpe. That page is highly educational for anyone wishing to understand how to apply COMMONNAME.

In particular, it's very easy to make a pair of Google searches, write down two numbers, and come to a conclusion. Those numbers are, however, very often misleading. If searches are analyzed as they were by User:Noetica in that discussion, the results are much more illuminating... but it requires significantly more work.

Actually clicking through to the last page of results, to get an actual count rather than an estimate, is a basic technique. A more advanced - but often enlightening - technique involves looking at the sources themselves via page previews in Google Books or at Amazon. This will reveal to the curious reader that some sources are not in fact about the correct topic, and possibly other surprises, such as the fact that Google often gets it wrong, especially when it comes to reporting on the presence or absence of diacritical marks.

Comments? -GTBacchus^(talk) 14:59, 16 August 2011 (UTC)[reply]

The point about diacritics has come up before and was discussed (now archived) on the talk page of Wikipedia:Naming conventions (use English). Hence the paragraphs in the guideline Wikipedia:Naming conventions (use English)

• Search engines are problematic unless their verdict is overwhelming; modified letters have the additional difficulties that some search engines will not distinguish between the original and modified forms, and others fail to recognize the modified letter because of optical character recognition errors.

• Google hits are generally considered unreliable for testing whether one term is more common than another, but can suggest that no single term is predominant in English. If there are fewer than 700 hits, the actual count (gotten by paging to the final page of hits) may be accurate for Google's particular corpus of English, but whether this represents all English usage is less certain. If there are more than 700 estimated hits, the number gotten by going to the last page will be wrong; Google only loads a limited number of hits, no matter how many there are. Counts over 1000 are usually estimates, and may be seriously wrong.^{[n 1]} If several competing versions of a name have roughly equal numbers (say 603 for one variant and 430 for another), there may well be divided usage. When in doubt, search results should also be evaluated with more weighting given to verifiable reliable sources than to less reliable sources (such as comments in forums, mailing lists and the like). Do consult reliable works of general reference in English.

Note

1. ^ "Climategate, Tiger, and Google hit counts: dropping the other shoe". Language Log, December 7, 2009

--PBS (talk) 00:06, 17 August 2011 (UTC)[reply]

I agree some qualification is needed. Raw search result data can very easily be misinterpreted. Google is most useful when the difference is very pronounced. But it should not be entirely discounted. In particular, Google analytic tools such as http://ngrams.googlelabs.com/ and http://www.google.com/trends can help to inform a discussion, although these shouldn't be viewed as definitive any more than raw hit counts. older ≠ wiser 01:08, 17 August 2011 (UTC)[reply]

Yes, COMMONNAME does not mean "more hits on Google". Actually reading sources is important. Blueboar (talk) 01:37, 17 August 2011 (UTC)[reply]

I agree fully. In fact, google search is actually completely useless in this sense for common names that are phrases, for those with slightly different spelling variations, and for comparison when one of the terms has homographs/homonyms. I think and always get A LOT of pushaback for this, that there is absolutely no substitute for reading the sources. --Cerejota (talk) 09:50, 17 August 2011 (UTC)[reply]

• "Actually clicking through to the last page of results, to get an actual count rather than an estimate, is a basic technique." No, it isn't: the "last page of results" only shows how many distinct pages there were in the first 1,000 results. No Google search will you give more than 1,000 results if you click through to the last page (you can check it with a search for "Wikipedia" or "Microsoft" or any term with thousands of actual results). If you start out with a supposed 200,000 resulst and the last page comes after 75 results, then the original count is probably way off. But if the "last page" comes after a few hundred results, then that is not the actual count, but just the way that Google ends after 1,000 results, i.e. a few hundred distinct results. Fram (talk) 10:24, 17 August 2011 (UTC)[reply]

This is the essence of the problem with just about every invocation of COMMONNAME I have seen. As a policy, it doesn't really help in most substantive naming disputes. If you need to go to Google to establish what the common name is, rather than consulting with a range of authoritative sources on their use and their recommendations, then there probably isn't a common name case. Indeed, what I've seen recently is the abuse via Google of COMMONNAME to try reinforce certain contemporary political discourses, just at a time when manipulation of discourse is a studied weapon in the armoury of campaign groups and political parties, particularly in the US (which dominates the English language internet).

There is a very good reason we should go with the practice of authoritative sources rather than our own estimates of general usage: we are not lexicographers, we are (essentially) anonymous internet users. They, on the other hand, have experts, stylebooks, consultants and exposure to public criticism. Having COMMONNAME as an overemphasised policy unfortunately makes us vulnerable to the pushing of original (and often very amateurish) research. The truth is, if authoritative sources don't agree in more or less even measure, we typically find criteria other than COMMONNAME for making a choice - such as NPOV, WORLDWIDE, ENGVAR and so on. "Whichever is more common" is a criterion, but it tends not to be the primary one when it's a close call. (And as people point out above, Googlehits can't be trusted to finesse beyond the bleeding obvious.)

I would rather the ambitions people had for the policy were reduced to pointing out that we make choices like Caffeine (not 1,3,7-trimethyl-1H-purine-2,6(3H,7H)-dione) and for some individuals we make choices such as Hulk Hogan (not Terry Gene Bollea), and occasionally Bill Clinton rather than William Jefferson Clinton even though the latter is probably recognised and is the formal title. Beyond that I don't believe the community puts such an active stress on COMMONNAME (rather than NPOV), except when some users find this particular policy page useful in the middle of a dispute. I get worried by talk of using commonname as a basis of "simplifying" naming disputes through "metrics". It's legislating, not reflecting considered practice.VsevolodKrolikov (talk) 12:45, 17 August 2011 (UTC)[reply]

It may be that COMMONNAME isn't often used to title articles in the first place, but when you look at the discussions that come through Wikipedia:Requested moves, it is mentioned an awful lot. Perhaps that's something we should be discouraging?

I do realize that looking at RM discussions for evidence introduces some kind of selection bias, because at RM, we only see the tiny minority of articles that (a) have a title that someone wants to change, and (b) where the move is blocked by a history existing at the proposed target. Most articles never see such a discussion. -GTBacchus^(talk) 18:33, 17 August 2011 (UTC)[reply]

Pretty much. WP:COMMONNAME gets dusted off to defend or attack "first namer bias". I have been in some pretty contentious RM's and you can basically identify the POV people as those who are screaming COMMONNAME or NPOV without argument, because in the end, thats all this, as written today, does: give a trout with which to beat your opponent, rather than a tool to improve the encyclopedia.--Cerejota (talk) 11:58, 18 August 2011 (UTC)[reply]

That's exactly the problem that I've been trying to get us to address. Can you propose a modest revision to COMMONNAME that would help make it clear how it is to be used, or with what weight, or something that would move it in the right direction? Dicklyon (talk) 15:18, 18 August 2011 (UTC)[reply]

Is my above suggestion of any help? Jakew (talk) 15:31, 18 August 2011 (UTC)[reply]

• Remember the predictive value of sampling up to 1,001 actual webpages: When counting data in 1,000 Google results webpages, then the results can often show a 97% accuracy (WP:SIZE1001). The trends found in those webpages have a 97% confidence level of predicting the data in all similar webpages (see essay: WP:Sample_size). Hence, if there are many cases of a target spelling actually used within those 1,000 webpages, then it is extremely unlikely (97% unlikely) that the target spelling is not really used in the reported "141,500 results" that Google has suggested. For that reason, when spelling-1 is used in "405 webpages" and spelling-2 is used in "425 webpages" then it is 97% correct to claim both spellings are used about equally, even without looking at the other 140,500 reported webpage hits. For that reason, even though Google only displays 1,000 search-result pages, for many cases, that data is sufficient to make a 97% solid conclusion. Moral of story: Google Search is a fairly good tool. The key issue is to warn when the search-results are not sufficient, because in many cases, they are good for a 97%-accurate prediction. -Wikid77 (talk) 23:13, 17 August 2011 (UTC)[reply]

The problem with clicking through to the last page is that Google (including Google Books) will never show more than 1000 results, even if there are millions. (Searching on never itself, and clicking through, results in 382 hits, although Google's searchbot hasn't gotten out of the titles with Never in them.) If we understood the search algorithm better, it would be a useful control on duplicate hits; it's an accurate control on rare search phrases. Septentrionalis PMAnderson 15:56, 18 August 2011 (UTC)[reply]

• Never say never: Please do not claim, "Google will never show more than 1,000 results" (because there are ways). When I searched in Google for "never" and went to the last page (19) of results, there were 938 matching webpages. Then, I searched for "never never" (Eureka!) and got a list of 857 matches. So, next I searched for "never quit" and got 919 matches. You do the math. No, here it is: 938+857+919 =2,714 matching webpages. For searches where the results are more likely to be repeated for different searches, then use the Google-minus/exclude operator, such as searching for "never" but exclude "never quit" (Google parameters: never Template:J Hence, keep running other similar, thoughtful searches, until there are over 5,000 results, as Sample size=5,000 with a margin of error (MoE) of 1.2%. Google Search is a powerful tool when people know the tips for extracting the data to analyze trends. -Wikid77 07:42, 19 August 2011 (UTC)[reply]