Wikipedia talk:WikiProject Stub sorting/Archive 7#Stubs templates and WikiProjects

Yes, this is not a typo. I think a completely different approach should be used for stub marking. The current situation is a total mess where even experienced users find it difficult to choose the appropriate stub tag(s) for a given article. Actually, we already have categories, so why having "stub categories"?. If an article is, let's say, an Italy-related article and it is a stub then it is, as a plain logical consequence, an "Italy-related stub". Therefore, MediaWiki should manage everything. We should only mark the article as stub and the software should provide a way to browse either all stub articles or all stubs by categories. And it could use a special symbol (let's say [stub] or [s] to indicate stubs in categories pages. The current "solution" is absolutely unmaintainable and error-prone and wastes everyone time. —Gennaro Prota^•Talk 17:15, 24 May 2006 (UTC)

:This comes up periodically, and indeed, there's certainly some redundancy between the two. But firstly, mediawiki doesn't manage any of this, so this'd have to go via the developers.

:: So?^(gp)

:Secondly, if you think the stub sorting system is a total mess... have you seen the permanent categories lately? Stub-sorting is at least marginally easier than categorisation, as the range of options is a good deal smaller, and at least somewhat managed; and because tagging with a template is a little more convenient than with a category.

::This argument is moot. Categorization is a concept on its own. What if someone then comes with something like {{Italy-politician cleanup}}, {{Computer software cleanup}} etc.? I simply think MediaWiki has been developed without any stub marking idea in mind. Fortunately that's trivial to add. About the granularity issue, it's up to the reader to decide how deep in the tree he wants to go. And if categories are a mess that's another problem.^(gp)

:Lastly, it wouldn't achieve the same effect. Perm-cats are allowed to be small to the point of silliness (I've in the past nominated singleton cats for deletion, and had that opposed), and filtering for the stubs only would exaggerate that: you'd end up with a very deep, "stringy" system, very different from the situation of having only "viable" stub types with at least 60 articles in them each (supposedly -- naturally some people aren't happy when their "pet" over-categorised stub type is culled on grounds of mere size). Alai 22:39, 24 May 2006 (UTC)

::Now you are totally inventing rationales, sorry.

--

(gp)=Gennaro Prota^•Talk 12:06, 25 May 2006 (UTC)

I am not "inventing rationales"; please assume good faith. I've also no idea what you mean by your reply to my second point; I think I've demonstrated that the issues are far from "trivial", but as you don't address any of my points, further comment would seem, well, pointless. Alai 14:35, 25 May 2006 (UTC)

:It was not meant to be offensive. You are doing that in good faith, but you are inventing rationales, in the sense that what is allowed/what is not isn't written anywhere: you just said what you *imagine* to be the rationale for having a separate stub classification. And even if it is written somewhere, it is absolutely unfeasible, so it was an experiment which failed. As to the second point, I meant that if we want a separate tassonomy for stubs, why not having one for articles to be cleaned up too? And then why not a separate one for articles that have a bad punctuation? You see, this is an unmanageable approach. The fact that an article is a stub and the fact that it belongs to a given category are totally orthogonal (read: independent); thus the corresponding markers should be independentely applicable and usable. I also don't understand why each time a software modification is required everyone seems to be afraid. If Mediawiki developers refure to implement features which improve our work on Wikipedia we can always create a fork; MediaWiki is free software. Please, let me know if I have still not addressed all your points. —Gennaro Prota^•Talk 15:09, 25 May 2006 (UTC)

You haven't addressed the main concern - which is the viability of stub categories on the basis of size. You may look on it as inventing a rationale, but it is the main single reason why stub sorting exists. Otherwise we'd have done it by the method you suggest or similar ages ago. There is simply no point in calculating and marking stubs in cases where there may be one or two stubs per category. The whole purpose of stub sorting is to help editors to find stub articles that they can expand. To do that, it is no use to list the one or two stubs per category that need sorting - it is far more useful; to have reasonable numbers of stubs assembled into one place. Stub sorting does that by creating categories only when the number of stubs hits a reasonable threshold and splitting these categories only when the number of stubs becomes excessive. That can't be done by the methods you suggest, which is why similar suggestions have been rejected in the past. Look at it this way. You're an editor looking for stubs to expand. You might be able to expand 1/10 of the stubs around in a moderately broad subject area. Which is easier - to look through a category containing 100 stubs, picking out the ten or so you can work on, or looking through five to ten main categories each of which have one or two stubs before even finding one that you can edit? Grutness...wha? 01:19, 26 May 2006 (UTC)

:I'm wasting my energies in a pointless discussion. Please, cling to your ideas. Given that stub categories don't even follow a consistent naming convention, 80% of the times I'll not know what name to use and I'll just attach {{tl|stub}}. Then, maybe, someone else who has tons of time to waste will sharpen the category. Some of the people who contribute to Wikipedia subvert all the good rules of data processing and are even so pretentious to state they are right. What's more, they do that in a tone that almost mocks at the person who is trying to highlight their errors. What could we do? Either we should reserve maintenance tasks and design choices to people who understand how to maintain an enormous information base such as Wikipedia is (and that computers have been created to spare time, not to waste it) or we have to be contented with the a bloom of parochial approaches we already have and which don't help anyone. —Gennaro Prota^•Talk 17:08, 27 May 2006 (UTC)

::Is that another way of saying that - rather than answering the points raised - you'd prefer to attack the people raising the points? The stub categories follow as logical a naming convention as we can - every time we find a new one that someone has created out of process with a dumb name, we try to rename it according to the standard stub naming guidelines. That we have trouble keeping up is not our fault. In any case, you don't need to know the names of the categories to add stub templates. All you need to know is the names of the templates - which again are as logically formed as we can make them (you should see the fights that have occurred here in the past with people who have arbitrarily named stub templates any olfd thing rather than try to follow and easily understandable pattern!) And no-one here has made any attempt to mock you. Then again, you haven't "tried to highlight our errors", either - all you have done is point out to us a system that would not perform the same tasks as we are trying to achieve. If you can come up with a method which fulfils all the requirements of stub sorting (which so far you haven't done), come back and we'll listen. Grutness...wha? 23:26, 27 May 2006 (UTC)

I have a proposal to use per-month trancluded pages for WP:WSS/P, like Wikipedia:WikiProject Stub sorting/Proposals/June 2006. Hopefully this will simplify archiving and reduce the size of the page you load when editing.

Another proposal is to change logging scheme for WP:SFD. Currently, closed discussions are cut-and-pasted to the logs. What about transcluded per-week logs? I think that per-day logs (similar to WP:CFD) would be clearly too short (sometimes there are 0-1 nominations per day). Per-month logs, on the other hand, are going to be too long (about the combined size of these two). We can also create templates similar to {{tl|cfd top}} and {{tl|cfd bottom}} to mark closed discussions. Conscious 07:47, 27 May 2006 (UTC)

:The first, on /P, seems a clear win; it doesn't materially affect the appearance of, or procedures on, the page, and should make things easier all 'round. The second I'm not quite as sure about; it'd involve significant changes to how the page operates; in particular, actually closing discussions inside of a week, admittedly no bad idea in itself. Perhaps we should start off with the closed-discussion templates as a first step. Alai 15:17, 27 May 2006 (UTC)

::I'd definitely support it on /P. Not sure it's really necessary on SFD - it's relatively short - but {{tl|sfd top}} and {{tl|sfd bottom}} would be worthwhile, I'd think. Grutness...wha? 23:14, 27 May 2006 (UTC)

Well, is there any sense it {{tl|sfd top}} and {{tl|sfd bottom}} if one has to move discussions to logs anyway? I mean, the discussion is closed if and only if it's been moved to the log. Conscious 20:03, 28 May 2006 (UTC)

:Not strictly so; the discussion can be closed, and still be awaiting orphaning, or deletion, which currently involves c'n'p within the page. If we're going to go with transclusion, I'd also like to see the closure tags already in place and seen to be working smoothly, rather than changing several experimental variables at once, as it were. Alai 20:29, 28 May 2006 (UTC)

The reason I think they'd be useful is because of the number of times that discussions have been moved down to "old business" and been ready to deal with, then someone comes along and adds further comments. Most of the time the extra comment isn't helpful (some times, if it's a close call it may tip the scales one way or the other, but usually it makes no difference at all). Grutness...wha? 01:15, 29 May 2006 (UTC)

:I've created the templates. Artists, feel free to change the colour. Conscious 07:01, 29 May 2006 (UTC)

::looks good - but is there any way of making the text a little darker, or is it a standard "deletion template grey"? It's a bit difficult to read against that background. Grutness...wha? 10:43, 1 June 2006 (UTC)

:::The text is black, so it would make sense to change the background colour. Conscious 19:21, 7 June 2006 (UTC)

::::Once its in the box it appears as mid-grey on IE. Grutness...wha? 08:35, 8 June 2006 (UTC)

What about three-day logs, by way of an admittedly arbitrary compromise between the possibly-too-short per-day, and the possibly-too-long week? Alai 18:40, 7 June 2006 (UTC)

:Well, first, do we agree that it's a good idea to have transcluded logs at all? Re your question, I'd favour weekly logs. In addition, they could use names like /Log/2006/Week 23, generated with {{CURRENTWEEK}}. Conscious 19:21, 7 June 2006 (UTC)

::Agreement in principle's hard to factor out from agreement on the practicalities, since the whole point is what'd be the better in practice. In particular, weekly logs would mean that some items would necessarily go unlogged for that long, and half a week on the average (as opposed to the current "however long it takes", admittedly often longer). It'd also mean that any unclosed items for the whole week would hold up all those, making that both more likely, and making it affect more items when it occurs. So my current feeling is, "not if it means doing them on a weekly basis". Alai 20:22, 7 June 2006 (UTC)

:::And I just don't understand how 3-day logs would be implemented :) Conscious 06:13, 8 June 2006 (UTC)

::::It's not clear to me how much work said generation of names would save, but hypothetically, why not /Log/{{CURRENTMONTH}}/Period {{#expr:({{CURRENTDAY}}+1)/3 round 0}} or something along those lines? Alai 11:30, 9 June 2006 (UTC)

:It's not really clear to me that it would be that helpful on WP:SFD, which isn't the largest of pages. I'd be far more pleased to see it used at WP:WSS/D. Grutness...wha? 11:37, 9 June 2006 (UTC)

::I don't see that logging on any given schedule is much of an issue at /D (which is the real point of transclusion, not size). Alai 11:55, 9 June 2006 (UTC)

:::Size not an issue? You don't have a dial-up modem, by the sound of it! I start loading the page, then make a cup of coffee, come back and it's still loading. Size is a big issue as far as I'm concerned. On WP:WSS/D, where you really only need to see the latest items at the end of the page, individual month transclusions would be extremely useful. Grutness...wha? 12:31, 9 June 2006 (UTC)

::::That bears no resemblance to what I actually said. What /D needs is to be archived slightly more often than at ten-monthly intervals; that's hardly an argument for transclusion, especially if people continue not to archive. SFD, however, actually is logged on a semi-regular basis, but doing so by copying and pasting on an item-by-item basis is distinctly tiresome, and transclusion on a sensible basis could help with that. Alai 12:51, 9 June 2006 (UTC)

:::::There's no doubt we have a problem with /D. It needs to be archived at the very least, many of the entries are resolved. But what about abandoning it at all in favour of a stubberg-type list? Alai and I maintain such a list, it is generated from each database dump and contains all stub categories not listed at /ST. Discussions at /D aren't generally very lively, and aren't binding. Now imagine we transfer this load to SFD (I mean, people take some categories from stubberg to SFD from time to time). What do you think? Conscious 13:02, 9 June 2006 (UTC)

::::::I'd broadly agree with your diagnosis, not quite sure about the prescription. While I think the "unlisted list" is very handy, trouble is it's sufficiently large (approximately 450 entries) that it itself presents "management issues". Separate lists of each batch of new unlisteds from each dump might be more digestable, and would be handy both for spotting the "incoming", and as a means of tracking the undersized-but-might-grow contingent, several iterations along. We probably would still require some sort of discussion page, but if we move away from the not-very-successful model of a "tracking system" at /D, we'd probably be ahead on the deal. Alai 13:46, 9 June 2006 (UTC)

Yes, a good consideration is needed. For example, a list to watch shouldn't contain newest entries (newer that 2 months, for example) or should just be sorted by date of creation. I'm just throwing in an idea, maybe there's a way to get /D straight. Conscious 14:11, 9 June 2006 (UTC)

On third thoughts, I'm inclined to say that we should actually go with per-day transclusion and logging. It's potentially slightly more work, but equally one should be able to do a number "at one gulp" pretty readily, without the complication of a whole week being held up by one "problem case". Alai 03:39, 1 July 2006 (UTC)

We seem to have stalled somewhat: I suggest we implement some version of this, starting on the 1st August. If I'm left to my own devices, naturally it'll be done my way... I'll leave a note at WT:SFD, too. Alai 22:23, 14 July 2006 (UTC)

What do we think of the above? The coding looks dodgy to me, given the unclosed div tag, and I'm not too sure about the visual appearance, but the general idea seems to be on the lines of something we've been discussing for a while. The main stumbling block seems to be that unless we start messing around with the coding of the stub types themselves, or duplicating same, there's still the same amount of spamming with redundant information (this is a stub; please expand; yadda). If anyone has any bright ideas on how to improve... Alai 01:50, 30 May 2006 (UTC)

I hestitate to do this, since we've plenty to do as things stand, but... Preliminary poking around in the db dump indicates over 10,000 very short articles not tagged as stubs. I'm going to try to produce a cleaned-up list, possibly after the next en. db dump, but before I do so I thought I'd better check a) are people ready for the shock, and b) exactly what criteria should be used? Bluemoose was previously using a size threshold of 350 bytes; I'll use the same, unless someone has strong preference otherwise. He was also excluding any page with a template inclusion; I'm more likely to exclude on the basis of category membership. In particular, anything that looks like a disambiguation page, copyright violation tagee, deletion or wiktionary candidate. More marginal would be categories indicating a lack of sources, expansion request, merger, cleanup, incomplete lists, wikification, context, verification, and importance. I'll include or exclude these or the basis of whatever feedback I get here. Alai 03:50, 30 May 2006 (UTC)

:I'd exclude dabs, redirects, listdevs, and deletion candidates (CV or AFD) but include the rest. it'll be a long list and there are bound to be non-stubs on there, but it's more likely to get all the ones that are stubs. Grutness...wha? 05:13, 30 May 2006 (UTC)

Further investigation indicates about 7,500 of these without any categorisation, so that aspect's something of a nicety: plenty to be going on with, possibly excessively so. I won't upload those for the time being unless there's a major clamour, as it's by this stage about 12 days out of date, and I don't want to get a pile of hate-mail about whipping you into a frenzy of action(?) on the basis of possibly dud info (though I'd be surprised if it's changed that much). My crystal ball's a little misty as to when the next en. db dump will be (and no-one on meta is letting on); presumably not until after a de. dump has finished, which itself will take a number of days, and it's not yet begun; but I won't in any event be downloading it before the 7th. Alai 00:42, 31 May 2006 (UTC)

All hands to the pumps! {{Cl|Stubs}} is back to almost 400 articles... Grutness...wha? 06:48, 30 May 2006 (UTC)

Seems that User:220.253.3.186 has discovered a new way to vandalise using stub templates. Have a look at [http://en.wikipedia.org/w/index.php?title=Hopkins_School&oldid=55873348 this]. To paraphrase the great Alexei Sayle, "Stub templates are very important to this story - see how many you can spot!" Grutness...wha? 08:37, 30 May 2006 (UTC)

The following discussion at the pump about a mysterious glitch with stub icons has produced the answer to a question I've wondered about for a while. It'll be worth keeping an eye out for any other icons that have this problem!:

----

Putting the politics and economics stubs together indents the second one. Is this attributable to the size of one of the graphics?

RyanEberhart 01:41, 5 June 2006 (UTC)

:We've had that problem with some stub icons but not with others. You might want to ask this question over at Wikipedia talk:WikiProject Stub sorting - because it's worth finding out which templates cause that problem and fixing them. I think it is a size thing, but I'm not technoboffin enough to know for sure. (BTW, I "substed" those templates to remove the categories). Grutness...wha? 05:33, 5 June 2006 (UTC)

::It's the "left" in the image link for the first template (removed below). This causes the image to "float" left, allowing subsequent text (and images) to flow around it. -- Rick Block (talk) 01:47, 6 June 2006 (UTC)

:Excellent. Thanks for that - I'll try to remember that if I see it happening again! Grutness...wha? 00:00, 7 June 2006 (UTC)

----

The template {{tl|GuantanamoBay-detainee-stub}} is quite long, well over one line on my small screen. Ideas on shortening it are appreciated. - CrazyRussian talk/contribs/email 15:56, 11 June 2006 (UTC)

:Thanks, Alai. Looks good. - CrazyRussian talk/contribs/email 18:31, 11 June 2006 (UTC)

:(ec)And on mine too, at least at the default browser width. I've had a go. Alai 18:33, 11 June 2006 (UTC)

Looks persistently fried. Anybody know what's up? - CrazyRussian talk/contribs/email 00:28, 12 June 2006 (UTC)

• Wish I did. I'm guessing more problems with toolserver's database replication, but I have no more information than you. Asking on meta rarely produces much of a response. If anyone's really stuck, I may be able to run off-line queries to give similar results, if people can describe exactly what they want, and don't mind seriously ad-hoc turnaround times. Alai 18:11, 12 June 2006 (UTC)
• I took a peek at Interiot's page, and it looks like they are attempting to reload the corrupted data to the toolserver. It looks like they've taken the toolserver's entire en-wiki database offline until they've finished with the experiment. Valentinian ^(talk) 21:17, 12 June 2006 (UTC)

There is an active discussion at Template talk:Stub about changing the way stub templates are worded (originally the suggestion was to change all references to the word "stub", but that seems to - thankfully - have died down. The gist of it is that templates saying "This article is a stub" are not instantly understood by casual editors, who would be far more likely to understand something like "This article is short and undeveloped". Input from more of the old hands would be welcome! :) Grutness...wha? 08:36, 13 June 2006 (UTC)

Since I know how heavy a use this WPmakes of them, I thought I shouid point this out: [http://en.wikipedia.org/w/index.php?title=Wikipedia:Village_pump_%28technical%29&diff=55538604&oldid=55537496] (For those who don't know, Brion is the Wikimedia Chief Technical guy...) - SoM 10:51, 15 June 2006 (UTC)

:I wish he'd make up his f!@#$%^g mind. First he says "no problem", then he says this. How long before he gives a third opinion on it? Grutness...wha? 14:56, 15 June 2006 (UTC)

::I guess we're all complete and total idiots. So much for assuming good faith. Crystallina 04:02, 16 June 2006 (UTC)

:::Since clearly being Wikimedia Chief Technical Guy exempts one from normal considerations such as consensus, civility, and answering a simple question in a straight way, I dast not speculate. I'm no strong objection to them being subst'd, but if it's not an efficiency issue (and I thought that the main hit here was in regards to 'editing' heavily-used templates), then it's a little pointless to flip this from "must not" to "must" in any particular hurry (cvue robots making tens of thousands of edits). Alai 18:08, 16 June 2006 (UTC)

::::Let's not fall into the trap of returning incivility with the same coin. I *presume* he just had a bad day. I couldn't care less if we stubst these themplates or not. Valentinian ^(talk) 20:35, 20 June 2006 (UTC)

I was searching for a stub template name recently and it took a while to find it. I had a look for something like :category:stub templates (like in our Turkish Wiki) but found nothing. I went to #wikipedia on IRC, had a chat and they proposed to ask a bot to do some work...

If one analyses for example :Category:Animal stubs one needs maybe one :Category:Animal stub templates category for all those stub templates listet there but it would be nice to have a bot do it because it may be way too much work to add a Category:stub templates (or even sub categories) to every single stub template. Any ideas? --katpatuka 10:07, 24 June 2006 (UTC)

:(maybe better in Wikipedia:WikiProject_Stub_sorting/Proposals ?)--katpatuka 10:07, 24 June 2006 (UTC)

::A user has tried this one before but opinion was pretty strongly against, so the categories were deleted again. I don't think the situation has changed since then. Valentinian ^(talk) 10:31, 24 June 2006 (UTC)

:::I think this would be fairly profoundly redundant; they're listed on WP:WSS/ST, an automatically generated list is produced by a script Conscious runs to detect the "unlisted" ones as well as the above, and they're included in the stub categories themselves (both rooted at {{cl|Stub categories}}, and comprehes. Yet another method seems unnecessary, work to maintain, and clutter in the templates. Alai 18:35, 24 June 2006 (UTC)

We have {{tl|Stub Category}} and {{tl|regional stub category}} to provide some basic boilerplate for categories. I'd like to propose a third {{tl|Biography Stub Category}} to provide boilerplate that would be appropriate for all the stub biography categories. I have a proposed version at User:Caerwine/Template:Biography Stub Category that in addition to what {{tl|Stub Category}} does, includes text recommending the three basic categories all biographical articles should ideally have, year of birth, year of death/not dead yet, and what they are. Any thought for either possible improvement or abandoning this idea? Caerwine Caerwhine 01:13, 25 June 2006 (UTC)

:Seems a good idea to me. I think as well as a comment on categorisation, this might be a good place to note that occupation-based stubs types should also have a regional tag, and vice-versa. Alai 01:53, 25 June 2006 (UTC)

Over on the Commons they have a category :commons:Category:Stub pictures that has some nice pics that people might find useful for stub templates, especially the ones that mix a picture with a wiki puzzle piece. Caerwine Caerwhine 03:13, 25 June 2006 (UTC)

I'm wondering if we should have a designated place for "noting" that a stub type is undersized. At present, if a stub type is already on the list, or has been proposed, or has already survived an SFD, it's not really a "discovery" any more, though we could use that page regardless certainly; or else a section on the todo page (long-standing undersized types). I suggest this as it seems that often we take a "tiny" type to SFD, whereupon it's either populated, there's suggestions we might wait and see if it grows, or else there's protestations that it's soon to be. A less formal process might have similar effects, and perhaps with less agro on occasion. Alai 22:03, 29 June 2006 (UTC)

:A couple of points - firstly, what you describe is a very similar thing that happens on AFD - articles taken there are either deleted or "saved" by sudden rapid growth, so i don't see that as too much of a problem. As for a list of small stubs, I though someone (actually I thought you!) kept a list of those on a user sub-page. If so, that page could easily become a de jure part of WP:WSS just by adding it to the WP:WSS category, same as I have with my geo-stub count and splitting lists. If necessary, mover the user subpage to an official WP:WSS pagename. Grutness...wha? 02:33, 30 June 2006 (UTC)

::Yup, I do indeed (I assume you're referring to this, which is at any rate the most recent incarnation of much the same thing). I could certainly move it to the WSS project-space if people would find that useful and convenient. (Can't do much about the update rate, though: I'm at the mercy of whatever the heck is going on with the db dump.) That doesn't track creation date, whether there's a wikiproject, whether it's already been on SFD, etc, etc, so it's not a complete or integrated solution. I'm not suggesting types being populated while on SFD is a "problem", or otherwise a bad thing at all, I'm just wondering if there's a better way to get the same good result. Alai 02:50, 30 June 2006 (UTC)

Actually, reversing myself yet again, {{tl|popcat}} pretty much covers this. On the other hand, might it be worthwhile splitting out the stub types, say with {{tl|popstub}}? (With associated category, which one could of course then request to be populated...) Alai 22:36, 1 July 2006 (UTC)

:I've gone ahead and "forked" this as {{cl|Underpopulated stub categories}}. Alai 06:45, 18 July 2006 (UTC)

We have a mess here, especially in, but not limited to, UK and US stub types. I'd like to suggest the following additions to the naming guidelines.

1. If the parent non-stub category is Fooian X's that the stub category be Fooian X stubs.
2. If the parent non-stub category is X's in/of Foo that the stub category be Foo X stubs.

:I realize that this won't matter much to Grutness since for some inexplicable reason, Wikipedia uses New Zealand rather than New Zealander as the Fooian of New Zealand in the main categories.

Anyway, this is nice and straightforward, and I hope it will attract the support of others. Caerwine Caerwhine 15:36, 2 July 2006 (UTC)

:Well, Wikipedia's correct in that. New Zealand is one of those rare countries where the correct adjective is the same as the country's name. It's only the noun form of the people who live there that is different. I'm a New Zealander, that is, a New Zealand person, who lives in a New Zealand house in a New Zealand city. Oh, and support BTW, although it'll probably mean a lot of renaming. Grutness...wha? 06:45, 3 July 2006 (UTC)

::I thought this for a long time, wondered whether it was all due to a peculiar technical rule :-) TheGrappler 10:42, 4 July 2006 (UTC)

::) BTW, another reason this should be done is that we're slowly getting more flak in places like CFD for the differences between stub ncategory names and non-stub category names. Things like this, for instance. Grutness...wha? 12:49, 4 July 2006 (UTC)

:Strongest possible oppose. This would institutionalise the pointless flip-flopping we already see between for example, "France this" vs. "French that" on no rational basis, rather than actually do anything to fix it. Does anyone really think that {{cl|France buildings and structures stubs}} is an especially sensible name for a category? Why on earth should we utilise awkward-sounding names that would never fly in the permanent category space, to purport to be consistent with same? While not actually being so, of course, since we'd be rearranging the component words with blithe disregard to what's in any way a customary usage. To take the implications of this scheme on its face, if "geography of France" were renamed to "French geography", we'd then flip between from "France geography stubs" to "French geography stubs"? That makes no sense whatsoever. There has to be some internal logic (or indeed, any logic) to stub category names, beyond "a given part of speech appeared in a parent, let's use it somewhere or other in the stub name, after throwing away the prepositions that caused that word to be used in the first place". I'm frankly boggled Grutness would support this, having previously said exactly the opposite in at least one context, though perhaps in hindsight I shouldn't have been since we do have a lot of these "Awkward Attributive Noun geography stubs" kicking around at present. Alai 07:56, 7 July 2006 (UTC)

::My support is more for the suggestion that we should try to somehow get the stub category names in line with the main categories than it is for the actual suggestion of "Foo X stubs"/"Fooian X stubs". The problem with importing the standard category names wholesale is that it would lead to some even more cumbersome combinations such as "X of Foo stubs". But the closer we can get things to consistency with the main categories the better. Grutness...wha? 11:05, 7 July 2006 (UTC)

:::I'd gladly support "somehow in line with" myself: on some issues, they're pointlessly wayward, both internally, and wrt the perms. Case in point would be the whole people/personnel/biography/biographical business, which is completely needless. Of course, the perms are often all over the shop, too, so this is trying to hit a moving target. I'd rather have internal consistency than external, which is in fact essentially what the perms do themselves: have rules for each "layer", and horizontal consistency within that (... on a good day...) rather than fully-vertically-integrated common usage, particularly on nouns vs. adjectives. Alai 02:27, 9 July 2006 (UTC)

My argument is simply this: Abolish stub types. Articles would be tagged as stubs by any user, either just like they are now (eg {{stub}} appearing in the code) or perhaps by some other mechanism, such as Flickr-like tagging (see folksonomy). Then the articles would be able to be listed as stubs under the category (or categories) the article has had applied to it. If no category has been applied to the article then it will appear under an 'uncategorised' stub list.

This simply removes the problem of thousands of stub types doesn't it?

reinthal 07:09, 3 July 2006 (UTC)

:This proposal crops up occasionally and does sound appealing, but there are problems with it - notably that quite a lot of the time stub articles don't have categories and an uncategorised stub list would be extremely long (even if only 5% of stubs have no category - a very low estimate - that would mean a list of 20,000 stubs). See Wikipedia talk:WikiProject Stub sorting#An alternative to stubs for a recent similar suggestion. Grutness...wha? 08:24, 3 July 2006 (UTC)

::That problem would only be temporary, though. Articles with a non-general stub but no categories were presumably tagged by someone who took the time to look up the correct stub for their article, but didn't bother to add a category. This proposal would essentially force people to add a category, and remove the redundant effort involved in giving an article a specific stub only to later add the categories that are clearly implied by that stub. At its most extreme, we could try to work something out so that putting {{tl|stub}} in an article with no categories causes glowing red text 24-point text to appear at the top of the article that says "THIS ARTICLE IS UNCATEGORIZED. {{tl|stub}} requires a category. Please add a category to this article along with the stub!", much like transcluding subst-only templates does now. Of course, that wouldn't work with the number of existing stubs with no categories we have... We could always decide on a few categories that are decided by each existing stub, and run bots to slowly turn them into generic {{tl|stub}} while adding the necessary categories. This would be almost necessary in implementing this suggestion, anyway. Once those bots had finished running, the only things left in cat:uncategorized stubs would be what's currently in Category:Stubs. --Aquillion 22:54, 21 July 2006 (UTC)

:This would be a great solution (and the only really reasonable one in the long run, in my opinion), but it requires major code changes in MediaWiki to implement properly, since MediaWiki does not support category intersections at the moment. If you can write up a patch that does it, great, but in the meantime, the status quo is what we have to use. -- grm_wnr Esc 17:13, 3 July 2006 (UTC)

:There are several issues here: is a "quick technical fix" going to be available at some point; is it OK to get rid of some of the criteria of stub-sorting, like keeping stub categories within a certain size range, which the category-intersection model would make impossible, unless there was some after-the-fact way of fixing this up; and would such a system be more or less usable from the point of view of the "sorters"? None of these are clear at first inspection. I imagine we'll persist until they become moreso, or else the whole edifice collapses under the weight of the sheer amount of work required to prop it up. There's no question that there's much inefficiency built into the current sorting of sorting and re-sorting articles, even after they have permanent categories. Alai 03:01, 9 July 2006 (UTC)

Can people explain exactly what their rationale, and exact criteria for "guessable" stub category names are? Who is being assumed to be doing the guessing, and on the basis for what body of knowledge? People working from the particular permanent category? Other stub categories? One or other naming conventions page? Without further elucidation I don't know what "guessable" category name is supposed to be, much less why they're desirable, in particular above and beyond other desidirata. Alai 06:04, 11 July 2006 (UTC)

As you can see from the To Do page, we're not getting too far with stub sorting and indeed seem to be going backwards with every update bringing progressively more categories and stubs. Is there any way to fix this other than just "sort more"? Crystallina 18:06, 11 July 2006 (UTC)

:For stub sorters? Probably not. For editors in general? Expand stubs until they are no longer stubs (and remove the stub notice). Simple. :) --TheParanoidOne 20:12, 11 July 2006 (UTC)

:I'm seriously tempted to suggest redefining "oversized" to mean over 1000 articles in size... Alai 04:56, 13 July 2006 (UTC)

::Hrm. On second thought and upon reading the edit summaries I probably should not have said anything at all, carry on. Crystallina 17:48, 20 July 2006 (UTC)

I have worked with a few individuals to form a project to eliminate poor stubs and duplicate content in favor of more comprehensive articles. I agree some articles will always be "stub" in size, but those that are suspect to development in more than 1 article for the same subject matter should be merged, and maybe split off again when deemed necessary.

WP:ʃ

Cwolfsheep 23:15, 12 July 2006 (UTC)

is a quad stub. kzz* 21:45, 13 July 2006 (UTC)

:Unfortunately this regions covers five countries and we normally stub geographical articles with one template per "affected" country, so this one could have had five. This method does look a bit stupid but is an attempt to make the article appear in as many relevant categories as possible. In cases such as this, using a template for each country seems to be only possibility to treat all countries equally. But it is a prime candidate for expansion. Do we have any experts around? Valentinian ^(talk) 00:09, 14 July 2006 (UTC)

:The other option is simply to revert it to its region as a MEast-geo-stub. Grutness...wha? 01:22, 14 July 2006 (UTC)

::I agree with Grutness. If the feature covers that many countries of the Middle East, it would be better to use the single broader-term stub. -Acjelen 14:15, 14 July 2006 (UTC)

:::I can easily buy that solution. The max. is probably around three to four countries, so this one might be a bit too much. Valentinian ^(talk) 00:14, 15 July 2006 (UTC)

::I usually use a limit of four, FWIW. Grutness...wha? 01:55, 15 July 2006 (UTC)

:::I'd tend to concur. There may be circumstances where a convoluted structure of the stub types, and an unusual article argues for more stub tags, but a simple list of countries isn't one of them, nor is any similar use of "sibling" templates. Alai 03:35, 15 July 2006 (UTC)

FYI, I've made a request for approval for a bot-task for replacing double tags with "consolidated" stub types. With luck, this will help somewhat with the re-sorting backlog, though it certainly won't be a complete solution by any means. Alai 04:36, 18 July 2006 (UTC)

Talking of bot requests for approval, please note this. In extremis, what this could be is a stub-un-sorting bot, but what it really illustrates is the confused thinking about the whole "Stub-class" thing. Is "Stub-class" the same as "stub", or not? We were unable to get a straight and consistent answer to that when the question arose at SFD, with at least one person arguing to keep the two categories separate on the basis of being different (without really defining how, or addressing the issue of terminological confusion). This sounds like a possible trainwreck. Alai 16:12, 18 July 2006 (UTC)

:I've added by 2¢ to that bot request... hopefully others here might do likewise... Grutness...wha? 01:39, 19 July 2006 (UTC)

:[http://en.wikipedia.org/w/index.php?title=Category%3ASEPTA_stubs&diff=64849573&oldid=49713067 Two] [http://en.wikipedia.org/w/index.php?title=Category%3APhiladelphia_stubs&diff=64850026&oldid=42972259 changes] related to "Stub-Class" vs "Stub". The first seems the most contentious as it has replaced a legitimate stub category with a Stub Class category. --TheParanoidOne 20:52, 20 July 2006 (UTC)

::Said editor now seems to be merging Stub-class and stub categories, using stub-class on articles rather than talk pages (or indeed, on both), and generally creating a mess with them. While there's a degree of home-brew screwup at work here, I think it does also go to underscore the capacity for confusion from the terminological overlap. Alai 21:02, 22 July 2006 (UTC)

:::I've reverted these: the editor in question seems to think these are "artificial" distinctions (right, great idea having talk pages and articles in the same category, obviously). I sense this one will run and run, in one form or another, though. Alai 18:18, 4 August 2006 (UTC)

I've identified this project as a candidate for material to be analyzed by Wikipedia Integration methodology. Please feel welcome to offer suggestions and feedback. WP:ʃ Cwolfsheep 16:22, 22 July 2006 (UTC)

• Exactly what does this analysis (or indeed, methodology) involve? I see no discussion there of WSS, and the link to this project is amid links to such non-entities as Wikipedia:WikiProject Wikipedians for writing an encyclopedia. Alai 00:36, 23 July 2006 (UTC)

As well as merging double tags on stubs, I've also started using my bot to restub some articles on the basis of their permanent category, and parent stub type. See the current contents of {{cl|Non-profit organization stubs}}. This method might not find the most appropriate possible stub type, but it should at least find a notionally appropriate stub type. I'd be interested in hearing people's thoughts on this approach, and on what instances (if any, of course) they think it's appropriate to do this on, etc. Alai 03:05, 23 July 2006 (UTC)

• OK, this is now approved, so if anyone has any suggestions for to-do's for full-speed edits, please suggest away. Alai 05:03, 25 July 2006 (UTC)
• One possibility is the clothing-stub/fashion-stub mess recently mentioned at SFD. Half the items marked fashion-stub should be marked clothing-stub and vice versa (and no doubt some should be marked with both). Grutness...wha? 05:44, 25 July 2006 (UTC)
• "Ah, but which half...?" If you can suggest a "rule" for this based on the perm-cats, that may be a possibility. Alai 06:17, 25 July 2006 (UTC)
• Well, that's my point. Any with a permcat of Fashion should get fashion-stub, any with a permcat of Clothing should get a clothing-stub. It won't completely sort them out, but it should improve both stubcats immensely. Grutness...wha? 08:12, 25 July 2006 (UTC)
• Just trying to be as explicit as possible -- so I can blame you when it all goes wrong :). What about (perm-)subcats? {{cl|Fashion}}'s a subcat of {{cl|clothing}}, but I could put all contents and subcat contents of "Fashion" in that first, and then the remainder of those with other subcats {{cl|clothing}}, say. Alai 15:10, 25 July 2006 (UTC)
• This is now half-done: I've taken all the clothing-stubs with a fashion cat, and moved them over to fashion-stub. Doing it the other way around is a little more involved (I suppose that what we're looking for is fashion-stubs with another category that's also in the clothing hierarchy, but isn't also in the fashion sub-tree), but at any rate we should shortly be rid of the {{tl|clothingstub}} redirect. Alai 06:00, 7 August 2006 (UTC)

I've just created a new WPSS template for use at the top of category and template talk pages. occasionally WPSS has been tripped up by not knowing about discussion taking place on a template or category's talk page... hopefully {{tl|WPSS-talk}} will reduce the chance of that happening in future. Grutness...wha? 06:28, 24 July 2006 (UTC)

Would someone mind generating some support for your award? Otherwise the proposal will get archived. --evrik 15:39, 22 June 2006 (UTC)

• I have humbly submitted a design for a Stub Sorting Barnstar. Please take a look! Cheers, ♥ Her Pegship♥ 05:27, 27 July 2006 (UTC)

:I think it looks great :) SynergeticMaggot 08:48, 27 July 2006 (UTC)

• A stub sorting award proposal has been submitted by User:Kukini, and designed byUser:Pegship in WIkipedia:Barnstar and award proposals. After reaching consensus, the barnstar was approved and posted on WIkipedia:Wikiproject awards.--Ed 23:07, 27 July 2006 (UTC)

I think a lot of us have had the impression that it was always better to use very "light" stub images due to potential load problems with the image servers. This post from WP:VPT (Village Pump, Technical) says that this is not a problem at all, since the image is rendered only once. Apparently, we can use any type of stub image that we like.

(copied from a post with the same headline)

Q: Are some images more processing-intensive than others? I was under the impression that image resizing and rendering was done only once (say, the first time an image is requested at ##px), so even if complex svg rendering is required, it wouldn't particularly matter. However, a user has expressed concern that some large images should not be used for stub templates because of their high server load. Is there truth to this? ~ Booya ^Bazooka 14:47, 29 July 2006 (UTC)

:A: A large image is rendered into a thumbnail the first time it's used at a specific size, so a large stub image has the same load on the servers as a small stub image (assuming both were thumbnailed at the same size). The load is also much smaller than you might think, due to brower caches and proxies. There was a time (about a year ago IIRC) when a lot of stub images were removed due to server load issues (the image servers were not keeping up with the load); since then, new image servers were added and it's no longer a problem. --cesarb 16:28, 29 July 2006 (UTC)

So at least that is one less concern. :) Valentinian ^(talk) 21:52, 29 July 2006 (UTC)

• At least until the next time the image servers start being overloaded again. It's still a good idea if you have several servicable images of equal utility to choose from to pick ones that will keep the load down. Caerwine Caerwhine 17:25, 30 July 2006 (UTC)

:What surprised me was that it seems that the servers generate a small thumbnail once and then uses this image rather than the larger file. Anyway, if I have to chose between two good images and one is a lot smaller than the other, I normally go with the smaller image. Valentinian ^(talk) 18:06, 30 July 2006 (UTC)

Image size for stub icons has never been about server load (although whether there are images or not was). Icon images are kept small primarily because having large images in a stub template draws too much attention to it and away from the article. That's the reason why a small clear stub icon is always preferable to a large clear one, especially when the image is vertical, since that is more likely to take several line widths - which beomes a problem when there is more than one template. Grutness...wha? 09:10, 31 July 2006 (UTC)

Does it say anywhere that stub templates should not be subst'ed? I've had a user telling me that I should be subst'ing all stub templates. If there is an official policy on this, I think it should be put on the project page. --Bruce1ee 11:31, 10 August 2006 (UTC)

:This is the official policy: NEVER EVER SUBST STUB TEMPLATES. EVER. Martin 11:36, 10 August 2006 (UTC)

:Actually the user (aka me) was saying that he (I) had been told to subst. Happy not to - it takes fewer key clicks. Glad there's a policy. (Where?) --Dweller 12:12, 10 August 2006 (UTC)

::No harm done, really. It was pretty easy to fix. However, we probably need to write this rule somewhere easy to see. Valentinian ^(talk) 12:18, 10 August 2006 (UTC)

:It's written up somewhere on one of the WP Policy pages as to which templates should be substed and which ones shouldn't - I'll see if I can find it. Grutness...wha? 14:18, 10 August 2006 (UTC)

:It says on WP:SUBST, but that is a bit out-of-the-way. Martin 14:19, 10 August 2006 (UTC)

::That was the one I was thinking of. It appears to be guideline rather than policy on WP, but it's regarded as policy at WP:WSS. Subst'ing stub templates causes no end of problems so it is strongly discouraged in a "don't cross the beams"-type way. Perhaps putting it in big friendly letters on WP:STUB would be a good idea. Grutness...wha? 14:25, 10 August 2006 (UTC)

:::... or how about here: WP:WSS#General_rules. --Bruce1ee 14:38, 10 August 2006 (UTC)

::::Done, in both places :) Grutness...wha? 14:53, 10 August 2006 (UTC)

:::::Thanks for that! --Bruce1ee 15:01, 10 August 2006 (UTC)

Alai made a comment about how this WikiProject is no fun, so I was wondering if there could be informal titles or something. Something like a leadership council, but without actual responsibilities. Alai already suggested one title: Stub Approval Group Moderator Director For Life [http://en.wikipedia.org/wiki/Wikipedia:WikiProject_Stub_sorting/Proposals/2006/August#.7B.7BIndian-food-stub.7D.7D]

Just something fun to think about. *shrug* ~ Amalas rawr ^{=^_^=} 20:10, 11 August 2006 (UTC)

:Then, seeing as how I hang out in {{cl|Rail stubs}} and work on them most, I hereby claim "Chief Rail-stub Mogul" for my own title. B-) Slambo (Speak) 20:33, 11 August 2006 (UTC)

• I'd like to be 40px Wearer of the Sorting Hat. ♥ Her Pegship♥ 21:15, 11 August 2006 (UTC)
• Oh, good grief. Grutness...wha? 01:16, 12 August 2006 (UTC) (Geo-stub Coordinator-General and NZ-stubber Pursuivant)
• I seem to be The Countess. At least my talk page suggests it. Either that or She Who Is On The Verge Of Getting Fired For Stub Sorting At Work. Crystallina 22:42, 12 August 2006 (UTC)
• Mind if we call you Bruce to keep it clear? B-) Slambo (Speak) 01:50, 13 August 2006 (UTC)

See {{tl|Maintenance}}, and weep. It's probably not SFDable, either, as it's not just a stub template. (The technical term, to paraphrase Slavoj Žižek, would be "nightmare".) Alai 20:29, 14 August 2006 (UTC)

• It looks like it's only in chemistry type articles (amino acids, chemicals, cell biology, drugs, etc) and it has pretty much only one editor: User:BorisTM. I also found this old discussion but it doesn't have many answers. ~ Amalas rawr ^{=^_^=} 20:56, 14 August 2006 (UTC)
• Doesn't "nightmare" per definition involve the tax man? (Rambo go home, you never met Ole Stavad). Valentinian ^(talk) 21:15, 14 August 2006 (UTC)
• Seriously, TFD ? Valentinian ^(talk) 21:15, 14 August 2006 (UTC)
• Yes, seriously TFD. Grutness...wha? 22:48, 14 August 2006 (UTC)

Category:British royalty stubs says it is for stubs relating to British royalty - however, only actual royals are currently in it. Can I add members of royal households and courtiers, etc? Dev920 12:42, 15 August 2006 (UTC)

Wouldn't it make sense for stub tags to be displayed on yellow backgrounds, just as page tags are? It would seperate them from the article, look smarter on the bottom of each page, and would make them more prominent to the eye. Crimsone 20:19, 15 August 2006 (UTC)

:*I vote no, if that's an option.

:#stubs usually already have plenty of colored boxes - {{tl|cleanup}}, {{tl|wikify}}, {{tl|uncat}}, {{tl|sources}}, etc

:#The colored boxes are distracting, not helpful. People come to wikipedia to read the articles. I think that the helpful editing messages should be as unobtrusive as possible.

:~ Amalas rawr ^{=^_^=} 20:28, 15 August 2006 (UTC)

:::That being the case, would you have an alternative suggestion for the separation of the stub message from the article, so that it is clear that the sub message is not part of the article itself, and is clearly seperate from it in a visual (aesthetic) sense? Crimsone 20:39, 15 August 2006 (UTC)

::::I actually had that discussion on the AWB talk page (link to discussion), and you can just leave 2 blank lines before the tag, and it looks fine. Cortex (archaeology) is a good example. ~ Amalas rawr ^{=^_^=} 20:56, 15 August 2006 (UTC)

:::::A fair comment. While I still feel the yellow box background would look better, I can't disagree that the two additional spaces fulfill the need for differentiation. The two spaces are certainly adequate though. Thanks :) Crimsone 21:58, 15 August 2006 (UTC)

:FWIW, this suggestion has come up frequently in the past, and the consensus has always been as Amalas suggests. Stub templates aren't meant to be noticeable - anyone can see if an article is too short. All they're meant to do is provide a means of categorisuisng the articles and a quick link to any relevant information on how to extend the article. Making them stand out is a bad idea, esopecially in those cases where the article is so short that any boxed/coloured template would dominate the article's actual text. Grutness...wha? 01:55, 16 August 2006 (UTC)

A further thought - I notye that {{tl|Aviation-stub}} uses grey text rather than black to separate it from the main body of the text. (This was never proposed, it seems, but still... ). Perhaps a similar thing could be done with other stubt emplates. Admittedly it's a bit too grey at the moment, but perhaps something like this would work? Grutness...wha? 01:33, 19 August 2006 (UTC)

I would like to propose that articles be placed in categories at the same time they are stub sorted. For example, if an article tagged with {{stub}} is sorted to {{US-novelist-stub}}, which places it in :Category:American novelist stubs, that article would also simultaneously be placed in :Category:American novelists. The former categorization will disappear when the stub tag is removed. But the later category should stay, so its placement cannot be part of the stub template. Please let me know what you think and how to best implement this. Thanks! — Reinyday, 23:48, 17 August 2006 (UTC)

:In theory a good idea, until you look at the details. Often the appropriate permanent category is not the one that is the parent of the stub category, but one of the parent's subcategories. There's no good way to handle permanent categorization than other than by manually placing the categories. Now if we can come up with some ideas on how to encourage people to place the permanent categories, that would be great. Caerwine Caerwhine 00:36, 18 August 2006 (UTC)

::It'd be possible to create a "stub-plus" template designed to be subst'd into the article, which for each type has a) the code for the stub-tag itself (not itself subst'd, of course), and b) some additional permanent categories, where that's possible. But it'd be a good deal of work to set this up for every stub template, and if it were only done for some, it would cause more confusion than it would help things. And it'd depend on people using it correctly, and not (for example) failing to subst it, or getting confused and substing "normal" stub templates instead. A different approach would be to do something similar by bot: if an article has a stub tag, but no perm-cats, add an appropriate one (where available). Or there's the traditional wiki approach: masses of volunteer labour will get these things done sooner or later... Alai 03:00, 21 August 2006 (UTC)

:::There will always be way more categories than stub types, so I don't think it can be done automatically. Piet 12:52, 21 August 2006 (UTC)

::::It would be more of a problem if it were the other way around... The above observation just means that the "automatically" added perm-cat won't be the most specific, and will typically have to be further refined by hand, and obviously there will be categories that don't correspond to a stub type at all, and hence will have to be added by hand. So the marginal benefit over the present system (where it's all done by hand, but with the stub category placing it somewhere it the right general neighbourhood) may indeed be slight, if any. Alai 16:31, 21 August 2006 (UTC)

:I like the spirit of the idea, but I think in practice it would be somewhat redundant. Generally -- perhaps there are exceptions -- "Category:XYZ stubs" is already a subcategory of "Category:XYZ"... thus when a stub is sorted it is already sorted into the parent category. As noted above, of course, this is not usually the most appropriate category, but it is a start. What might be useful to WikiProjects and the like would be a list of "XYZ stubs without real categories." -- Visviva 17:27, 25 August 2006 (UTC)

::I could certainly generate such a thing. One subtlety is how you define "real" category: there are any number of "maintenance" categories and the like used in the article-space. A often-useful technique is to start looking at "Category:XYZs", as you say: those may not be the only appropriate categories, but they should have one of those at least. Often more pressing though is that if an article is tagged with a -bio-stub template it may lack any indicated of occupation or notability, either as a stub type or as a perm-cat (and vice versa). Do you have a "customer" Wikiproject in mind for a "pilot scheme"? Refined categorisation is also useful for driving re-sorting stubs when they get oversized, so it'd be in "our" interests to facilitate and promote this. Alai 18:44, 25 August 2006 (UTC)

I'm not sure this is the right place for this, but at least some people will see it. I saw some stub templates with a parameter, and realized that could be used as a sortkey for the categories. Of course this would require changing all the stub types (or just biological, since those should usually have sortkeys for their other categories). The appropriate section would be changed to (for {{tl|bio-stub}})
Category:People stubs{{#if:{{{1 (for some reason using | didn't work). TimBentley (talk) 02:34, 19 August 2006 (UTC)

:This is the perfect place for this comment! Actually, I've thought in the past that adding a sortkey pipe would be very useful for stubs (especially, as you point out, bio-stubs - biographical not biological, though :). It would require a helluva lot of work, though. Anyone know whether it would cause server issues for such high-use templates, or any other problems? If not, it's definitely worth considering - though, as I said, it'd be a LOT of work. Grutness...wha? 03:02, 19 August 2006 (UTC)

::My main objection is that it would considerably complicate the basic stub template code, making it even less comprehensible to casual users and coders than it already is. We'll need to carefully document why it's there at Wikipedia:Stub#Creating_stub_template and if we can get {{tl|metastub}} to produce the code after being substed, it would be even better. Caerwine Caerwhine 04:01, 19 August 2006 (UTC)

:I think this would largely wasted effort, myself. There's over a hundred thousand bio-stubs just for starters, and ultimately, this is being done just to be thrown away. Better to spend the effort on expanding the article. It might be marginally easier to find the right stub if they were sorted by surname, but better that they be sorted uniformly lexigraphically than a hodge-podge of the two in a single category, which is the most likely outcome. If people really must do this (and it's already unilaterally in place in some cases, then please used a named parameter, and not just "1". Every so often someone gets a bright idea for a stub templates that's parameterised in some way or other, and of course invariably, "their" parameter is "1", too. Alai 02:52, 21 August 2006 (UTC)

I notice that Instantnood has added an extra parameter to {{tl|Stub Category}} to cater for this, which assumes that the sort key is indeed parameter #1. As noted above, I think this is pointless at best, and potentially confusing. In fact, there's been a lot of hither-and-yon edits on this template. As this is used about on about 2,500 categories, it's probably well into "high risk template" territory, and we should probably think about protecting it on that basis. Alai 18:41, 1 September 2006 (UTC)

:Sounds like a very good idea. go for it. Grutness...wha? 23:51, 1 September 2006 (UTC)

I'm wondering if we should systematically employ either {{tl|CatDiffuse}}, or some bespoke equivalent, on "once and future oversized" stub categories. I don't know if people pay any attention to these things, but it might be worth a shot. A more far-reaching version would be to modify the templates to indicate they should be replaced by something more specific, though in such cases we'd have to make sure that in every case there was a suitable sub-type that would be applicable. An interesting existing example: {{tl|football-stub}}. In addition, we should probably be more systematic in out use of {{tl|verylargestub}}. Alai 03:11, 21 August 2006 (UTC)

• -Good idea. A "bespoke equivalent" would be better, pointing discussion of wholesale changes to WP:WSS/P rather than to the category talk page, but other than that, a good idea. Grutness...wha? 05:44, 21 August 2006 (UTC)
• I hereby mass-nominate everything that is now, or has ever been over ten listings pages (i.e., north of 2000 stubs) to get this, for a start. And working down from there... Alai 06:19, 21 August 2006 (UTC)

I'd more usually stick something to this effect on the /To do page, but "actor stubs still oversized and need sorting" isn't exactly in the "news" department. Specifically, I think I've "diffused" these as much as is possible on current data. The good(ish) news is that the actors and US actors are down to five pages, and the UK actors down to seven pages (I think the last may not be as well-categorised as the others). Caveats would be that these have been moved in several directions at once, so there's lots of stubs that probably ought to be double-stubbed according to specific medium and nationality combos that have not, and that the judgements as to "primary" medium are rather crude: I've moved anything with both a film and TV category (but not a stage) to "-screen-", and anything with only one of those two to that stub type; the "stage" types don't look viable on that basis. The bad news is, first as I said I may not be able to do much more, short of a manual slog through each of those parents, dealing with the cases that have either too little categorisation, or "too much". And secondly, that this is going to make types like UK-tv-actor- worse and worse from here... Alai 01:20, 22 August 2006 (UTC)

• I currently go through some actor stubs each day, but hopefully these posts get more people to help out. Actor stubs is too big, but I agree with what Alai said... the more actor stubs is sorted, the bigger the other stubs are (US-movie-actor, US-actor, etc). Maybe people need to work on getting some pages past the stub status? I will try to do what I can, but I currently sort other groups of stubs as well. RobJ1981 01:56, 22 August 2006 (UTC)

In case anyone is feeling at a loose end: I've compiled lists of UK actors allegedly of big screen, small screen, and stage and of none of the above. Obviously tagging the former with all three, or with "-screen-" and "-stage-" rather defeats the purpose of the exercise, while there's no category clue what to do with the latter at all. So per-article inspection and re-sorting required. (Feel free to delete chunks of the pages if you do a "batch".) Alai 06:35, 22 August 2006 (UTC)

I'm not sure if this is the right place to say this, but I find it a little POV to associate a Ying-yang with martial arts. I mean, chinese people weren't obsessed with ying-yangs and korean people didn't even know what they were until the rest of the world did. Assosiating a ying-yang with martial arts is like putting the american flag on a category called "smart people". After all, we have pretty smart people. How long is it until that is not a biased, POV statement? Has anyone realised Taekwondo is more popular than Karate? An image associated with chinese

popular culture does not belong on a category full of martial arts stubs.Daniel_123 | Talk 13:58, 23 August 2006 (UTC)

• I've continued this discussion on the template's talk page. I would suggest that all further discussion be conducted there so as to keep it in one place. ~ Amalas rawr ^{=^_^=} 16:01, 23 August 2006 (UTC)

I'm rather boggled that yet another bot proposal has shown up to create talk-page spam on an industrial scale, announcing that stubs are (coincidentally enough!) "stub-class articles" -- whodathunkit, hrm? And what's more, this one has already been bot-flagged, and has made over 150,000 such edits already. No wonder the db dumps are getting larger and slower, that's probably almost as many new talk pages that have been created, with remarkably little information content. This is admittedly benign to stub-sorting per se, as it's making no automated changes to existing stub templates, but isn't the duplication getting a little ridiculous at this point? If these things are essentially the same, why do we need two parallel structures, and if they're crucially different, why does mass-duplication of current contents make any sense? Alai 05:50, 28 August 2006 (UTC)

:Well, I'm glad that someone else is unimpressed with the assessment tags. Obviously the people behind kingbotk think what they are doing is worthwhile. Perhaps efforts further up the chain will be required. -Acjelen 15:14, 28 August 2006 (UTC)

::Person singular, AFAIK. A manual assessment, adding some actual information content along the lines of "this is a blah-class article", where "blah-class articles" are supposedly likely to make it into WP1.0 (don't ask me what the "blah-class" threshold might be, the 1.0ists are as clear as mud on this, and for all I know are going through this exercise just for the sheer fun/hell of it) might be fair enough. But I can't for the life of me see the point in tagging every single stub in the db, that was never at any point in any number of being selected for 1.0, with talk-page spam explicitly stating this factoid. BTW, the discussion is now at Wikipedia:Bots/Requests for approval/Kingbotk. While listed as "on trial run", I'm inclined to believe that the bot-flagging has been construed as de facto approval. Alai 16:06, 28 August 2006 (UTC)