Wikipedia talk:Naming conventions/Proposal

Proposal

Copied from Wikipedia_talk:Naming_conventions#Proposal for further discussion:

Okay, it looks like the cat is out of the bag anyhow, so here it is. Following on from the #Disputed section above, I have drafted a proposed rewrite; enter via Wikipedia:Naming conventions/Proposal. As far as I'm concerned only the basic premise is fixed—that a range of encyclopedic values are brought to bear in naming articles, and editors can be trusted to balance those values appropriately. As long as that premise remains immutable I am happy to see the prose rewritten from the ground up. Hesperian 23:21, 22 February 2009 (UTC)

I like it. It certainly seems an improvement on what we have now. I would use a bit more of the "ideal title should" type of language though, particularly in the list of values near the top, to prevent misinterpretations when sentences are cited out of context. And given that we have to put them in some order, I would put Accessibility first, since it does seem to be the concern that has most commonly been treated as overriding.--Kotniski (talk) 06:54, 23 February 2009 (UTC)

As hesitant as I might be about this change, it does appear to be an improvement. The Consistency section may need some work to deal with conflicting naming conventions. Also there are hints that titles should make clear what the subject is. This could lead to renames for titles that are correct but would be better understood by many more readers with a longer title. I don't see that as bad, but it is something to consider. Vegaswikian (talk) 09:16, 23 February 2009 (UTC)

I think that the wording is likely to cause as many problems as it solves, because there are so many areas where there are conflict between the principles it describes. I made some points on this subject some time ago (see Wikipedia_talk:Naming conventions/Archive 11#What we've all been overlooking (I think)) I would like to reply in more detail, but am not sure where to do so. Hesperian where would you like to conduct the conversation? Also such a large change would need very wide participation to build a large consensus, because the effects of such changes can be surprising and far reaching in areas of Wikipedia which can not easily be predicted without a working knowledge of that area (unforeseen consequences). --PBS (talk) 10:41, 23 February 2009 (UTC)

Re: "so many areas where there are conflict between the principles it describes", I think the most compelling reason to adopt something along these lines is that these conflicted areas are going to exist whether this policy acknowledges their existence or not. Every example I've used here is a conflicted area right now, under the present policy.

Re: "I would like to reply in more detail, but am not sure where to do so", I think here would be best, but I don't mind if others want to push it over to Wikipedia talk:Naming conventions/Proposal.

Re: "such a large change would need very wide participation", I agree. Eventually this will have to be tagged as proposed, advertised at the pump, etcetera. But I am under no illusions as to my inability to write a solid and elegant policy from scratch all on my own, so I would prefer to see this workshopped informally for a while first.

Hesperian 10:55, 23 February 2009 (UTC)

I am going to copy this section to Wikipedia talk:Naming conventions/Proposal as it could quickly come to dominate this page. --PBS (talk) 11:09, 23 February 2009 (UTC)

• Correctness: The title should be correct and accurate.

In many areas there are no correct names, other than the names given in reliable English language sources. For example what is the correct name for Kiev or Zürich ?

correctness would give no end of problems is with the naming of Soviet offensive during World War II, the Soviets military historians had a very specific naming convention which granulates offensives quite precisely, for example there is a difference between "strategic offensive operation" "offensive operation" and a "tactical offensive operation" (and presumably the opposites "strategic defensive operations" etc). There is one particular editor who has argued long and hard that we should use the names that Soviet military history use for these operations and not the names used by western historians because western historian base their names on German military biographies written during the Cold war so they carry a POV and they are not systematic. This argument has been refuted several times at Wikipedia:WikiProject Military history, because in a general encyclopaedia like this the public are more likely to know the Battle of Berlin by that name than as Berlin Strategic Offensive Operation consisting of the "Seelow-Berlin Offensive Operation", "Settin-Rostock Offensive Operation", "Settin-Rostock Offensive Operation", "Spremberg-Torgau Offensive Operation", "Brandenberg-Ratenow Offensive Operation" among other. When enemy forces are forced into a pocket (military) during one of these attacks then they kotel (cauldron), to reflect very large, strategic, size of trapped enemy forces; a meshok (sack) to reflect operational size of trapped enemy forces; a gnezdo (nest) to reflect a tactical size of trapped enemy forces.

Apart from the fact the English sources do not usually make these distinctions, (the word cauldron is sometimes used because it is the German military term for a pocket), so the terms are not often used they are used in some translations of Russian military texts. They are firmly established terms and often there is no common English equivalent, but do we really want an general English encyclopaedia using terms that only specialists use when it is generally agreed that we should cater to "readers over editors, and for a general audience over specialists." --PBS (talk) 11:23, 23 February 2009 (UTC)

I agree with everything you've said here. The first sentence of the correctness section makes it clear, I think, that much of the time there won't be a unique correct name. All this policy proposal says is that correctness is one of our ideals, so when there is a correct title, it must be taken into consideration. That is, in deciding on the name, we should take into account the relative correctness of the candidate names.

With respect specifically to an article in your Soviet offensive example, my take on the naming dispute would be something like this:

Accessibility: The western names are far more accessible to English people.

Correctness: Strictly speaking, the Soviet names are the only correct titles of these offensives.

Precision: The names are equally precise.

Neutrality: The argument that one set of names is biased is extremely weak.

Consistency: Not applicable here (?)

Stability: There would need to be a good reason to change from the original title.

Having considered how our values impact on this case, it is clear that this decision comes down to a conflict between a highly accessible name and a little-known name that is strictly correct. Personally, as someone who doesn't fully understand the full context, I don't see either side of the argument as compelling. I would be happy to stand back and let the military history editors come to a decision. So long as they respect our values, I trust them.

And that is the whole point of this proposal. Instead of saying "thou shalt use the most easily recognised name", we say "these are our values; do what you think best.

The other angle here is that the mere existence of an argument where people are arguing for an allegedly more correct, allegedly less biased title, rather than the allegedly more accessible title, is further proof that people really do approach these issues with more than one value in their mind. The point of this proposal is to document this accurately, rather than pretending that people only care what the most easily recognised name is. Hesperian 12:30, 23 February 2009 (UTC)

The trouble with military historians like all specialists, they use jargon between themselves (being military historians, like the military, they love acronyms and code names), which means that the names they would choose are not necessarily the names which the general public would recognise. Because of the naming conventions the Military History project has had to create its own guideline, specifically to stop Wikipedians who edit in this area from over using acronyms and code names. I put it to you that if this is a value based system there will be a tendency by specialists to prefer the "correct" name over the "common" name. This is the Balkanisation of Wikipeia I have mentioned before. --PBS (talk) 13:23, 23 February 2009 (UTC)

It would be a shame if the value of correctness came to dominate the value of accessibility; just as much as at present when the later dominates the former. If the community as a whole wishes, or tends, to move in a certain direction, do you see a role for policy in preventing them from doing so? Who has the right to make a policy like that? Hesperian 13:35, 23 February 2009 (UTC)

"Accessibility" does not dominate "correctness" because in most cases there is no contradiction between "accessibility" and "correctness", as the name of Wikipedia articles is reflected in that used in reliable English language sources. The only time when these come into dispute is when there is a divide in the reliable sources. For example the sources are split on the meaning of Big Ben. Some reliable sources state that it is only the Bell, others say that by association it is the Tower, Clock and Bell. Who is to say which is "correct" as all of them are nicknames? But one thing is for sure "Big Ben" is a better name than having three articles called Clock Tower, Palace of Westminster, the "Clock in the clock tower, Palace of Westminster" and the "Great Bell in the clock tower, Palace of Westminster". There is no such thing as correctness, there is informal and formal names, Wikipedia does not use either it uses the name most often used in reliable sources, and if that name is not clear do considerations like weighting towards a formal name come into play. --PBS (talk) 10:40, 24 February 2009 (UTC)

Your example is good. But your assertion that "there is no such thing as correctness" is an extremist view that I have difficulty identifying with.

Metallica's fifth studio album is actually self-titled, but everyone calls it "The Black Album". How do you think it ended up at the title Metallica (album)? If you infer that the people who made that decision, and continue to maintain it, give some weight to the fact that "Metallica" is the more correct name, then you must concede the point that correctness is a value that people use in choosing article titles.

Hesperian 23:07, 24 February 2009 (UTC)

• Precision: The title should precisely identify the topic and scope of our articles.

This clashes with correctness in many areas the correct name. For example take the Bosnian Genocide Case a succinct title, its correctformal title is Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro) --PBS (talk) 11:23, 23 February 2009 (UTC)

Yes. The proposal acknowledges that these values are constantly in tension. The tension exists in the names themselves, in the disputes over names, in the mental process of weighing up what is the best name. The proposal doesn't create these clashes; it accurately reports their existence. Hesperian 12:30, 23 February 2009 (UTC)

At the moment they are not in tension because we use the common name, precision only comes into play if the common name clashes with another common name, or we are using descriptive names (see WP:PRECISION and WP:Disambiguation). --PBS (talk) 13:31, 23 February 2009 (UTC)

They're still in tension. At present the policy says that the tension should be resolved by ignoring all other values and focussing solely on accessibility. There is plenty of evidence that this is unsatisfactory to many editors, and unsatisfactory in many cases, and that the tension is being resolved in other ways. Hesperian 13:48, 23 February 2009 (UTC)

I think that you are using "precision" in a different way from me. I mean precise as in succinct. what do you mean? --PBS (talk) 10:40, 24 February 2009 (UTC)

Precise does not mean succinct. It means detailed or specific. In our context you may take it as synonymous with sufficiently unambiguous.

I don't believe succinctness is a value; succinctness is good to the extent that it improves accessibility.

Hesperian 22:58, 24 February 2009 (UTC)

• Neutrality: A neutral point of view is a cornerstone of Wikipedia.

This is covered by WP:NPOV#Article naming, yes we use neutral names for descriptive names but not for proper names. So it is the Black Hole of Calcutta even though it is archaic (now Kolkata) and was a creation of Victorian propaganda. --PBS (talk) 11:23, 23 February 2009 (UTC)

Septentrionalis has said the same thing, but that is not my reading of the content you've linked to. I too have concerns about how to phrase this section, which I've raised independently at WT:NC. Please help me thrash this out. Hesperian 12:30, 23 February 2009 (UTC)

Strike that; you and Septentrionalis are quite right; it is covered by the text "Where proper nouns such as names are concerned, disputes may arise over whether a particular name should be used. Wikipedia takes a descriptive rather than prescriptive approach in such cases, by using the common English language name as found in verifiable reliable sources." I will update the section. Hesperian 12:41, 23 February 2009 (UTC)

Where are you discussing this? --PBS (talk) 13:23, 23 February 2009 (UTC)

Septentrionalis brought it up at WP:NC#Disputed. I've also opened and subsequently closed a thread on his talk page. Hesperian 13:48, 23 February 2009 (UTC)

• Accessibility: The title should recognisable and meaningful to as many readers as possible.

• Consistency: Relates articles should have related titles.

Does this mean that all the articles about colour should be spelt one way or another? I know that this is not the intention but that is how some will understand the wording. Further there are lots of articles at the moment that could have a consistent names but because they are under commonly used names they are not consistent should they be? Probably not as consistency for consistency sake is attractive, and something that humans like (canals on Mars) and this is likely to unleash lots of disputes. --PBS (talk) 11:23, 23 February 2009 (UTC)

You seem to be missing the point that these are values, not rules. The proposal does not demand that every title achieve all these values simultaneously. That would be impossible. And since it is impossible for every title to achieve all these values simultaneously, we much compromise. The present policy compromises by saying "forget the rest, just choose the most accessible name". That sucks. It is horrible prescriptive. It doesn't work. People are ignoring it. This proposal steps back from that, and says "do your best, find the best compromise, we trust you." As I've said, that is what people are doing anyhow. They are taking other values into account irrespective of what the policy says. I've given plenty of examples of that.

With respect to your colour/color example, yes, having inconsistent spellings of "colour/color" does not score very well on our consistency value. But it makes up for that by scoring well on accessibility and neutrality. Conversely, mandating a consistent spelling would score well on our consistency value, but it would alienate readers who felt biased agains, scoring an F- for accessibility and neutrality. Hesperian 12:30, 23 February 2009 (UTC)

Then I had been missing the point! Rules make it simpler to apply the naming conventions. If these are are values then every name will always be open to debate at any time. Further different people will have their own POVs on the values. At the moment rules mesh well with Purpose of consensus. This seems to me to mean that all names are a matter of opinion because it is value laden, at the moment it is much easier to determine the name even if one does not like the outcome. --PBS (talk) 13:23, 23 February 2009 (UTC)

I would like for you to go over to the flora naming convention and say "your convention is much better than the general policy because it is much easier to determine the name even if one does not like the outcome."

But seriously, I see your point, but it has long been established that policies are descriptive not prescriptive. Editors really are bringing all these values to naming discussions. Decisions really are being made on the basis of values other than accessibility. How many concrete examples of that have I offered now? Lots. It is both inappropriate and impossible to enforce rules that the community has no intention of following. Hesperian 13:48, 23 February 2009 (UTC)

See my comment below--PBS (talk) 10:40, 24 February 2009 (UTC)

• Stability: The title should not be changed to another controversial title without consensus.

This is covered by the Consensus policy, and is not directly relevant to the naming conventions. It is suggested in some of the guidelines as a possible solution when other things criteria can not be agreed upon (for example color/colour) but to elevate stability to the same level as other criteria is I think a mistake. --PBS (talk) 11:23, 23 February 2009 (UTC)

This was added by Wsiegmund a little while ago. I am not yet sure how I feel about it. I think it could be made to fit in here fairly comfortably, but I also think the policy (if it becomes one) will be just fine without it. Hesperian 12:30, 23 February 2009 (UTC)

I see stability as distinct from the Consensus Policy. From the reader's point of view, it may be disconcerting and thought unprofessional if a title changes frequently for no obvious reason. The first line in the draft under this heading (copied from the existing policy) makes this point, more or less. Walter Siegmund (talk) 20:59, 24 February 2009 (UTC)

Many of the rules put into the naming convention policy and guidelines were added before the middle of 2008 when before we added to the policy "Wikipedia determines the recognizability of a name by seeing what verifiable reliable sources in English call the subject." This addition fixed lots of problems that previous specific fixes had tried to tackle. Many of those fixes are now redundant as the outcome when one looks at reliable sources is the same as that which the specific guidelines advocated. This was an example where a small change to the policy proved to be very beneficial to the policy and guidelines as a whole. --PBS (talk) 11:54, 23 February 2009 (UTC)

Yes, I can see how the reliable sources clause has wide-ranging implications, covering accessibility, correctness, precision and neutrality. Possibly some of this could be recast so that it is clear that the value emerges naturally from a commitment to follow the lead of reliable sources when it comes to naming. Hesperian 12:30, 23 February 2009 (UTC)

Yes, we use common usage of reliable sources in good English, not the web. This is not new. Septentrionalis PMAnderson 00:16, 24 February 2009 (UTC)

And we agree that this may not be "the most easily recognised name"?

Yes, there is a real elegance to this line of reasoning. If all your reliable sources on the Black Hole of Calcutta refer to that event as the Black Hole of Calcutta, then that is its name and it is not biased for Wikipedia to call it that too. If all your reliable sources on a mathematical concept spurn the most easily recognised name in favour of a more precise term, then precision must be necessary in this context, so use the more precise term. If all your reliable sources on birds use standardised common names, then by all means impose the same consistency on Wikipedia articles. If all your reliable sources ignore the actual title of the book and simply use Gulliver's Travels, then who are we to insist on pedantic accuracy?

The policy would perhaps capture this elegance if the fundamental principle was "Use what reliable sources use" rather than "Use the most easily recognised name". That we are directed to assess the latter by recourse to the former is no comfort. To put it another way:

The policy says: All we care about is what name is most easily recognised. Assess this by seeing what reliable sources use.

It ought to say: We care about lots of encyclopedic values: accessibility, accuracy, precision, consistency, neutrality. Find the ideal balance between these values by seeing what reliable sources use.

Such a change would be much nearer to a clarification (<-- there's that word) than a major policy change. It would, I think, maintain the premise that you're looking to maintain, whilst eliminating the false premise that naming is purely a popularity contest.

Remaining issues around this:

1. We want to use reliable sources that share our values: it is wrong to base our naming decision on biased sources. Do you agree?
2. This line of reasoning has nothing to say about what to do when sources don't agree on what name to use, and (perhaps equivalently) fails to describe the practices of Wikipedians when faced with this problem (e.g. the royalty and nobility convention)

Hesperian

Replying to Hesperian last posting to the bullet point Consistency: Relates articles should have related titles. which starts "But seriously, I see your point, but it has long been established that policies are descriptive not prescriptive.."

When asked directions to a place for which the giving of directions are complicated, there is an old Irish saying to such an enquiry "If I wanted to go there, I wouldn't start from here". The method of naming pages on Wikipedia has been stable for 8 years and the key paragraph has existed unaltered for just under 7 years. It would have been possible to have used other naming methods, for example each page could simply have an index number and all names could be redirects. This was not chosen as the method for naming pages, if it had then the naming of pages would not have been an issue (as all links to a page would have been through redirects). This would have put some extra strain on the servers (double look ups), but it would have been easier for editors.

But given that every page has to have a name and not an index number, the vast majority of pages are at their common name and most editors use the algorithm as described in the naming conventions for most articles. The algorithm we have to decide on a page at the moment is a simple one.

• Is there a common name in English?
• If yes then that is the name the page has -- unless it clashes with another page with the same name (see precision).
• If no then we get into a whole string of other conditions. Quite a few of those conditions have been made redundant by the introduction of determining the name by reliable sources as the reliable sources in many areas nearly always come out with the name described in the further guidance. (The reliable source criteria could not have been introduced in 2002 because WP:V did not exist until 2003 (see Wikipedia talk:Naming conventions/Archive 11#Change to policy).

For example the only English king since 1066 for whom there is any question of using a cognomen is William the Conqueror, even Richard the Lionheart is usually known as Richard I in reliable sources. The only guideline which seems to totally reject the current common convention is WP:NC (flora) and AFAICT less than 20% of those pages would not be unequivocally at their scientific names if commonly used name was to be implemented in that area. Of that 20%, given all the other general guidance, far less than that would end up at a name other than that of the name used by specialists in that field. And that is before any specific flora guideline is used. Given that, I see no particular advantage of fixing something that is not broken and introducing a new method of naming pages which AFAICT throws more problems than it purports to fix. However I am still open to being persuaded that this is the way forward, but at the moment it seems better to me to make Wikipedia:Naming conventions/Proposal/Draft an essay rather than a policy. --PBS (talk) 10:40, 24 February 2009 (UTC)

Another problem with "Correct" occurs with the diacritics, see Wikipedia talk:Naming conventions (use English)#Since when did we agree with this wording? for the latest instalment see the entry that starts "I subscribe to everything Húsönd says. ..." --PBS (talk) 14:46, 25 February 2009 (UTC)