Wikipedia talk:Category titles/Archive 4

From Wikipedia, the free encyclopedia

Archive This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
Talk archives for Wikipedia:Category titles (current talk page)
<< 1 < Archive 3 Archive 4 Archive 5 > 5 >>

Contents

Previous discussion

I've archived previous discussion in /Archive 3

Perhaps we can all agree with this?

Can all agree with the use of soft redirects for these category titles? If "Fooish things" isn't the category title, it would point the user to "Things of Foo" and likewise, "Beings of Foo" would point the user to "Fooish beings", whichever the case may be.

For an example of a soft redirect see: Category:Authors which has a soft redirect to Category:Writers. There is already the template {{categoryredirect|<target category>}} which sets this up. Even if we come up with the most brilliant standards for determining category titles, people will categorize things wrong, or look in the wrong place. Soft redirects just help point people in the right direction. As an added bonus, there will be less CfD discussions about the same fooin' categories time and time again. -- Samuel Wantman 08:26, 12 August 2005 (UTC)

  • Note that there is presently a bot request open to automatically move wrongly categorized articles following cat redirs. For instance, if Category:Foo redirects to Category:Bar, then this bot should move any article accidentally put in the former, into the latter. Without such a bot, cat redirs may actually be a bad idea, as the editors will never notice that they put their article in the wrong cat (because a non-existent cat would show up as a redlink, letting the editor know that he made a mistake; a cat redir would be blue). Radiant_>|< 08:40, August 12, 2005 (UTC)
    • Yeah such a bot arangement would be a must. We do have to be wery carefull to have a working "infrastructure" in place to avoid abuse of such a system though. If for example a vandal runs around and redirect a few dozen categories all into the same cat it could cause quite a mess. A means to "undo" all bot actions will be needed (I think Pearle already has that though). An additional safeguard might be to make the bot check the category's history to check when the redirect template was added, and then only take action if it's been in place for at least 2-3 days or so (if no one removes the tag by then it is fairly safe to asume it's not in dispute). --Sherool 16:22, 12 August 2005 (UTC)
      • I periodically check the redirected cats category and fix them manually. If/when I get my bot running, I can easily set it to do this. ∞Who?¿? 22:07, 20 August 2005 (UTC)
  • I imagine we would all afree to this, but I am unsure that it solves anything since we still have to determine which cat would soft-redirect to which. -Splash 17:00, 12 August 2005 (UTC)

A thought on moving forward

Instead of seeking consensus on category titles unilaterally, would it not be possible to use a method similar to that used at Template talk:Europe, whereby we break categories into general areas, and have polls on each area using the criteria for vote closing that each poll question must be voted by a minimum of 15 users. 70% of support is required for a position to be declared a winner (or 30% or less to be declared a loser). Once a position is declared a winner, the appropiate change must be made to the category.

I know Rick Block listed areas of categories before at Wikipedia:Category titles/Categories by country.

  • 30 in "Nationality x" format
  • 26 in "X of country" format
  • 10 inconsistent (subcategories have no dominant name format)
  • 26 in "X in country" format
  • 7 in "Country at" format (all Olympics-related)

Rick's Note: This does not count the 75 or so subcats of Category:Occupations by nationality (which are at least mostly of "nationality occupation" format, although there are some exceptions) or many of the subcats of Category:Subdivisions_by_country (which splinter by country first, so are significantly harder to classify).

So what we could do is where they are inconsistent, a poll could be made to standardise them as per the criteria mentioned above. Where they are already in X of country format a simple poll of the above to standardise that format within the category could be run. The Nationality issue could be settled by first creating some sort of consensus around whether we are to use Nationality in any category names. If we can then agree on a position, we can apply that position. The X in counter and Country at, how do we want to standardise them? Note, however, that we shouldn't unilaterally create a poll without discussing it first. Hiding talk 10:02, 15 August 2005 (UTC)

  • The idea in principle is good. However, we need an agreed-upon system to do centralized discussions such as these (and they should always start with a discussion rather than a poll). Wikipedia:Standards once more? However, since we're talking about renaming categories, seeking 70% support is far too much. A rename is basically the choice between two (or more) names, which generally only requires a 51% majority support for either choice. Note also that several people feel more strongly about have a standard than about having the standard of their particular preference. An intuitive way of doing this may be to have a poll with four options: "1.standard A", "2.standard B", "3.standard but don't care which" and "4.no standard" (one vote per person). This way, we have a consensus to standardize if 1+2+3 combined have 70% of the votes, and we can pick the standard A or B that has the most votes. Radiant_>|< 11:46, August 15, 2005 (UTC)
  • I think the reason we should set a fairly high bar on this is because it is a contentious move, and one that seems to over-ride the common usage policy. If we are setting a standard which is to be applied in the future, then I believe we should seek the highest consensus possible to strengthen the position. We are not talking about renaming categories, we are seeking to resolve a standard to which categories are renamed, which is different. It would remove the opportunity to rename a category to something other than the standard, and would therefore become a naming convention, and thus policy. I agree it is worth seeking a consensus on whether to standardise names within a category, which is why I suggested we use Rick's breakdowns, seek consensus on standardising, and then seek consensus on what to standardise to. I think your idea as to polling is not without merit, though, as long as we use the categories in groups such as those broken down by Rick. Hiding talk 15:18, 15 August 2005 (UTC)

I'm not quite understanding the proposal, Hiding. I think you mean:

  • If a group of topically related cats is inconsistent: poll among the existing options and then rename to the successful option;
  • If a group is already consistent: hold a poll to establish that format for the future.

Is that about what you mean? This could usefully incorporate Radiant!'s suggestion along the way; voters would then still have the option of rejecting any kind of standardisation at all which some are bound to want. -Splash 17:01, 15 August 2005 (UTC)

I'd like to suggest postponing areas that are not consistent until after a conclusion about any formal standards that are already consistent. This way the easier stuff is handled first and may give guidance about the more difficult areas.
Also, especially for the areas that are now inconsistent, there should be a way to both state a preference for "no standard" and a preference for which standard, if any, is chosen. I'm not sure how well I said that.
For example, I should be able to state that I prefer no standard, but if a standard is chosen, I prefer standard B.
One way this might be done: Have support and oppose votes for each option, and whichever has the highest ratio over 70 percent wins. Maurreen (talk) 17:10, 15 August 2005 (UTC)
  • Another way this might be done is holding a simple yes/no vote on whether or not to standardize (>X% support needed to standardize), and simultaneously old an approval vote between several options; vote with the most options becomes the standard (iff the vote-to-standardize passes).
  • However - since standardizing basically involves renaming and editing pages, which are 'no big deal' and can be done by any bold editor - I would like to hear why people would oppose standardization. They may argue that it's a lot of work, but as long as other people volunteer to do the work that's not a big argument. I've heard the argument that it may be confusing or harsh on newbies, but having an unstandardized jumble is more confusing, and the only thing that happens to the newbies is that their articles may end up edited or renamed. Radiant_>|< 09:29, August 16, 2005 (UTC)
  • I agree to the first point, however, I feel I have already addressed your second point, or at least my opinion on why a high consensus should be needed to set a standard. It is not simply a question of renaming, as you put it. What is being sought is a policy which will be enforced in renaming polls, let us at least be clear about that. If something is <of foo> in a category for which the standard is <fooian>, we are proposing that it be policy that that change can be made, without argument. If someone opposes the move they have to change the standard. Therefore, if we want this to become official and we go the poll route, according to Wikipedia:Consensus we need at least a 2/3 majority, although the 70% benchmark I set out is also mentioned as popularly used. As for other arguments against standardisation, see Wikipedia:How to create policy, especially the Guidelines for creating policies and guidelines section and Instruction creep, which should point out the danger of standardising too heavily. Hiding talk 11:19, 16 August 2005 (UTC)
  • That wasn't really my point. I was, out of curiosity, asking what reasons people would have to oppose standardization. I do seem to be the only one here who finds a lower percentage than 70% acceptable; we'll have to see if 70% is workable. I hope people who oppose a standards will do so for a reason pertaining to that standard, (e.g. standardizing British vs American English sounds entirely unworkable), rather than the general principle. Radiant_>|< 11:59, August 16, 2005 (UTC)
Radiant, you say above "since standardizing basically involves renaming and editing pages, which are 'no big deal' and can be done by any bold editor - I would like to hear why people would oppose standardization." The difference is that it would make renaming in one direction nearly automatic, and a much bigger deal in the other direction -- something that can't be done just by being bold. Maurreen (talk) 12:16, 16 August 2005 (UTC)
  • To the 70%, note that either side needs to achieve 70% before the poll is closed, this would be a poll with no time limit on it. I'd suggest we start on standardising a category already heavily standardised through common usage, probably X of country. I don't think on this instance we would need to specify a standard, as it is already commonly standardised. Hiding talk 12:11, 16 August 2005 (UTC)
  • I don't think it's a good idea to have no time limit, because if there isn't one then the status of the standard can vary between accepted and rejected at any given moment, which may cause confusion in the ranks. I have no objection to reopening, rediscussion or revoting at a later stage, of course. Radiant_>|< 12:28, August 16, 2005 (UTC)
  • Well, it seems to work at Template talk:Europe, but I'm open to opinion. Hiding talk 12:52, 16 August 2005 (UTC)
    • Oh, I see now. TL:Europe closes each poll once it reaches 70% support and has a minimum of 15 voters. That sounds reasonable. But see the bottom of Wikipedia:Bible verses for a example of why it's bad to have polls with no clear closing criterion :P . Radiant_>|< 13:05, August 16, 2005 (UTC)
      • Good grief, yes, we'd want to avoid that. Well the other option is to set a time limit of two-four weeks, perhaps, as votes were still coming in on this straw poll after closing, because if support for a standard fails to achieve either two thirds or 70%, whichever we choose, it automatically defaults to no standardisation, doesn't it? We can't have a position whereby we enforce a no-standardisation policy, can we? I'm beginning to see your point on why would someone oppose a standard for category naming specifically. Hiding talk 13:51, 16 August 2005 (UTC)
        • Of course, that Bible Verses thingy is an extremely badly formed proposal. My personal preference would be to simply put a time limit (one or two weeks) on the poll; then we can close the issue tidily either way, and move on. I might prefer one week, even. The recent WP:CSD expansion poll saw no substantial change in results during the second week. I'd say that if a proposal has majority support but no consensual support, then it should be considered a "good idea" but we shouldn't be enforcing it. Enforcing non-standardization sounds like an oxymoron to me :) Radiant_>|< 14:08, August 16, 2005 (UTC)
          • We could split the difference and say ten days. The problem is, as I see it, if we set a two thirds or 70% target and fall short by a very small amount, it would seem silly to have to put it to one side as a good idea when consensus may have been reached during a longer voting period. What happens to a "good idea"? Hiding talk 15:00, 16 August 2005 (UTC)
            • A "good idea" is one of those things Wikipedia has plenty of, and would be a nice reference in any vote on the issue (e.g. "I vote to do this because that page says it's a good idea"). In other words, nothing really happens with it. I believe it's reasonable to say a vote can be extended if the margin is close (e.g. "vote lasts for 7 days, 70% passes. If vote has 60%-70% support, it will be extended to 10 days, then to 14" or a variation thereof). Radiant_>|< 15:20, August 16, 2005 (UTC)

I'm not wild about extending voting. Maurreen (talk) 17:39, 16 August 2005 (UTC)

I have no particular preference on how the vote is carried out except this: The vote will have to be organised in such a way that multiple simmilar or unclear proposals don't "dilute" the vote. If we have several simmilar proposals then we should pit them against each other first untill one remain or a compromise proposal is reached and not try to resolve everyting in one huge vote (or have "alternatve" proposals pop up in the middle of a vote). The options in the final vote should be as "binary" as possible, otherwise a high consensis is next to impossible. It should also be explicitly said wether it's a simple "vote for the one you prefeer" type thing or if "oppose" votes are allowed on your least prefered one too and such. --Sherool 11:10, 18 August 2005 (UTC)

That is the thunking behind picking a group of categories already fairly well standardised, so that we can offer a binary poll on whether to standardise all exceptions to the rule or not, and use that agreed standard as a speedy rename criteria. The next step would be to move outwards until inconsistent categories are reached, where we will first ask if they should be standardised in a binary poll, and then ask what standard in a binary poll, or perhaps split the poll into two binary proposals of whether to standardise or not and to what standard. Hiding talk 11:27, 18 August 2005 (UTC)

A different voting system?

Can I suggest we use the single transferable vote system? In this, you rank your options in order of preference 1 to N. At close of poll, the votes are initially assigned according to 1st preference votes. The option with the fewest 1st preference votes is elimiinated and its votes distributed according to the second choice votes. Then the option with the fewest total votes is eliminated and its votes redistribuyted according to 3rd choices etc etc. Eventually, one option will emerge the winner.

For this system, I would suggest simply a majority among the final two options is enough for adoption since the various preference routes should be more than enough to satisfy people's various objections and the need for consensus. Note that consensus is not only "getting 70%" but also having the opposition to the chosen proposal being content-enough with it — this STV system would probably achieve that.

It'd be harder to count the votes at the end, but I'd be happy enough to do it, given a few hours. The voting might also be fractionally more complex, but not very much so. -Splash 17:20, 16 August 2005 (UTC)

I might need a little time to digest this idea before I can form an opinion. Maybe before deciding on this, it'd be good for us to take a few days to review the grouping of consistent categories. Maurreen (talk) 17:36, 16 August 2005 (UTC)

Speedy renaming

For those advocating speedy renaming, how would you allow for any flexibility or evolution? Maurreen (talk) 15:26, 15 August 2005 (UTC)

  • Sorry, how do you mean? Hiding talk 16:43, 15 August 2005 (UTC)
  • I think, though am not sure, that the question means to suggest that once a speedy rename criterion (for whatever particular case) is established, that it would be immutable from then on. However, as is the case with all policies, guidelines, conventions etc, any of them can be rediscussed at any moment — we just recently had a major discussion and some reform of the CSDs; possibly the most contentious policy of all. So even once established, they could be changed by discussion and, hopefully, consensus forming. In the meantime, the speedy rename would save on time, repetitive discussion and, in certain cases I'm sure, repetitive conflict. -Splash 16:53, 15 August 2005 (UTC)
    • Yes. Existing guidelines can be changed. And that's exactly why I wanted a (simple but effective) set of rules-of-thumb on Wikipedia:Standards. Radiant_>|< 09:34, August 16, 2005 (UTC)

As an aside, and partly to address concerns over possible exceptions, can I suggest that speedy renaming is added to the two-day-rule presently applied to most speedy deletions? That way we avoid any posssibly accidentally incorrect speedy renames but can still expedite the process. Whether or not the 2-day-rule currently applies to speedy renaming is slightly unclear. It does appear to in Wikipedia:Categories for deletion policies but appears not to in the "Speedy" section of CFD. -Splash 16:53, 15 August 2005 (UTC)

Thanks, that sounds good. Maurreen (talk) 16:58, 15 August 2005 (UTC)
  • I am unaware of a two-day-rule for speedy deletions. However regarding category renaming, I have no objection to delaying two days - I just wish to avoid seeing the same discussions over and over again on CFD. For the sake of argument, however - suppose that we have consensual support for a standard, and according to that standard Category:Foo is nominated for renaming to Category:Bar. Now suppose that User:User did not like the standard. How is this person prevented from objecting to every single rename nominated according to the standard, and starting the debate over and over again even if it is already clear that a strong majority disagrees with him? Radiant_>|< 09:34, August 16, 2005 (UTC)
    • It would have to be stipulated somewhere that if something is nominated for renaming according to the standard, the only objections to the rename would be where it doesn't meet the standard. Objections to the standard should be raised on the talk page of the standard. Any object votes that don't state why it doesn't meet the standard should be ignored. --JimmyTheWig 10:41, 16 August 2005 (UTC)
    • Wikipedia:Categories for deletion policies is badly phrased and refers to a renaming as a speedy deletion. It also says things like an empty cat can be deleted 24 hours after "blanking" — though this is hard to tell with a cat. -Splash 17:15, 16 August 2005 (UTC)

Jimmy, would it be acceptable under your criteria to object to a speedy renaming on the grounds that "This should be an exception to the standard because ..."? Maurreen (talk) 12:08, 16 August 2005 (UTC)

  • Sounds reasonable to me. For instance, assuming for the sake of argument that we were to adopt "United States foo" as a naming convention, then an obvious exception would be "American football" since that's a proper noun ("United States football"? Nah). The only thing I want to avoid is votes like "this should be an exception because I don't like the standard" if the standard has consensual support. Radiant_>|< 12:15, August 16, 2005 (UTC)
  • This sounds reasonable to me, too. So we'd have a two-day-rule for speedy renaming and would usually only entertain objections on grounds of reasonable claims to exceptions, such as Radiant!'s example. Sounds good. -Splash 17:15, 16 August 2005 (UTC)

Postpone inconsistent categories

I'd like to suggest postponing areas that are not consistent until after a conclusion about any formal standards that are already consistent. This way the easier stuff is handled first and may give guidance about the more difficult areas. Thoughts? Maurreen (talk) 14:29, 16 August 2005 (UTC)

  • I don't have a problem with that. --Kbdank71 14:46, 16 August 2005 (UTC)
  • I agree and suggested above using Wikipedia:Category titles/Categories by country#x of country from Rick's list as the first grouping of categories to run a poll on, in which we simply propose to standardise along the common usage already inherent within that group. Is that acceptable? Hiding talk 14:53, 16 August 2005 (UTC)
    • So you're saying we propose that for the "X of country" section of the page above, any future additions use the same format as what is already there? Radiant_>|< 15:20, August 16, 2005 (UTC)
      • Yes, sorry, was I not clear? Although for that group Rick noted that category:Subdivisions by country had many subcats, not analyzed, so how we take that into account I don't know. It's probably best to get Rick's thoughts before moving on. We should probably look at how to co-opt Wikipedia:Categories for deletion/Category:By country into such a proposal, if neccesary, either by noting it is the place to discuss what form the "of country" should take, or seek consensus on the standards resolved there. As to Wikipedia:Naming conventions (country-specific topics), should we discuss whether that is deprecated or to be deprecated, or is it still standing? Is there a template tag for that page to discuss what we're doing? Hiding talk 16:52, 16 August 2005 (UTC)
        • I think we should consider Wikipedia:Naming conventions (country-specific topics) deprecated as I said before. We've had a lot more input (and likely output) than that proposal, so we can probably move on from it.
        • I'm still not sure I've understood which bit of which proposal you mean to poll on. I don't want to make you repeat yourself, but I think you are saying: for, and only for, the 26 "X of Country" in Rick's list, we propose to establish by poll the existing structure as the future structure i.e. the poll would be a formalization rather than a reorganisation. I can see some sense in that, and in adopting the list of country namings suggested. I imagine we could do that without a poll though; is there anyone here who objects to the current form of those 26 "X of Country" subcats? -Splash 17:11, 16 August 2005 (UTC)
          • You've pretty much got it. Although I'm not sure if it is a formalisation or a reorg, as I don't know how well they all follow the naming convention as outlined. I believe we should have a poll for four reasons:
  1. We are seeking to create a standard to be used in speedy renaming and possibly a naming convention.
  2. We will set a precedent for polling conventions when it comes to inconsistent category titles, and will have generated good will within the community.
  3. We may well be expanding the categories coming under the scope of the formalisation when we reach the inconsistent categories, and poll on them.
  4. It will give us a strong base for moves such as deprecating Wikipedia:Naming conventions (country-specific topics) and co-opting Wikipedia:Categories for deletion/Category:By country. I would imagine that when we conclude this process we will either have no standards, a number (2-4) of standards specific to groups of categories, possibly with some categories for which no standard can be agreed, or one standard.
          • I would think once we have run this poll, we could run two polls concurrent to formalise conventions on the remaining groups Rick identified that had a convention, if we so agree, and then run one final poll to sweep up the rest, albeit one which could be broken into many proposals, one for each grouping of categories if neccesary. Since this is, on the face of it, the least contentious poll to run, it seems the best one to run first. I think we have yet to reach agreement on whether we should categorise by nationality, even if only for people related categories. Although I am happy to be corrected there. Hiding talk 17:57, 16 August 2005 (UTC)

Hiding, I essentially agree with you: "using Wikipedia:Category titles/Categories by country#x of country from Rick's list as the first grouping of categories to run a poll on, in which we simply propose to standardise along the common usage already inherent within that group."

But I think you're also suggesting that first we review Rick's analysis for any potential problems (for lack of better phrasing on my part). I agree with that also. Maurreen (talk) 17:07, 16 August 2005 (UTC)

  • Yes, that sounds sensible. Might want to consider when to use "of" and when to use "in" for example. -Splash 17:11, 16 August 2005 (UTC)
    • Well, I believe Rick grouped all the in's to another section, but yes, if anyone wants to sift through and have a look, please do. Hiding talk 17:57, 16 August 2005 (UTC)

Another thought: ultimately, do we see the intention of this discussion and polling to be that the consensus we achieve be listed as a naming convention and a speedy deletion criteria? If so, we should make a note of that in the poll description, yes? Hiding talk 17:57, 16 August 2005 (UTC)

I like the idea of starting with some subset we suspect is non-controversial. Can we perhaps generalize the set of category types for the first poll to something like "natural features of foo" (i.e. "rivers of", "mountains of", "deserts of", etc.), perhaps throwing in "government of", "history of", "languages of", and "politics of", rather than all of what seems to predominantly use "of country" from Wikipedia:Category_titles/Categories_by_country#x_of_country (I'm not sure how I'd characterize everything in the list other than "everything in this list" which is kind of unappealing to me). I think "of foo" is a de facto standard for these categories already, so all we'd be doing is making it a de jure standard as well. BTW - Hiding is correct about "in foo" vs. "of foo" - I already separated these (interestingly, most of the "in foo" seem to be "man-made objects in foo"). Followup to Hiding's question - where exactly will we codify these naming rules? IMO, Wikipedia:Categorization#General_naming_conventions seems like the right spot (Wikipedia:Naming conventions refers to it). The procedural rules around how the naming rules are used at WP:CFD seems like a separable issue, but must also be addressed. -- Rick Block (talk) 01:03, August 17, 2005 (UTC)

  • Okay. In that list, I think we should take a look at two sections: "thing OF country" and "thing IN country". While this is somewhat nitpicky, we can probably set a simple criterion for OF vs IN. At a first glance, I'd say anything geographical should be IN (that includes lakes and roads etc, and also localized events such as racing) and anything related to government should be OF (and also, history, economics and other abstract concepts). Then propose that as a standard. Radiant_>|< 14:21, August 17, 2005 (UTC)
  • By my count, our earlier poll has 29 people in support of "of country" (or "in country") when related to anything physically located within the borders thereof; 7 people oppose, 3 oppose standardization in general, and 2 do not express an opinion. That would be 74% support. Radiant_>|< 14:28, August 17, 2005 (UTC)
  • Oh yes - we are in effect amending the naming convention (and this poll has already shown over three times as much interest). I've marked the convention as 'under discussion'. Radiant_>|< 14:28, August 17, 2005 (UTC)
I think "of" vs. "in" is not nitpicky at all, and I'd prefer if we didn't try to initially institute a rule requiring wholesale renaming. In particular, the "natural feature of foo" categories (rivers, lakes, mountains, etc.) are currently nearly exclusively "of" - in fact, as I recollect these were fairly recently standardized to "of" (from a hodge podge of "in foo" and "fooish"). Are you really proposing changing these to "in"? -- Rick Block (talk) 18:20, August 17, 2005 (UTC)
I think I must repeat again that I see no reason at all to feel restricted by what we already have. Just because that's the way it is does not mean that is the way it must always be. Instituting a wholesale rename is easy, if we ask Beland nicely.
The geographic features, my gut reaction is to say "in" and allow them to appear in several countries if necessary; they do in 'real-life', after all. However, if CfD itself just recently agreed to all these (and I do seem to remember seeing the debates though opting out), then I guess we should respect that. -Splash 22:43, 17 August 2005 (UTC)
Standarization to "rivers of foo" was from April 8 (archived here), some of the "lakes of foo" were renamed May 11 (archived here), and some more "rivers of foo" on June 30 (archived here). The June 30 discussion includes a reference to "the geographic categories standard at the country level", which I assume is referring to the earlier discussions here. I truly thought "natural features of foo" (with "of") would not be controversial, since they're basically all in that format already. As far as I know we haven't written this down as a standard before, but choosing to change it seems a little perverse (we're here because we don't have standards, but in this case there is an established informal standard that has been previously cited at CFD to create consistent naming). Writing down a standard we already have, and changing procedures to make it easily enforceable, seems like a good first step. If this is actually going to be controversial, how about if we pick a different one - perhaps "history/economy/government/geography/politics of foo"? Out of 455 categories of these forms, 31 (mostly in Category:Politics_by_country) do not conform to the "x of foo" naming convention. Codifying just this much creates 31 renaming tasks and I hope would let us all think we're making at least some progress. I would really, really like to pick something to start with that we can all immediately agree on. -- Rick Block (talk) 01:14, August 18, 2005 (UTC)
I don't really mind where we start since we will have to present a binary poll to the community at large anyway. On the geography point: all those discussions reference one another and the fact that it's already like that. So it's all just internalised. I mean, where is the Amazon river? "In South America", or "of South America". Are the Alps "of France" or "in France"? "In" seems like a no-brainer to me, and "of" is gramatically dubious at best.
I would have no problem with "history/economy/etc of Foo" so yes, if you like we could set that as a new speedy-rename first. -Splash 20:13, 18 August 2005 (UTC)

Summary of consensus on procedure so far

I think this is what have agreement on so far:

  1. Any cats that go to speedy renaming will have two-day window in which users may state why the category is an exception to any standard decided upon. At that point, the cat goes up for discussion.
  2. Any poll would need a 70-30 ratio for such a standard to be set.
  3. Polling about consistent categories would be done before polling about inconsistent categories. Maurreen (talk) 17:16, 16 August 2005 (UTC)

Thoughts:

  1. I hadn't realised that was the intent of the exception. It can be gamed too easily. "I think my favourite category should be an exception" and we're into a 7 day discussion. We need to be firmer than that. I'd suggest that, if there is a no-consensus outcome after the 2-day period that only then should it get the remaining 5 days (or maybe the full 7). It'd be much neater. -Splash 22:31, 16 August 2005 (UTC)
  2. Well, ok, but I'll put money on us standardising no more than one third of what we'd like to, just because we don't agree one way or the other. -Splash 22:31, 16 August 2005 (UTC)
  3. Yes. -Splash 22:31, 16 August 2005 (UTC)
My thoughts:
  1. I think you've misunderstood this point. It won't allow gaming in that way, as a reason why has not been stated. You would have to state something like American Revolution should be exempt from this naming system as Revolution in the United States is incorrect and Revolution of the colonies of Great Britain (it might have been the UK) in North America against the Crown is too long. It's not enough to say my favourite category should be an exception without a reason, and such a vote can be disregarded. Note also that we should make some mention that this standard will over-ride common usage, so that the reason can not simply be one of common usage.
  2. That's slightly negative. It behhoves us to find a middle ground we can go out and sell and achieve consensus on. I'm not convinced we don't fundamentally agree, and it's worth finding the areas we do agree on.
  3. Yep. Hiding talk 07:34, 17 August 2005 (UTC)
OK, speedy renaming needs to be discussed further. Maurreen (talk) 07:50, 17 August 2005 (UTC)

Speedy renaming 2

Here are ideas raised so far for exceptions to speedy renaming for cats that don't conform to whatever standards are decided:

  1. Any cats that go to speedy renaming will have two-day window in which users may state why the category is an exception to any standard decided upon. At that point, the cat goes up for discussion. Inappropriate reasons include "I want." Appropriate reasons include factual accuracy and length.
    1. Common usage?
    2. Other appropriate or inappropriate reasons?
  2. If there is a no-consensus outcome after the 2-day period that only then should it get the remaining 5 days (or maybe the full 7.
  3. Within two days, there must be an objection (or maybe an objection with an "appropriate" reason) to the speedy by at least two users (or some other number to be determined). Each of these users must have been registered before the speedy nomination was made. (or some other requirement to prevent sockpuppetry) Maurreen (talk) 07:50, 17 August 2005 (UTC)
    • These suggestions taken to Wikipedia:Centralized_discussion/Speedy_category_renaming. Please respond there. Radiant_>|< 14:07, August 17, 2005 (UTC)
      • I'm not sure that's appropriate. The purpose of any poll this page is discussing is surely to create a standard that will be applied as a speedy renaming criteria. So we need to discuss how to frame the speedy renaming criteria. I think the only question we have is what makes a valid objection, although that should be clarified within the two day peeriod and decided by a closing admin. I think we all agree one can not simply oppose without expressing a reason why the standard should not be applied. I think we need to state the standard will over-rule common usage except in areas where the standard is inappropriate, as per the American revolution exaple I gave above. Hiding talk 09:48, 18 August 2005 (UTC)
  • My reason was that Maurreen's suggestions relate to speedy renaming in general, which is a broader issue than this discussion. There were some other points on SR'ing elsewhere, too. Radiant_>|< 10:34, August 18, 2005 (UTC)
    • Ah, okay. I take it we still discuss them here with regards to how they affect any poll we are proposing here? It makes sense to have all discussion relevant to the poll here, yeah? Hiding talk 11:35, 18 August 2005 (UTC)
      • Yeah, I think we should have the parts of the discussion that apply to here, here. And I'm not entirely happy at the removal of SR from the policy page as it stands; it seems like making policy vanish without asking. Still, the discussion on SR does need to be had — and it seems to have died off in its new location. -Splash 20:07, 18 August 2005 (UTC)
        • Where is the policy page that SR was removed from? Radiant_>|< 07:52, August 19, 2005 (UTC)
          • Wikipedia:Categories for deletion policies in this diff. If I thought it important enough, I'd have put it back. But whereas prior to that diff there was at least some rule (if in dodgy English) there is now only a mention of a discussion of a rule. Still, non-admins would still have to list at the appropriate section on CfD. -Splash 16:03, 19 August 2005 (UTC)
            • I'm not sure if I understand you; Speedy Renaming is explained down the bottom of that page. I'm not sure where the assertion came from that speedy deletion must wait for two days, but I've removed that as it is clearly contradicted by WP:CSD. Radiant_>|< 13:07, August 22, 2005 (UTC)

I will have to miss the fun

I am taking an indefinite break. Maurreen (talk) 09:35, 17 August 2005 (UTC)

Proposed poll question?

Okay, just to push things along, here's a proposed poll question. I'm not sure about a suffrage restriction, do we need one? If so, I'd suggest users registered prior to the 17th July, as that was when this current debate was kickstarted at the village pump.

Wikipedia has a large number of areas where the topics are categorized by country. These frequently appear on WP:CFD with a request to rename them for consistency. After much discussion and a straw poll, it has been decided to begin seeking consensus on consistent naming of categories by sections. Therefore, this poll seeks to codify the standard already inherent within the following categories and subcategories there-of, and also make such a standard a criteria for speedy rename, with the following condition:
  • Any categories that go to speedy renaming to be standardised to the agreed format will have a two-day window in which users may state why the category is an exception to said standard. If there is valid objection to the speedy by at least two users, each of these users having been registered before the speedy nomination was made, the category gets the remaining 5 days. Inappropriate reasons include "I want." Appropriate reasons include factual accuracy and length.
The group of categories we seek consensus on standardising are as follows:
The standard we are seeking to apply is foo of country, a standard already inherent within these categories through common usage of wikipedians.
This is a binary poll to last 10 days and seeks a 70% suppport consensus with the question having been voted on by a minimum of 15 users. If the poll is succesful the standard will be listed at Wikipedia:Categorization#General_naming_conventions.

What's everyone's thoughts?

We should separate the speedy-rename procedure from the standardisation proposal in terms of how it is presented. People might support one part of this proposal, but not the other. Apart from that, I think I'm happy enough with the phrasing, and things, but wonder if the suffrage might be relaxed a little — we're talking more than a month of contributions. That's not unreasonable I suppose, but what about being registered prior to the first poll: it would seem unfair to allow a hypotheical new user to vote in that one but not this. -Splash 21:54, 20 August 2005 (UTC)
Feel free to edit the proposed question to how you'd like to see it. That'll be the best way to get it where we want it, won't it? As for suffrage, yes, that sounds reasonable. That was the 4th of August, off the top of my head. Hiding talk 22:11, 20 August 2005 (UTC)

I've some questions over the speedy rename procedure, too: who will decide what is a "reasonable" objection (an admin, I presume) and how will they decide it? -Splash 21:54, 20 August 2005 (UTC)

I don't know. I'm half of the opinion that they should just be standardised and then have the requests for rename from the standard debated upon WP:CFD. We could look through all the categories and see if anything troubles us before we list the poll, though. That would allow us to determine any exceptions. Hiding talk 22:11, 20 August 2005 (UTC)
I like most of the proposal, I am not sure about how to work the suffrage. As far as SR and valid objections, I would think a valid one would be most of what was discussed here, and not I want, listed above. I wouldn't say it would require an admin to decide, as anyone who has spent a fair amount of time on any of the discussions would know a pointless objection over a valid one. Plus every comment and/or objection would be seen by other users/admins in that discussion, so it could be "un-vetoed" if need be. ∞Who?¿? 22:30, 20 August 2005 (UTC)
  • In reponse to the above... I would like to make this actionable; too many repetitive matters appear on CFD these days. Anyway, some comments. First, we don't need suffrage. Wikipedia:Suffrage has an "informal" version that is used throughout the wiki unless stated otherwise, and it basically means we can discount sockpuppet votes. Second, I'd remove the objection "by two users" part; a single objection is good enough. And third, we should give a couple of examples of valid exceptions (e.g. "Football of America"). It'd be mostly common sense. However...
  • My main problem here is that we're bureaucratizing the issue. In my opinion, we should be able to set a clear and simple guideline ("use X of country for any categories relating to political, organizational, social or historical articles"). The current list makes the issue bulky and confusing, and makes people likely to vote it down as instruction creep (I also note that our current list has 14 cats, while the main list where it came from has 25). Thus, see below.

Counterproposal

We have a couple of clear structures here. It is easier to define those structures and establish a clear guideline for that, than to establish individual guidelines for each individual category tree. Thus, a counterproposal that feels more intuitive. Basically,

  1. Geographical categories should use "in foo"
  2. Abstract categories (government, organization, etc) should use "of foo"
  3. People categories (professions, sports, etc) should use "fooish"

Note that this roughly corresponds to what is already there. Radiant_>|< 13:07, August 22, 2005 (UTC)

By country or by nationality?

Some of the meta-cats are named "thing by country" and others "thing by nationality". Uniformity would be nice. Since this seems mainly a matter of semantics, it will probably boil down to a binary vote. The vast majority of such categories use "by country" (72 to 11, or 86%).

Geographical

These categories represent objects that physically exist within the borders of their country - e.g. lakes, cities, volcanoes and libraries. There seems a tendency towards "in foo" but "of foo" would be an acceptable alternative. None of these uses "fooish".

Abstracts

These are abstract concepts, generally related to sociology, structure and/or government, that are clearly tied to one country or another. Tendency seems to be towards "of foo", with some exceptions. In particular, for some of them, "in foo" simply sounds better. It's probably a good idea to split this list in two on that criterion. The few "fooish things" here are arguably out of place, and the recent poll seems to confirm that.

Business and organizations

This has an obvious crossover with the above (and we may need to swap some of them) but this basically concerns organized groups of people. The issue seems split between "fooish" and "of/in foo", however given their abstract nature it would be easiest to group them with the former.

People

Individual people, professions etc. Tendency is obviously towards "fooish", and the recent poll was split on the issue. It seems difficult to get a consensus on this group, but especially here "fooish" simply sounds better.

Unknown

A few oddbal categories that didn't fit into my system.

Comments

There is a bigger question implied in all of this. Some "Fooish" categories use "fooish" because they are talking about the origin of the members of the category, and not their current location. So Canadian actors would include all of the Canadians now living in LA. Can this quality of originating in a place, but not necessarily still located there be what determines "Fooish"? I notice that this is not always consistent with "Fooish" categories, but I suspect it should be. The question should be "could these things that originate in Foo be found in other places? If the answer is yes, then "Fooish" is probably the best choice. If wars are sometimes fought by people of one country in someone else's country, this might be a good choice for "fooish". This would also imply "American Corporations". This distinction also sometimes seems to be the difference between "of" and "in" for the geographical categories. Rivers of Canada are sometimes also Rivers of the United States. My point is that rather than just determine which categories use "Fooish", we should also say that the reason for using "fooish" is because it has a specific meaning -- things from a country related to people, that aren't by definition located in that country. -- Samuel Wantman 16:35, 22 August 2005 (UTC)

To be honest, my first impression is that this counter proposal is a complete sidestep of all the discussion we've just engaged in over the last week where it seemed to me we had agreement on starting on one group of categories that had a clear standard already arrived at by common usage and seek to codify that before moving on to the next grouping. I'm not clear on how this counter proposal addresses that, nor exactly what it is proposing. If you are proposing we have all polls at once, then I was under the impression we had discarded that notion as per Maurreen's summaries. Otherwise, I don't find the listing of counter proposals helpful if the intention is to run them one after the other, as we are muddying water we don't need to as of yet. I thought a decision had been made to leave the nationality issue until after we had standardised other areas that didn't need to address such a split area, and then we could utilise the good faith we had built in each other and the community to tackle that area, and then similarly move on to the inconsistent categories.


I appreciate that you ultimately want to see universal guidelines, but I don't think my proposal is aimed at a single category tree, nor will the outcome neccessarily entail a single category tree approach, but rather end up with a grouping that had grown organically. Hiding talk 18:10, 22 August 2005 (UTC)

I agree with Hiding that this proposal is likely too sweeping, but I'd prefer the first poll to be some "meta" category of categories other than simply "the categories in this list". If we can't find an uncontroversial meta-category to start with (and, sigh, I fear we can't - but I still think "geographical features of foo" is a reasonable candidate) I'd rather start with a defined list. I think the problem is simply too big to tackle all at once. -- Rick Block (talk) 00:48, August 23, 2005 (UTC)
Right. The reason I trimmed the list from 25 to 14 is because the geographical ones were causing concern with debate over in or of, so that left the 14 above. My own feeling is that we should be seeking small areas where we can agree and then expand outwards. I'm quite happy for that list of 14 to be added to, if we can all agree on the additions, or for a meta-category to be chosen, if we can. However, we had previously agreed not to attempt inconsistent categories, and furthermore, I don't actually see this codification as an instance of instruction creep, and would suggest that that argument is going to be used to reject any standardisation attempt. I would have thought going to the community with a group of categories which are already of x and asking if we could standardise that was better than going and asking if we could unilaterally rename all categories, which seems more of a case of instruction creep. Hiding talk 09:00, 23 August 2005 (UTC)
  • No, we don't need to hold all polls at once (although you have to admit it's easier to get attention that way). For instance, we could start with the geographical list above, which almost universally uses the same standard. My point is that a proposal like "geographical categories should use <this> name" is more workable than a proposal like "The following N categories should use <this> name". The latter sounds arbitrary, and non-extensible (to the effect where we'd require yet another discussion on each addition to the list). If you cover the traject slowly, people will lose interest. If you ask to standardize something that is already 100% standardized, people may wonder what you're trying to accomplish. Thus I'm proposing we standardize something that is 80%-90% standardized, and clearly defined; that sounds like something most people would agree on. Radiant_>|< 09:29, August 23, 2005 (UTC)
The step from "no codified rule" to "any rule" seems like a big one. I think we should try to make it as easy as possible. I proposed "geographic feature of foo" and there was some resistance. Hiding's proposal seems like a pragmatic response to that resistance. I haven't added up the categories in the 14 classes Hiding proposes, but (per above) within only the 5 classes of "[history|economy|government|geography|politics] of foo" there are 455 categories with 31 not using an "of foo" name (93% conformance). If was can't agree on "geographic feature of foo" and Hiding's list is too unsatisfying, how about "man-made objects in foo"? This would be include:
and perhaps some others not specifically listed in Wikipedia:Category_titles/Categories_by_country (like category:Airports, which should clearly be category:Airports by country, but it uses "of" rather than "in"). I suspect any general rule will end up hitting at least one class of category currently standardized in some other form. Perhaps we want to avoid this, but perhaps we don't. -- Rick Block (talk) 14:36, August 23, 2005 (UTC)
  • Regardless of what we do, we will at some point hit upon something either not standardized or standardized in the wrong way. So I don't think that'd be a problem. Can we simultaneously propose that any categories about "man-made objects in foo" use "in foo" as a suffix, and that the parent category should be called "thing by country" (rather than "by nationality)? Radiant_>|< 14:41, August 23, 2005 (UTC)
I am back intermittently. I can't address all the details now, but I prefer Hiding's proposal to Radiant's counterproposal, mainly because of the consensus to handle the already-consistent categories first. Maurreen (talk) 17:39, 23 August 2005 (UTC)

Ok, so I feel slightly responsible for the removal of geography from this discussion. However, most if not all of these are binary questions. Do we want (a) or (b)? As such, there's not a whole bundle of point us discussing this among ourselves, since we are going to have present precisely the same two options to the community once we're done. Even supposing we were to agree on whether to use (a) or (b) there's no reason to suppose the community at large will do so. Basically, given this, I wouldn't have a problem with the purpose of this discussion being to carve the categories up into such binary questions (which most seem to be) and present a mass poll. It's either that or a series of smaller polls, and people will get tired of voting and will begin telling us that polls are...etc -Splash 17:55, 23 August 2005 (UTC)

Okay, I'm getting slightly confused here. I thought we had agreement for leaving contentious categories until last. I also thought we had agreement on trying to find a group of categories where a standard was almost inherent and seeking consensus on codifying that. If that is no longer the case, then what are we actually in agreement on? I don't agree at the moment that they are all binary polls, to the extent that some categories are either format (a), format (b), or (c) no standard, which is a three way choice. This whole discussion is starting to get circular, if I'm honest. Hiding talk 20:20, 23 August 2005 (UTC)
No, no, I'm happy enough to do the non-standardized one first if people want to. I'd just been thinking about it a bit, and had reached the conclusion I mentioned above. Still, even with that 3-way choice, it won't matter much what we agree here we think is best, it will still have to go to the community at large. So the difference any discussion here is likely to make is what the options are, rather than a determination of which is the best. But I'm ok with going with the smaller-scale version, as long as people don't think there will be poll-fatigue. -Splash 21:16, 23 August 2005 (UTC)
  • Well, geographical categories for instance are not contentious and have an almost inherent standard. But Splash is correct about vote fatigue. I'm sure we shouldn't do twenty polls simultaneously, but neither should we restrict ourself to asking one single question per week. Radiant_>|< 08:25, August 24, 2005 (UTC)
  • Hmm. It seemed as though Rick and Splash had differing opinions on the geographical categories and it was decided they were, for the time being, contentious. I'd also point out that I'm not sitting here suggesting we have twnety poll questions per week. I thought the plan was to do uncontentious categories that were settled in a format, and then work out a proposal for contentious categories that were settled in a format, and then work out a proposal for inconsistent categories. I saw it as, at most, a four stage process. I think at the minute though we all seem to be pulling in different directions. Perhaps we should take a breath and try to start again at common ground. Hiding talk 10:25, 24 August 2005 (UTC)

The geographic ones which initially seem mixed between "in Foo" and "of Foo" actually form a distinct pattern. With only two exceptions (both borderline areas - regions and parks) they are "Natural features of Foo" and "Human-made features in Foo". Given that that seems to be a de facto rule, I don't really see what the fuss is about there. Grutness...wha? 02:23, 26 August 2005 (UTC)

Lost

Having followed these discussions (and added the odd comment along the way) I am now totally lost as to who is saying what and why. I'm taking this page off my watch list.

If someone can come up with a summary of the discussion then please let me know.

--JimmyTheWig 08:44, 24 August 2005 (UTC)

American/United States

In that mind, it occurs to me to propose that on the American issue we could propose to allow the term American to be used to represent the United States on the basis of first usage, similar to the British/American spelling. If the category is named American, do not rename, and the same for United States. US and U.S. should however, be renamed to United States, and we could also propose placing a notice on categories which use American to the effect that the term is used to convey the nationality of the United States rather than to denote continental nationality. This would prevent systematic bias which could be argued if either side were chosen, and represents the split that seemed apparant on the issue. Thoughts? Hiding talk 20:05, 28 August 2005 (UTC)

This seems to presuppose the discardment of the perfectly good discussions and progress we are making chunk-by-chunk above. I don't see the reason for the sudden changing of heart. -Splash 20:17, 28 August 2005 (UTC)
I'm not sure about this. I suggest defering it. Maurreen (talk) 20:20, 28 August 2005 (UTC)
Fair play. The issue hadn't been raised for a while, I didn't really notice any discussion of it, to be honest, and this suggested itself as a nice compromise. It would be nice to have thoughts on the compromise, but I shall kick it back into the long grass. Hiding talk 20:24, 28 August 2005 (UTC)