Wikipedia talk:WikiProject Stub sorting/Naming conventions/Redraft1

First draft log
Improvements, listed in roughly chronological (i.e. not importance) order of their implementation: ''First draft finished. Whew!'' &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 10:04, 20 February 2007 (UTC)
 * Intro now discusses stubs and stubcats (similar edits elsewhere, when needed, e.g. in the summary at one point where only categories were mentioned)
 * Intro now tells people how and were these guidelines are used
 * Cross-reference to list of stub types is now formatted like a cross-reference
 * Fixed triple redundancy in the summary heading
 * Summary converted to bullet list; its points are not logically sequential, so need not be numbered
 * Replaced "guideline(s)" and "rule(s)" with "recommendation(s)" until such time as WSS/NG is in fact a Guideline per WP consensus; "guideline" is presently a misleading claim, and has led to a lot of counterproductive strife in SfD; it was also contradicted by the overall language of the original document anyway
 * Redundant wording fixes ("several of the most important" &rArr; "the most important"; "may later" &rArr; "may"; "entirely in lower-case letters" &rArr; "entirely lower-case"; etc., etc.)
 * Typo fixes
 * Removed repetitive use of "and redirect" after "stub template"; just address redirects once.
 * Replaced computer-geeky "foo" examples with actual ones.
 * Grammar/awkward wording fixes ("can have exceptions", etc.)
 * Logical wording improvements ("necessary" &rArr; "warranted", for example &mdash; nothing here is strictly "necessary" at all; etc)
 * Disambiguated overuse of "category/categories": "intersection of two large categories" &rArr; "intersection of two stub-sorting topics" &mdash; doesn't intrinsically have anything to do with categories per se, nor their size which is a concern for WSS/P, not naming; usage, also changed elsewhere, was also confusing in that "categories" referred to stub-sorting topics, Wikipedia article categories, and stub categories, all now disambiguated)
 * NPOV fix ("are more appropriate" &rArr; "more closely follow these recommendations")
 * Increased wikilinking
 * Reduction of implied WP:OWNership (e.g. removal of "by WP:WSS", as if only WSS can propose stub/stubcat renames)
 * Fixed overspecificity/contradiction, where "xx-stub &#91;or xx-xx-stub&#93;" without "etc." implied no xx-xx-xx-stub exists
 * Wordiness fixes (e.g shortened to "used where (and only where)", instead of "used where" followed by a later "no exceptions" explanatory passage; another example is reduction of "however, if necessary by" to "by"; and so on.)
 * Similar fixing of logorrhoeic constructions, e.g. "examples of these stubs are things like")
 * WP Manual of Style formatting fixes.
 * De-listified a list with only 2 very simple points that thus didn't need to be a list (also per MoS); used emphasis instead
 * Fixed the above to make three points, because on close inspection it was in fact trying to make three, but as two list items
 * Ambiguous "Templates" heading (misinterpretable as "labeling a template as a stub", etc.) changed to "Stub templates"
 * Replaced confusing "xx-stub" and "xx-xx-stub" subheadings, which did not really apply to their content, with a unified "Components" subhead, followed by subsections that better address the components.
 * Replaced confusing run-on "hybrid/complex/composite/compound" with "compound"; as the original said, "these terms are four names for the same thing"
 * Merged split-up discussion of compound stub type; was seemingly confusingly discussed in two different sections.
 * Combined examples where they offered no new insight and were thus redundant
 * Provided additional (real) illustrative examples.
 * Important logic fix, changing "allows for more consistent creation of complex stubs" to "allows creation of shorter compound stubs", because there is nothing at all more "consistent" about "name squishing"; if anything the direct opposite is true, among other problems not addressed here; meanwhile "shorter" provides a rational reason for the practice; the fact that it doesn't improve "consistency" is self-proven by the necessity to have a list in this document of all the weirdly abbreviated ones in first place
 * Reduced unnecessary and long explanation of "parent" and "child" nomenclature, by simply using the terms in situ naturally.
 * Reduced verbiage relating to multi-part topic names, to no longer state the obvious with regard to abbreviations, which are covered elsewhere already.
 * Eliminated unnecessary "Use of" in the usage subheadings, since we already know we are talking about usage
 * Changed "Lower case" heading to "Character case", since section is not about just lower case.
 * Totally overhauled that section, which listed "exceptions" that were not even exceptions, and had a lot of redundancy.
 * Resolved "they" ambiguity in "Abbreviations" section
 * Plurality twiddle, editor/reader &rArr; editors/readers; same section
 * Paragraph flow fix, same section
 * Merged discussion examples of acronyms into same section, where they belong
 * Added emphasis to key point (ambiguity), same section
 * Fixed confusing section nesting levels.
 * Reorganized "Components" section to have a more logical order
 * Actually explained what the terms mean and in what order they go; same section
 * Inverted order of major & subdivisional component sections for logic flow; made "The subdivisional component" plural for obvious reason
 * Moved "Character case" and "Abbreviations" into "The subdivisional components" (while both are briefly discussed under "The major component", they do not need entire sections there &mdash; oh, actually it didnt address this at all in "major"; fixed that)
 * Changed "Order and usual form" to "Order and form", to avoid "begging the question" and inviting exception demands; same with removing "many" from "There are, of course, many exceptions to these rules" &mdash; that's just asking for conflict
 * Noted that the major components list is not complete
 * Overhauled with plain language in "Order and form" to avoid forking the entire piece into two major sections for simple and compound stubs
 * Clarified that stub-sorting topics do not universally correspond to Categories
 * Clarified "The major component", which rather oddly suggested by implication that subdivisional components do not usually represent categories
 * Moved "Of current stub topics, the most ambiguous are probably..." where it belongs, in "major", the only section to which it applies
 * Cleaned up plurals, etc., in "major" list
 * Copyediting in "subdivisional", with Wikilinked examples; generally explained the whole concept more clearly
 * Indicated that the acronym/abbrev. list in "subdivisional" is possibly not complete
 * Created subheading for the list in this section
 * Removed "Non-geographic subdivisions" in the list, moved the one unique example to "major" where it belongs
 * Fixed indent levels in the list, and some formatting cleanup there (and alt. name of Taiwan)
 * Clarified "Character case" with examples; made less wordy; made more specific (only applies to subdivisional components)
 * Added wikilinks (changing examples if necessary) and annotations to "Mix and Match" section
 * Removed the entire "Children of" section; as the document now clarifies, they are not exceptions at all, it's just that "major" was inadequately documented; the "common speech" rationalization for explaining them as "exceptions" was particularly open to "I want an exception too!" interpretation
 * Moved "Exceptions" to end, where it logically belongs because it isn't only about the templates.
 * Simplified subheading of "Foo stubs or Foo-related stubs?"; avoided more use of "foo" - only geeks (like me) know what that means; fixed overgeneralization here as well
 * Updated language that indicated the material in question was very old ("now accepted" as if happened yesterday etc.)
 * Changed "Categories" main section to "Stub Categories", per above disambiguation with regard to "category/categories" overuse
 * Changed "Redirects" to "Stub template redirects" for disambiguation/clarity
 * Added emphasis to "General form"
 * Changed "recommendations" to "conventions" throughout, per discussion below. &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 20:16, 21 February 2007 (UTC)

"Conventions" vs. "guidelines"/"rules" or "recommendations"

 * &#91;The first message in this topic was ported over here from Wikipedia talk:WikiProject Stub sorting.&#93;


 * Well done! A lot of work, but a much needed overhaul - that's the problem with these sorts of pages being built up gradually over time palimpsestuously (if there is such a word). One suggestions - since, as you say, 'recommendations" is more neutral term than "guidelines", but perhaps that term is a little too weak? Since the page is listed as part of Wikipedia's naming conventions, perhaps referring to them as stub naming conventions would be better - recommendations suggests that other names are allowable, yet we're doing what we can to stop that happening since it adds to confusion in the system. These naming styles are the conventionally used names for stub templates and categories, and they have been formed over a long period of time and considerable amounts of discussion, so perhaps it is a better name. The page may also need a new title, whatever is decided on that. Grutness...wha?  23:08, 20 February 2007 (UTC)
 * Thanks! If palimpsestuous isn't a word, it certainly should be!
 * Convention: I think that's perfecto for now (and can be upgraded to Guideline later perhaps. Implies acceptance and usefulness w/o implying a particular policy status.  Re: Rename, maybe so, but wouldn't bother with that now.  Should I make the "convention" edit or do you want to? &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 01:34, 21 February 2007 (UTC)
 * I'm only online for a couple of minutes, then I'm probably not going to be back for the best part of a day (probably not until about 04:00 UTC Thursday) so feel free to make the changes :) Grutness...wha?  07:07, 21 February 2007 (UTC)
 * Done. &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 20:16, 21 February 2007 (UTC)

"Conventions" might be confused with the "naming conventions" (for articles), which of course are guidelines... (Or in a couple of cases, policy, IIRC.)  As these largely concern template names, they're not really strictly speaking within the scope of that set of pages. "Recommendations" does sound like a weakening, I'd have to agree (though, see below). "Guidelines" does admittedly invite confusion with guidelines qua guidelines, which obviously wouldn't have this location. (I do agree that having NC/MoS subpages at wikiproject space is generally a bad idea, not least because they're generally highly underadvertised, and thus might represent a very "local" consensus indeed, completely contradict guidelines with a wider scope, etc (say helllooo 'U.S. highways'). This is less of a concern here, but I'll grant it's not ideal practice.)  So I'm not currently quite sure which is preferable/least worst.

At any rate, my personal objective for these is "as much consistency as possible", especially as regard template names, for which there's obviously very practical reasons. (Category names are in a sense just a matter of stylistic tidiness, plus of course scoping clarity.) As such, I'd not be keen on making them sound more "optional", and if anything, I think there's an argument for "guidelinising" these, with appropriate consulation. (Probably not adding them to WP:STUB, since that's more than long enough as it stands, but perhaps refactoring that into two separate pages.) The point of guidelines is, after all, to document existing practice, and as so far as I know the /NG does that pretty accurately (to the point of in the past of having listed instances that were wildly against-pattern, until we got rid of them (said deletions being my proudest 'contribution' to the current /NG)). It's not as if they're ever implemented directly and unilaterally anyway; rather they're just taken into account (or not) in /P and SFD discussions, so their notional "standing" is in some sense not especially important. I'm not aware of any exceptions that ignore the NGs being granted to "vested contributors", but I'm open to correction on that, and more to the point, so are the offending types... Alai 03:39, 26 February 2007 (UTC)
 * That all sounds plausible to me. If we can stick with "conventions" for now as better than "guidelines" and "recommendations" (for different reasons), any other issues with current draft?  I'd love to get on to the post-cleanup, logic-fixing Phase II, after letting this percolate a little longer (I just thought to "advertise" it at TfD and WSS/P today, so a few more days to wait for further input is probably warranted). &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 04:23, 26 February 2007 (UTC)
 * Reading through the list nothing leapt out at me as outrageous, but equally to be quite candid, nothing leapt out at me as particularly needing to have been done in the first place, either. (Had I been able to see them as diffs, it might have been clearer, but as you c'n'p'd...)  If you're planning larger "logical" changes, what's the point of the "style" changes?  (And in particular, why do those first?)  Alai 06:08, 26 February 2007 (UTC)

Ambiguous regional names
Grutness brought up at SfD that ambiguous subnational region names should use the form NameDigraph-geo-stub (e.g., PunjabPK-geo-stub, PunjabIN-geo-stub). My only question is should this be generalized any further, to other ambiguity cases? Just asking before adding it in to the draft. Was going to add it in as one of the HTML-commented unlikely-to-be-controversial suggestions for improvement to be saved for Phase II re-draft, since Phase I is just a cleaned up but substantively unchanged version of the original. &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 08:22, 26 February 2007 (UTC)
 * Not quite sure what you mean by generalisation here, but if you mean by type of stub, ISTR that at the moment it's limited to geography. I can't think of any cases where it's currently used for other stub types, but it could happen - ah - correction: There are Georgia-stub and GeorgiaUS-stub (no digraph is used for the country because it's not a subnational split), and GeorgiaUS has a politician-stub too. As far as whether it's only regions and not other subnational entities, I can see the possibility of it being used for cities at some future point (e.g., BirminghamUS and BirminghamUK). The only possible quibble there would be whether they would use the national digraphs or (in the case of the US one) the state digraph. Off the top of my head I can't think of two cities within the same country that are likely to both be big enough for their own stub types, so perhaps using only the national digraph is the better choice? Is that the sort of "generalisation" you were referrring to? Grutness...wha?  11:32, 26 February 2007 (UTC)
 * Pretty much. Just wanted to make sure it wasn't missing anything likely to come up next week if we added it in. :-) &mdash;  SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 11:42, 26 February 2007 (UTC)

Terminology simplification
A maybe Phase I, maybe Phase II idea: Let's change the "major" and "subdivisional" components to "basic" and "topical" (or "basic" and "specific"? "basic" and "narrowing"?) Anything but the current terms. "Major" implies significance rather than, um, basicness. "Subdivisional" is just long. Conforming edits might ultimately need to be made to Overcategorization. &mdash; SMcCandlish &#91;talk&#93; &#91;contrib&#93; ツ 08:44, 26 February 2007 (UTC)