Wikipedia talk:WikiProject Stub sorting/Naming guidelines

From Wikipedia, the free encyclopedia

Contents 1 HYPHEN MINUS is the most useless character in Unicode 2 Singular vs plural 3 Redrafting, stage 1 3.1 Given no objections, let's make current re-draft go "live" 4 Redrafting, stage 2 5 "Major" vs. "subdivisional" 6 Countries: adjective or noun? 6.1 Idea #1 6.2 Idea #2 6.3 Idea #3 6.4 Discussion

[edit] HYPHEN MINUS is the most useless character in Unicode

Can someone please explain to my why words in stub template names are separated by the HYPHEN MINUS character? It is an illogical choice from several standpoints:

• The words should be separated by spaces as they are in the English language; im-sure-you-wouldn-t-like-it-if-all-words-were-separated-by-HYPHEN-MINUS-now-would-you? {{some stub}} is eminently better than {{some-stub}}
• HYPHEN MINUS is the scourge of the universe. Neither a hyphen nor a minus. It is a legacy of the US military's ASCII data exchange format and is not part of said English language.
• The HYPHEN MINUS character is almost always much harder to type than a space character (Option–KpdMinus versus the Spacebar in my case).

— Nicholas (reply) @ 11:28, 27 April 2006 (UTC)

The "hyphen minus" is - on almost all computer keyboards - simply a hyphen (lower case, next to the zero, and as easy to type - if not more so - than the space bar). The hyphen became standard for stubs back in the days when spaces caused problems in templates. It is still widely used in templates in general and is so widely used for stubs that it would be one hell of a hassle to change everything. It is also very useful to reduce ambiguity in stub types - a combination of camelcaps and hyphens provides a clear differentiation between the levels of a stub split that using spaces would not. A hyppothetical example: if we had {{Canadian football stub}} would it be for football stubs from Canada, or for stubs related to the distinct sport of Canadian Football? Now comsider Canadian-football-stub (a direct child of football-stub, and CanadianFootball-stub (a direct child of stub). All ambiguity vanishes. Grutness...wha? 01:48, 28 April 2006 (UTC)

I realize the topic is a bit moribund (and I can offer some theories as to why, but that's another topic), but I have to strongly agree with Nicholas on using space instead of dash (ie. changing WP:WSS/NG, from both a usability and language perspective. Have to disagree with Grutness (again!); the proposition that it is easier to reach all the way to the top-right of the keyboard with the hand's least-flexible finger than it is to tap downward with the thumb, requiring no other motion, not only defies basic logic, it's been amply disproven by ergonomic studies. Let's not be silly, please! If you're going to defend a counterintuitive and user-unfriendly naming convention full of odd hyphenation, please do so with real reasons. PS: No opinion on the where-it-is-on-keyboards issue; there are a very wide range of keyboards in use around the world, and many of them are radically different from "typical American" ones; I think it unwise to make assuptions about what other people's computing environments are like... Also no opinion on its usefulness in Unicode; I'm not sure I get the point with that one. — SMcCandlish [talk] [contrib] ツ 22:16, 19 February 2007 (UTC)

I can think of several minor reasons, such as the undesirability of trying to defend the space as a hierarchy-indication, as against... an actual space (the same issue does arise with hyphens, but less often), but mainly, after 3000+ stub templates, I think this has greater inertia than Imperial units, and simply isn't worth the trouble of changing them all over. And the idea of "let a hundred flowers bloom" would be unutterably horrible, even for just a 'transitional period'. I think that's very much a "real reason", contrary to the above dismissive broadside. Alai 02:28, 20 February 2007 (UTC)

The "too far to turn back" reasoning kind of works for me (I suspect however that a bot could do this without too much trouble); still the "investment" level is already high, so fair enough. The hierarchy indicator take doesn't work for me, because as you point out yourself it doesn't actually do what it should; the naming conventions would have to change again, presenting the same "rename a zillion things" problem, so in a twisted way the 2nd argument almost works for me, too). NB: I didn't say there could be no real reason, I asked for one. So, thanks for more-or-less providing one (not so much thanks for the misinterpretation and such, but I'll chalk that up to fallout from our arguments elsewhere.) — SMcCandlish [talk] [contrib] ツ 11:17, 20 February 2007 (UTC)

Re: {{Canadian-football-stubs}} - there are other ways to deal with this amiguity when it arises to such an extent that is has to be addressed. One is simply avoiding constuctions that are ambiguous. If the WP:CUE articlespace were rich enough to need all sorts of new stubs, e.g. for the game of English billiards, I'd follow the permcat nomenclature, which seems to have evolved to be "NATIONALITY players of NATIONALLYAMBIGUOUS game" (I came up with that for English billiards in particular, and then it spread to either American or Canadian football or both, at CfD about a month or so ago.) So, if there were ever a need for a stub and stub cat for Cat:Australian players of English billiards the stub would be {{Australian-Englishbilliards-bio-stub}} and the stubcat would be Cat:Australian players of English billiards stubs by my reckoning (I don't personally care at all for the running-together practice, as in Englishbilliards, but I recognize that is has some buy-in). In the Canada example, I think we'd have {{Canada-footy-stub}} for soccer in Canada and {{CanadianFootball-stub}} for the different game of Canadian football (the latter "Canadian" not "Canada" because the game is "Canadian football" not "Canada football", and the stub isn't necessarily identifying nationality; there are in fact American-born players of Canadian football, mostly pros who didn't make the NFL/AFL cut any longer. If there were so many of them they needed their own bio stub, I guess it would be {{US-CanadianFootball-bio-stub}}.)

[edit] Singular vs plural

In the interest of centralizing the discussion, please bring the discussion of {{sport-stub}} and its relatives here. Her Pegship (tis herself) 15:40, 15 February 2007 (UTC)

Um, not under that topic, please. It is too binary and adversarial. The issue has already been far, far too polarized. I'm presently thinking on a more flexible way to address the topic, maybe/maybe not with specific WSS/NG revision ideas. — SMcCandlish [talk] [contrib] ツ 22:53, 19 February 2007 (UTC)

Went a completely different route. I think that these sorts of angsty issues ought to be dealt with somewhere around the 3rd level of re-drafting of this document; won't even touch them right now. Instead I produced re-draft level 1, detailed below! — SMcCandlish [talk] [contrib] ツ 10:53, 20 February 2007 (UTC)

[edit] Redrafting, stage 1

Its become clear to me after being on this Wikiproject for a while, and trying to actually use it as a general Wikipedian more recently, that WSS/NG is is sore need of a total overhaul:

• It is outdated
• It is confusing
• It is open to misinterpretation
• It is open to differing but equally valid interpretations
• It isn't finished
• It would never survive the WP Guideline Proposal process
• It is directly generative of strife and discord
• among other problems.

Rather than just start editing it with my views on how it should work, and engaging in a load of argumentation, I'm proposing a totally different route. The back-and-forth editing and kvetching-on-the-talk-page process can work when editing a "live" guideline or other document, and even when it works poorly (cf. WP:N from Nov. 2006 - Jan. 2007) it can still produce good results (compare WP:N of Oct. 31, 2006, to the present version! Wow, what a difference.) But it's a slow, antagonistic, painful process, that mingles basic horse-sense edits in with highly controversial changes, and pits gnomes against opinion warriors.

Instead, I think this should be undertaken in clearly defined stages, and be done in draft copies of the real projectpage, which when consensus is reached on the current draft, the "live" page can be replaced with that draft, a new stage discussed and planned, and another draft begun starting with the text of the previous one, to achieve the goals for the next stage, and so on, until a very solid document is arrived at, reflects broad consensus, and is ready to become a real Guideline.

The first stage is obviously Cleanup. It calls for a total moratorium on substantive changes, i.e. changes to the nature, specifics, intent, purpose, scope, etc. of the document and the process it documents. It will be tempting to say "hey, while we're at it, let's fix this", or "I don't agree with this bit at all, and think instead it should require that", or "maybe we should expand this to include..." This temptation must be resisted or this process won't work, and the experiment will fail.

I, like the rest of us (or we wouldn't be in WSS!) get tired of "someone ought to...", "I wish it was like...", "why isn't this better?" kind of thinking. So I've just gone and done it.

The resulting first re-draft is a total overhaul, in almost every way. Yet (unless I made a mistake) has not changed any of the sorts of things under the moratorium (the closest it comes is that it replaced a totally implausible justification for something with a plausible one, without changing the "something"). Every major change has been documented here in great detail, so that the process and reasoning for the changes are independently clear (which is very rarely the case when someone makes a major edit to a "live" document). The goal was to take an old '72 beater and do the bodywork to turn it into a hot rod (on the ouside; haven't touched the engine at all, and it runs exactly as it did before, but will just turn heads now). Aside from inevitable quibbles, I think WSS at large will really like it.

Process: Please read it over, preferably after or while looking over the log of edits, fix up issues you see with it without making any substantive changes to its guts, and at some point hopefully soon, we can have a formal or informal consensus discussion about making it go "live". Discussions of process, or even the merits of the entire "WSS/NC 2.0" idea I'm proposing belong here, I would suggest, while those regarding the particulars of the new draft should go to its talk page.

PS: I would offer that the 2nd stage will be basic policy-functionality improvements that are not controversial (there are HTML comment notes in the source code suggesting some of those improvements); the 3rd, resolution of the controversial bits (such as mentioned on WSS/NC's talk page and which are raised, including by me, at WSS/P and WP:SFD); and the 4th, Guideline Proposal. I believe that getting this document to the level of a Wikipedia Guideline is very important.

— SMcCandlish [talk] [contrib] ツ 10:51, 20 February 2007 (UTC)

A perspective note: I did this redraft while pretending it was a copyediting job from a commercial client, a really good client, whose flagship product I felt disagreeable about in some aspects, and that I was not empowered in any way to change any facts, marketing strategies, intent of spin, slogans, or other pre-determined aspects, only copyedit to present the facts, distill the marketing strategies, wordsmith the spindoctoring, and highlight the slogans. Try it! — SMcCandlish [talk] [contrib] ツ 11:30, 20 February 2007 (UTC)

[edit] Given no objections, let's make current re-draft go "live"

Since there's no disagreement here or at the first re-draft's talk page I plan to replace the content of the "live" WSS/NG with the first re-draft within the next day (and without the "let's fix this" HTML comments in it). I think this will be a good step forward for WSS and it will enable the next round of naming convention improvements to proceed post haste. Huzzah! — SMcCandlish [talk] [contrib] ツ 09:43, 6 March 2007 (UTC)

Working on it now. — SMcCandlish [talk] [contrib] ツ 00:57, 8 March 2007 (UTC)

Done. — SMcCandlish [talk] [contrib] ツ 02:31, 8 March 2007 (UTC)

[edit] Redrafting, stage 2

At Wikipedia talk:WikiProject Stub sorting/Naming guidelines/Redraft2 I have listed a bunch of unlikely-to-be-controversial improvements for the NG document. Most of these were already clearly identified in Redraft1 as HTML comments, while a few come from Redraft 1 discussion. The HTML comments just mentioned are still (as of this writing) present in Redraft2, to indicate likely insertion points. Depending on when you read this, some of them may have alread been replaced with new text, or removed because controversial. I would propose that any item on the list that anyone feels is controversial in any way should be struck out and saved for Redraft Phase 3, the dealing with controversial stuff. Several of them may require a consensus discussion to determine what exactly they should say/advise. Let's do it! — SMcCandlish [talk] [contrib] ツ 04:50, 8 March 2007 (UTC)

[edit] "Major" vs. "subdivisional"

The latest redraft has promoted (at least section-structurally, and seemingly in significance, too) the "major" "subdivisional" distinction. On reflection, I don't think this made much sense in the first place, and hence making it that bit more prominent isn't at all a good idea. What do we actually need to say here, beyond that stub templates have a number of components (there's no shortage these days of instances with three or more), that they're approximately speaking hierarchical (though sometimes permutations of the actual hierarchy, it must be said), and that each component is a noun (phrase) echoing the language of the corresponding category, or some standardised abbreviation thereof. Alai 02:09, 19 March 2007 (UTC)

Hmm. The instructional/explanatory text in the two sections is pretty different. It seems to me that the "major" ones are all or mostly from the top level of the WP category hierarchy or one level deeper, while the "subdivisional" ones are mostly geographical. I get what you are saying, but I'm not sure what to do about what the text says. Probably bears re-reading it a couple of time and sandboxing with alt. text. — SMcCandlish [talk] [contrib] ツ 02:52, 19 March 2007 (UTC)

PS: Did you see proposed idea on Redraft2's talk page about renaming them, since the "major" and "subdivisional" names aren't all that helpful? — SMcCandlish [talk] [contrib] ツ 02:53, 19 March 2007 (UTC)

Hard to suggest how to redraft (or rename) if I don't know what the intent is, and I'm far from clear on that. My suspicion is that there's a large degree of holdover from the Good Old Days when there were a lot of X-Y-stub templates, but not much that was longer, so a "binomial classification" was at least somewhat sustainable. But at present, it'd be a bit odd to suggest that {{US-stub}} had no major components, and one subdivision, and that {{1860s-baseball-pitcher-stub}} had three major portions, and no subdivisions. Or are 'decades' and 'pitchers' subdivisions, too? (I don't think the "major" parts are are high up the category tree as you suggest, nor even any higher than the 'subdivisions'.) Alai 04:11, 19 March 2007 (UTC)

I'm not sure either. As I 'splained in "Phase 1", my redrafting intent was to simply clarify the existing material w/o changing anything substantive. That included the nomenclature. I agree that the descriptions of these things don't really fit reality any more, it just wasn't my intent to address that at the time. If you think it is time now, I don't see any reason to put it off. May require some headscratching as to what precisely to do in place of the existing dichotomy, but so it goes. — SMcCandlish [talk] [contrib] ツ 04:16, 19 March 2007 (UTC)

Those are the hazards of taking on said tasks. :) (I'm not sure to what extent it's a matter of you making the distinction more prominent, or just having drawn attention to it by the very fact of having made changes.) Essentially as I see it, sometimes we have an X-stub and a Y-stub, with child in common X-Y-stub (or equally, at Y-X-stub...); sometimes we have a top-level Y-stub subdivided into X-Y-stubs of various sorts (though occasionally actually at Y-X-stub for 'natural usage' reasons -- see above example, in fact). In the latter case, one can argue that there's a "major" and "subdivision" distinction, but given that it has no significance per se, and no strict implications for naming...

Perhaps we should just first, state the general principle of hyphenated elements corresponding to category hierarchy (except when re-ordered); then, enumerate those elements that are especially common, and/or that have 'standard' abbreviations; state some principles about the forms of some generic components (countries/nationalities as nouns, not adjectives; decades in full, not just last two digits; etc); and lastly, make some observations and give some examples about what order those elements come in, when there's an exception to the 'hierarchy' principle, or where the hierarchy could be read either way. Alai 04:52, 19 March 2007 (UTC)

That's making sense in my reality-tunnel. — SMcCandlish [talk] [contrib] ツ 05:13, 19 March 2007 (UTC)

[edit] Countries: adjective or noun?

Seeing as how this gets discussed at SFD seemingly constantly, I thought I'd bring it up officially here. In creating categories, is the country considered an adjective or a noun? I've created a table with a bunch of different examples to illustrate.

Ex. #	Permcat	Stub cat (noun)	Stub cat (adjective)
1a	Cat:Canada	Cat:Canada stubs	Cat:Canadian stubs
1b	Cat:Canadian law	Cat:Canada law	Cat:Canadian law stubs
1c	Cat:Geography of Canada	Cat:Canada geography stubs	Cat:Canadian geography stubs
1d	Cat:History of Canada	Cat:Canada history stubs	Cat:Canadian history stubs
2a	Cat:Israel	Cat:Israel stubs	Cat:Israeli stubs
2b	Cat:Israeli law	Cat:Israel law stubs	Cat:Israeli law stubs
2c	Cat:Geography of Israel	Cat:Israel geography stubs	Cat:Israeli geography stubs
2d	Cat:History of Israel	Cat:Israel history stubs	Cat:Israeli history stubs
3a	Cat:Burkina Faso	Cat:Burkina Faso stubs	Cat:Burkinabe stubs
3b	Cat:Burkinabé law	Cat:Burkina Faso law stubs	Cat:Burkinabé law stubs
3c	Cat:Geography of Burkina Faso	Cat:Burkina Faso geography stubs	Cat:Burkinabé geography stubs
3d	Cat:History of Burkina Faso	Cat:Burkina Faso history stubs	Cat:Burkinabé history stubs

• Example 1 shows a country with a fairly common adjective form.
• Example 2 shows a country with a less common adjective form. Still guessable, but not as obvious.
• Example 3 shows a country with a very rare adjective form. This is not guessable.

• Examples (a) show the base country cat. Following the permcat is easy in this case. We just add "stubs" to the end.
• Examples (b) show a permcat that uses the adjective form. Following the permcat is easy in this case as well. We just add "stubs" to the end.
• Examples (c) show a geography-related permcat that uses the noun form at the end.
• Examples (d) show a non-geography-related permcat that uses the noun form at the end.

Examples (c) and (d) (at least in Example 1) show that we are interpreting the "<blah> of/in <country noun>" permcat differently.

The question is: Which is more important?

1. Following the permcat and using the noun forms?
2. Being consistent with other stub cats and using the adjective forms?

[edit] Idea #1

• "<country noun> <blah>" permcats → "<country noun> <blah> stubs"
• "<country adjective> <blah>" permcats → "<country adjective> <blah> stubs"
• "<blah> in/of <country noun>" permcats → "<country noun> <blah> stubs"

This would seem to create the most consistency with the permcats. While, it does not create 100% consistency among all stub cats, at least all stub cats of the same type (history, geography, etc) will be internally consistent.

[edit] Idea #2

• "<country noun> <blah>" permcats → "<country noun> <blah> stubs"
• "<country adjective> <blah>" permcats → "<country adjective> <blah> stubs"
• "<blah> in/of <country noun>" permcats → "<country adjective> <blah> stubs"

This strays a bit from consistency with the permcats. However, it creates stub categories that are all similar (save for high-level country stubcats, (Cat:Canada stubs which can remain as they are).

[edit] Idea #3

(Proposed by Caerwine)

• "<country noun> <blah>" permcats → "<country noun> <blah> stubs"
• "<country adjective> <blah>" permcats → "<country adjective> <blah> stubs"
• "<blah> in/of <country noun>" permcats → "<blah> in/of <country noun> stubs"

This creates the most consistency with the permcats. While some people would find the construction Cat:History of Canada stubs awkward because it requires context to know whether it is a category for stub articles about the History of Canada or a category for article about the history of stub articles about Canada, the fact is that the context here and in all similar cases is blindingly obvious.

[edit] Discussion

I marginally prefer idea #2 because it creates consistent stubcats. Permcats often change, so it may be hard to keep up with their naming convention. As long as each stubcat is still linked to the appropriate permcat (using {{Stub Category}}), then we're okay. ~ Amalas rawr ^{=^_^=} 16:41, 28 March 2007 (UTC)

• As you'll know by now, I strongly favour option 1 (in fact, it'/s basically what I suggested as a system to parallel the permcats). Part of the reason I favour it over option 2 is that quite often permcats have deliberately avolided the use of the adjective for a good reason. As such, to use the adjectival form for the stub cats defeats the purpose. Option one uses the adjectival form where such has been deemed the most acceptable usage for permcats, and keeps a noun form where adjectival forms are avoided in the permcats. Sure, it's not 100% grammatical, but no system is likely to ever be 100% grammatical when we're trying to add the word "stubs" onto the end of an existing phrase. It is however logical, consistent, and easy to keep track of. What's more, although it doesn't completely remove the weird and often unguessable adjectival demonyms, it does remove them in those cases where they can't be instantly checked by referring to the permcat. I for one would prefer not to have to look up some of the weirder adjectival forms around (Kittitian-Nevisian? Burkinabé? Equatoguinean? I-Kiribati?) whenever I need to find out what to call a stub cat. Grutness...wha? 00:51, 29 March 2007 (UTC)
  • I see there's now a third choice - I'd favour that over option no. 2 as well, though I realise there have been objections to it on grammatical grounds in the past. Still, as I said above, nothing we do is likely to be 100% grammatical. Grutness...wha? 23:26, 29 March 2007 (UTC)
• I strongly favor Idea #3 (proposed above) but marginally prefer Idea #1 over Idea #2 for the reasons Grutness gives. Caerwine Caer’s whines 04:26, 29 March 2007 (UTC)

That didn't entirely parse. Grutness prefers #1 over #2; Amalas preferred #2. — SMcCandlish [talk] [contrib] ツ 00:05, 31 March 2007 (UTC)

Fixed so that it parses. Caerwine Caer’s whines 06:39, 31 March 2007 (UTC)

• I'd go with #3, too; just seems much easier to remember and is less format futzing for format futzing's sake. — SMcCandlish [talk] [contrib] ツ 00:08, 31 March 2007 (UTC)