Wikipedia talk:Manual of Style/Disambiguation pages/Archive 9

Archive 5Archive 7Archive 8Archive 9Archive 10Archive 11Archive 15

I'm slowly becoming unhappy with the way {{wiktionarypar}} works on dab pages. On a normal page with paragraphs of text, the first paragraph flows around the box that's created in the upper-right corner, and that works fine. Dab pages, on the other hand, have lines laid out in a list format, and the first couple of entries tend to get awkward line breaks because of the wictionary box and the result is a rather ugly layout. Does anybody have a good solution for this? --RoySmith 17:25, 9 October 2005 (UTC)

  • It wouldn't be unreasonable to create an alternate Wiktionary link-template for DAB pages exactly for the reasons you've alluded to .. the box works fine in a standard article but works awkwardly on DAB and list-type pages. I don't have any well-considered suggestions on an alternate format at this time. On a side note, I think that the Wiktionary template should be used on far more DAB pages than it currently is used on. Courtland 21:26, 9 October 2005 (UTC)

Flagging article links on a DAB page that are redirects

Consider FAT; in a number of cases like this one I've appended "article link is a redirect" to a dab-def line as I think that this is useful information. However, I haven't done it often and I'd prefer having the practice debated here before I do much more of it. So, what do you think? I've looked through the archives to see if this has been discussed in detail before, and it doesn't look like it has. Courtland 00:19, 10 October 2005 (UTC)

It is an important issue. Personally, I never like to trick a user by using a redirect (or piping for that matter) and I don't think explanatory notes should become practice - there is something wrong if you need to further explain the entry. For this case I would go for
but there is probably a better solution. --Commander Keane 02:21, 10 October 2005 (UTC)

I would have just done:

Keep in mind that many users may not even know what a redirect is. To them, your comment is just confusing gibberish. It's easy, when you're an expert, to remember forget how things were when you were a newbie. --RoySmith 02:33, 10 October 2005 (UTC)

Agree with either Keane or Roy. Another possibility is
In any case, please don't link IATA airport code. —Wahoofive (talk) 02:45, 10 October 2005 (UTC)

I've revised FAT based on your suggestions. Courtland 04:15, 10 October 2005 (UTC)

Proposal to modify MoS section 'The disambig notice'.

There seems to be a movement lately to categorize dab pages (see, for example, Category talk:Miscellaneous disambiguations. I believe this is a bad direction to be heading, and based on the discussions I've seen, there seems to be consensus that dab page cats are indeed a bad idea (although, to be fair, it's clearly not unanimous). I'm thus proposing that we change Manual of Style (disambiguation pages)#The disambig notice to read:


Place the template {{disambig}} at the bottom of the page. It produces the following message (as of June 2005), and also assigns the Category:Disambiguation to the page.

Don't use {{subst:disambig}} as the contents of this notice may change in the future.

There are two variations on the {{disambig}} template to be used in special situations instead of the standard {{disambig}}. TLAs use {{TLAdisambig}}, a slight variation on the {{disambig}} notice. There is also {{disambig-cleanup}} which marks dab pages that need improvement. Dab pages should not be categorized other than as provided by these three templates. If you feel the need to break this rule, please discuss your idea on talk page first).


I don't know exactly what process we're supposed to follow to change the MoS, but I propose we discuss this here for a week (until 26 October 2005). Please mark your comments with support, oppose, etc, so it's clear how you feel. At the end of the week, if consensus has been reached, I'll go ahead and make the change to the MoS page. If anybody would rather see a different process (maybe somebody other than me should do the wrapping up?), I'm fine with that. --RoySmith 12:58, 19 October 2005 (UTC)

  • NeutralWeak support: I think there's an argument to be made for subcategorizing dab pages, but it shouldn't be done higgledy-piggledy. As I proposed on Wikipedia talk:WikiProject Disambiguation, the WikiProject should take the lead in discussing subcategorization, in order to avoid lots of cleanup later. In the meantime, your proposed change is harmless. (We probably should acknowledge {{disambig-cleanup}}) —Wahoofive (talk) 14:57, 19 October 2005 (UTC)
  • Support with modification: I seem to recall seeing discussion of furthre breakout of dab cats along similar lines to TLA (for n-letter acronyms). I also support deference to Wikipedia:WikiProject Disambiguation in such matters. Therefore, I propose that the specific language of the modified text point users to a page to be maintained by that project listing dab cats that have consensus for continued use, while continuing to specify {{disambig}} and {{disambig-cleanup}} explicitly. Perhaps replace the final proposed paragraph with language similar to "By their nature, disambiguation pages tend to defy categorization, and the community consensus is that they generally should not be categorized as anything other than disambiguation pages (using {{disambig}}) or disambiguation pages in need of cleanup (using {{disambig-cleanup}}). The Disambiguation Project presently oversees modifications to this policy as necessary, and it is there that proposed changes should be discussed. Currently accepted disambiguation categories, intended to reduce the size of Category:Disambiguation in a sensible fashion, are listed at Wikipedia:WikiProject Disambiguation/Consensus Dab Cats." The actual location of such a page should, of course, be left up to the project community, which shold also go a long way toward preventing the kind of problem exemplified by Wikipedia:List of disambiguation types, which is worse than useless and should be killed ASAP. --Kgf0 18:50, 19 October 2005 (UTC)
  • Comment on the Disambiguation WikiProject: My feeling is that the WikiProject should be a place for the organization of activity and the coordinated implementation of guidelines. However, I think that this is a reasonable place for discussions leading to guideline changes and I wouldn't suggest deferring to the WikiProject in matters of guidelines. I'm a member of that WikiProject. In general (and the Stub Sorting WikiProject might be a special case, granted), WikiProjects should in my opinion serve in the Judicial and Executive capacities, while Legislative activities should reside in a place like this (using references to the American governmental branches). Courtland 01:50, 20 October 2005 (UTC)
  • I think I agree with Courtland; the wiki project is a way to organize labor and get stuff done, not to make the rules. In any case, I've added a sentence (underlined) to the proposed text urging potential categorizers to talk it up before being bold. For now, I've left it unspecified where the talking-up is supposed to happen. I'm leaning towards replacing talk page with Wikipedia talk:Manual of Style (disambiguation pages), but I'll leave open for the moment the possibility that it might be Wikipedia talk:WikiProject Disambiguation. Does that address your (plural) concerns? --RoySmith 02:23, 20 October 2005 (UTC)
    • Oh, sure. I'm not really fussy about where the discussion takes place. I just would like to avoid having a dozen editors independently deciding to create subcategories and turning it into a huge mess which has to be cleaned up later, à la Stub Sorting. —Wahoofive (talk) 15:14, 20 October 2005 (UTC)
  • Support. I don't see any real benefit to creating a lot of dab-page subcategories. Most of the ones that have been created recently seem superfluous. I agree with the wording proposed above, which doesn't try to prohibit creation of new categories, but encourages discussion first. I would expect it to be a rare case in which a proposal for a new dab-page category is justified. --Russ Blau (talk) 19:22, 24 October 2005 (UTC)
  • Support with revision. What I've come to realize recently is that there exist disambiguation resources in Wikipedia that are relatively unconnected to the activities conducted around Category:Disambiguation. Among these is the List of people by name resource (into which MUCH work has been poured over time). My thinking is that the recent spate of category blooming unwittingly undermines the efforts of topic-specific disambiguation efforts that might come in guises other than dab-pages. Another resource that I only looked at the other day is Category:Signpost articles which could solve some problems with certain dab-pages that don't seem to "fit the mold" (see the debate going on at Talk:Aborigine (disambiguation), for instance).
My suggested revision focuses on the explanatory text:
§ Begin suggested text
Choosing a disambiguation template
  • Do not use {{subst:disambig}} as the contents of this notice may change in the future (if you don't understand this statement, see Information on Transclusion).
  • The template {{disambig}} should be used for the vast majority of cases.
  • If a disambiguation page is composed of three letters only (such as ABC, CDC, or PDQ), please use {{TLAdisambig}}, disregarding whether or not "abbreviation" or "acronym" apply to all of the contents.
  • If a disambiguation page is in need of cleanup in order to bring it closer to the Wikipedia:Manual of Style (disambiguation pages) guidelines, {{disambig-cleanup}} can be used instead of a combination of {{disambig}} and {{cleanup-date}}.
  • If topical categorization of the disambiguation page seems to be needed (for instance, all entries on a page are placenames or all entries on a page are ship names), please bring this need up for discussion at Wikipedia_talk:Disambiguation. Ad hoc and un-discussed category creation is prone to create agitation among the many persons working in the area of disambiguation; be bold, but as that exhortation warns, don't be reckless.
§ End suggested text
That was pretty long-winded. Thanks for considering this suggestion, and thanks to User:RoySmith for reminding me that a comment isn't as good as a vote to a blind bat. Courtland 01:05, 25 October 2005 (UTC)

Thank you everybody for your input. I've updated the text, including many of the excellent suggestions contributed above. --RoySmith 02:04, 27 October 2005 (UTC)

Non-Engish words on disambiguation pages

Often, I run across disambig pages that look something like this:

Foo may be:

  • The Foo people of Fooland;
  • The Foo language spoken by these people;
  • Foo (dish), a food eaten in California;
  • A Hungarian term meaning "used tissue";
  • A Japanese term meaning "my brother's housekeeper's widow".

Now, sometimes these foreign terms are useful, such as when the non-English term has been at least partially adopted into English. For example, the obi disambiguation page points to the Japanese garment of that name. But what about words that are just "Foo is the Crunk-language word for boy" and the like? I know there are instructions for including Wiktionary links in a dab page, but what about situations where the term isn't in English? BrianSmithson 20:06, 20 October 2005 (UTC)

This seems like consensus. I'll add a short note to this effect to the style guide, as I'd like something stronger than a talk page to point to when someone gets made because I deleted a word from his pet native language from a disambig page. --BrianSmithson 11:31, 26 October 2005 (UTC)

Usage of List of people by name

I discovered a while ago List of people by name but hadn't ventured to use it until recently. I've made a proposal by way of action at Abbott where I've removed the names that appeared on this disambiguation page and replaced them with a {{For}} pointing at the appropriate page and section associated with List of people by name. In the process I a) added names from the dab page to the List of people and b) supplemented holes in the dab-page's listing through referencing what I'm taking to be the "master list" of names of people.

My proposal is a large one and I expect it will require long discussion here, at Villiage Pump and elsewhere. I propose that we drop addition of personal names where the surname is the focus of disambiguation from dab pages and instead refer to the List of people by name resource.

This change would have at least two effects:

  1. a positive effect ... only one list of people would need to be maintained for disambiguation purposes
  2. a negative effect ... it would make the list of people inaccessible to the pop-up assistant that has proved so valuable to me recently in revising links in articles that point at disambiguation pages
  3. a negative effect ... it will take another mouse-click for a reader to find a person. --Commander Keane 03:07, 25 October 2005 (UTC) (copied from below)

Personally, I think that positive effect 1 trumps negative effect 2. Likely there are other consequences big and small, and I encourage folks to add to this list to their heart's content.

Thanks for considering this major change ... and how a consensus as to whether it is something that could or should be undertaken could be reached. Regards, Courtland 00:17, 25 October 2005 (UTC)

Another negative effect: it will take another mouse-click for a reader to find a person. --Commander Keane 03:07, 25 October 2005 (UTC)
Perhaps a way around this is to include the list in both places via a template? Josh Parris # 03:42, 25 October 2005 (UTC)
  • Funny you should say this ... I did something similar to that with Bacon (disambiguation) where I used {{:List of people named Bacon}} to transclude List of people named Bacon into the dab page. This was an experiment that I was going to bring up under a different heading, but it works here too. This transclusion has a couple of problems in that a) it introduces the transcluded page's category to the dab page (with the undesirable consequence seen at Category:Lists of people by name) and b) the list needs to be in a particular format where no extraneous material is present such as introductory lines or footnotes or secondary wikilinks or in-line links are present; in other words, it has to be formatted like a dab page but with some slight modifications. Courtland 03:54, 25 October 2005 (UTC)
Well, I'd not try to make it a valid page in it's own right - in the same way templates aren't. In fact, perhaps best to put it in the template namespace, and then include in into both List of people by name and Bacon (dab). Then you don't have those pesky issues with headings and categories, etc. Josh Parris # 05:03, 25 October 2005 (UTC)
  • The problem there is a major expansion of scope in Disambiguation-related activities, basically saying that from date X forward all lists of (insert topic here) fall under a new style format and are no longer (officially) articles in their own right. Otherwise we just increase the redundancy by introducing a template that has overlapping content with a list that has overlapping content with another list. Courtland 12:35, 25 October 2005 (UTC)
  • You're right, but the only list that would be affected is the LoPbN. Not such a big imposition, is it? Josh Parris # 05:51, 26 October 2005 (UTC)
    • Well, that depends on who you talk to :) as usual. The main technical issue would be conversion of the current page system to a transclusion based system in which lists of persons with identical surnames appear as transcluded lists interspersed with names added individually (let me know if that doesn't materialize visually for you). There appears to be a single major caretaker of that system of pages and we've had some recent exchanges about how to contribute to them. I've made a few proposals that might bear following up on ... among these are the production of a Guidelines page to assist in governing them and initiation of a WikiProject to facilitate maintenance and expansion of the resource according to those guidelines. One of the major needs in that page set is the development of a maintenance robot, as well as one or more conversion robots if the route to list of transcluded lists is taken. I'm not really in a position to drive that forward myself right now unless I were to drop other Wikipedia-related activities, which I don't think I'll do at this time. Courtland 02:27, 27 October 2005 (UTC)
    • See my example at Fish (disambiguation)#People. Josh Parris # 04:13, 27 October 2005 (UTC)
    • I've just noticed it stuffs up the working of the pop-up tool, so not such a great plan. Josh Parris # 05:35, 27 October 2005 (UTC)
  • This could be partially addressed by working toward redirection of titles that primarily would require name disambiguation to the list of names, such as Abbott, where the majority of the non-red links were to people. However, this raises another secondary problem ... how to get back to Abbott (disambiguation) from the list of people by name, when the persons who maintain that list are verrry conservative about page size? Courtland 03:47, 25 October 2005 (UTC)
  • Transclusion is a bad idea, and I hope we agree not to use it. It basically means that we will have to maintain two pages, since using ""list of people named Foo" will never satify this MoS. I think the {{For}} notice is a good idea for Callaway, and for pages where people are not the key targets we could move the notice from the top to the "People" section.--Commander Keane 04:49, 27 October 2005 (UTC)
I spoke too soon, using the List of people by name: Caa-Cal is way too difficult to navigate, I thought the {{For}} template was linking to List of people named Callaway, which I thought would be ok. By the way Josh, you can transclude any article (not just templates), take a look at The 2005 Ashes, just above the "Records" section.
[That's User:Commander Keane again, at 04:58, 27 October 2005, per page history].
_ _ There are (at least) two things CK might have meant by
...using the List of people by name: Caa-Cal is way too difficult to navigate
  1. They followed the lk, and landed at the top of the Caa-Cal page.
    There are three reasons for someone to experience that or to label it a problem:
    1. The Form template-calls were miscoded (as i should have realized, when Ceyockey|Courtland said the two For calls could have been combined). I've recoded them (adding a third For, in fact). Now you can go not to the top of page, but straight to a section of that page that contains entries of interest.
    2. If your browser works as my MS IE does, opening a new window for Caa-Cal makes it ignore the section specification in the URL (no idea why). (Lk'g by replacing the window's current contents with Caa-Cal should get you straight to the section.) The workaround for this is to use ctrl-N to clone the window, then click the lk to go to the right section of the LoPbN page.
    3. Some users may consider being at the top of the LoPbN page a problem bcz they have display of ToCs shut off in their prefs. Nearly every LoPbN page large enuf to likely require some kind of scrolling has a ToC, which takes you with a single click to a section likely to be completely displayed without (further) scrolling. And pages whose ToCs grow to likely be taller than a screen get broken up. (It occurs to me that the table of lks to other LoPbN pages has no reason to be at the top of the page, other than making sure users know it exists, against the rare occasions when they find it useful. Maybe it belongs at the bottom of the page, so the ToC is essentially at the top. Hmm, or as the first section below the ToC; i'm going to try that out on the page in question.)
  2. When they got to the appropriate section, the Callaways were buried in the section and it takes a fairly linear search to find them. This is a matter that has grieved me for a long time (whether entry is via a general purpose or one like this, specific to one surname), and i've backed off from trying to solve it with routinely very small sections. But List of people by name: Caa-Cal is now one of a handful of LoPbN pages that i have worked over, experimentally, with each section becoming a multi-level bulleted list.
    For example, once the Calaway-For's lk gets you to the "Cala - Cald" section, the eye need not conduct a linear search of the names. Instead, the first top-level bullet reads "Cala", which is the start of "Calaway". The subordinate bullet just below it reads "Calat", which implies the need to look at the next bullet at the same level. That next bullet reads "Calaw", again the start of "Calaway", and both the entries (i haven't investigated whether LoPbN has been fully populated with WP's Calaways) subordinate to it are Calaway-titled bio-lks.
    The other two variations on the Callaway name, on the same page, are similar, requiring (after the For-lk click and maybe a click on the ToC) looking (no click needed) at 5 points each, with the 5th one being the first entry lk'g to a Callaway-variant bio.
_ _ So it is not clear that there are valid objections to Dab pages using the For tmplt to "incorporate by reference" names that appear only on LoPbN.
--Jerzyt 23:23, 4 November 2005 (UTC)

Let me see if I'm understanding correctly ... the general consensus is that it is in the best interests of users (at present) to continue forward with three list modalities that are neither interlinked nor styled in a complementary manner: dab pages including people, lists of people created by users everyday, LoPbN. I'm not trying to be facetious, but I also don't want to work forward under the illusion that people see a way of integrating (even minimally, such as via x-reference) these three modalities without major efforts and significant cross-activity consensus. Courtland 17:09, 27 October 2005 (UTC)

Templates for Integrating LoPbN & Name Dabs?

_ _ There has been no clear statement of a significant problem that two editors have hinted at:

  1. 'using ""list of people named Foo" will never satify this MoS' (User:Commander Keane 04:49, 27 October 2005 (UTC)
    This seems to refer to the paucity of cases where piping is sanctioned on a Dab page, as is clearly necessary in LoPbn entries.
  2. "... example at Fish (disambiguation)#People" (User:Josh Parris 04:13, 27 October 2005)
    • This demonstrates a Dab following the firmly established LoPbn practice of surname (rather than the equally firmly established surname-Dab practice of given-name first).

_ _ IMO objection CK may have intended is fatal to the superficial template solution of "each surname has a template that both a Dab and an LoPbN page call, in part bcz leading each entry with the same surname is boring, distracting, and annoying. _ _ However, there may be a much less problematic solution either available or achievable by our developers. If it is available with the template specs i am aware of, as far a i can see it would have to be done with meta-templates: LoPbN pages transclude via {{Smith|L}} and Dabs via {{Smith|D}}. Template:Smith would depend on pairs of templates like template:NameFL-L and template:NameFL-D, for which

{{NameFL-L|Joe|Blow|(b. 2005), American infant}} would produce
Blow, Joe, (b. 2005), American infant

and

{{NameFL-D|Joe|Blow|(b. 2005), American infant}} would produce
Joe Blow, (b. 2005), American infant

The doubts i've expressed center on whether a template (e.g. template:Smith) can construct the name of a template it calls (using, i suppose, syntax like {{NameFL-{{{1}}}}|Joe|Blow...), rather than being restricted to "hard coding" template names. There was a point, IIRC, where templates mentioning page-names had to have them hard coded, a restriction that is now lifted at least for non-template pages (as some templates i have used in the following subsection demonstrate); it may be lifted for template page-names as well. If it is not, there are at least proposals for introducing conditionals into the template vocabulary, so the question may become whether the need is urgent enough to put up with klugy solutions if better ones are coming. _ _ Such a scheme may also need to include template:NameLOnly-L, template:NameLOnly-D, template:NameDOnly-L and template:NameDOnly-D, with the L-L and D-D versions passing their one parameter unchanged, and the non-matching ones having null value. This would allow for different judgements in the LoPbN context and the Dab one about, e.g., I. Lewis Scooter Libby and Scooter Libby both need entries. (Offhand, IMO, two entries on LoPbN and one (unpiped one, per MoS) on a Dab.)
--Jerzyt 00:43, 28 October 2005 (UTC)

Close, but as per the names at John Moore, we have to be able to distinguish between identically named people, as well as list people with the same last name. Josh Parris # 08:23, 28 October 2005 (UTC)
_ _ While i see John Moore as flawed in places, i think i take your point: with few exceptions if any, the same names need the same total degree of dab'n whether they appear on LoPbN or a names-only Dab page or a names-et-al dab page; but identical descriptions don't suffice, since a well-wrought Dab probably shouldn't usually repeat the dab'g suffix of the article title, while LoPbN excludes that suffix from its piping.
_ _ I'm tempted to argue for a new feature for this: live links in Wiki comments. (Bear with me a moment! It's not that i've gone completely off the deep end; i'm guessing they have other applications.) A wiki lk, or an external lk, inside comment markup, at present becomes visible only by editing & becomes a link only by
  1. temporarily breaking the open-comment markup,
  2. showing the preview, and
  3. hopefully remembering to un-break the markup before saving the edit. (Well, that's no problem; we all preview before every save. [wink])
Instead, the Wiki parser could (during preview only) continue scanning not only for the end of the comment but also for lks (and maybe other non-global markups), and display at least commented lks (& maybe commented transclusions) in a new space between the preview and the edit pane, or (delimited by a special icon) embedded between or within lines of the preview. Thus instead of having, while editing, just the preview and the dead-text markup, commented-out lks become handy (editor-only) tools. If the John-Moore-persons stretches of {{lopbn-lrng|Mo|o|q}} and John Moore each have commented view and edit lks to the other, one editor can use such lks to draw others' attention to the opportunity/need to maintain them as parallel views of the same underlying information, and those others can conveniently open simultaneous edits on those two pages in two separate windows.
_ _ In fact my notion of two copies of identical or one-letter-different template calls then works better, without putting the collection of single-entry templates into a meta-template: copying individual single-entry templates still saves the ugly effort of reworking the lk markup when propagating new entries between LoPbN & Dabs, but now the copied-over descriptions can be reworked by hand in the cases where the LoPbN one needs a little more detail to make up for hiding the dab'g suffix. This also preserves, e.g., the option of putting the more significant John Moores at the top of the Dab list while LoPbN preserves strict alpha order (supplemented by ordering by probable-date-of-death within groups with completely identical name). I note that i am much more bored by changing the order of the parts of a name & doing the piping than by hand-ordering entries.
(I won't be able to try that out for a couple of days, and i would be the opposite of offended if anyone wants to try out a few templates on a few names on real pages. I trust it's clear that beyond my Joe Blow example, different template pairs are required for names like
  • Burton, Richard Francis, Capt. Sir
  • Spellman, Francis, Cardinal (That's Francis Cardinal Spellman, you know.)
  • others i don't want to think thru right now.)
_ _ I note that i may just be discovering a principle that probably has been well hashed out in the year preceding and the year following Jan 2001: Wiki projects are about abundant eyes, not automation. I think it's a mistake to try and make it work automatically, as by transcluding a block of names into two places, however thoro the automatic transmogrification. (Could this be why the apparently recurring idea that LoPbN should be replaced by a true database of persons shows no sign of implementation?) IMO, it is better to use the technology to break down impediments to more efficient work, than to try to eliminate work that (as we're seeing) requires human judgement. Which is why i end up wanting to lean in the direction of better tools, rather than the other direction: the direction of solving the BIG problem of coordinating related lists with as rigid a mechanism as transclusion of big chunks of text.
--Jerzyt 16:36, 28 October 2005 (UTC)

What is LoPbN & How Does it Work?

Speaking as the main editor attending to LoPbN (above the handful-of-entries level) for the last year or so, it may help if i respond to several of the points made in the #Usage of List of people by name section.

  1. "...what I'm taking to be the "master list" of names of people." (User:Ceyockey/Courtland 00:17, 25 October 2005)
    Pretty close to my understanding, but worth refining.
    • The page itself says in its first sentence (lks omitted)
      This page provides access to a list of people by name, that is, a list
      • of articles (biographies) about real (rather than fictional) people
      • arranged in alphabetical order by name (as discussed below the table).
    That's actually my version, but the focus on bio articles dates from that passage's first incarnation of 20 months ago.
    • De facto, and sometimes explicitly stated, it also serves to note (via rd-lk'd entries) the need for bios as yet unwrit. (This role has been given as one of the reasons that categories do not obsolete LoPbN.)
  2. "...another secondary problem [In quoting, Jerzy has omitted connecting details.] ... the persons who maintain that list are verrry conservative about page size.... (User:Ceyockey/Courtland 03:47, 25 October 2005)
    This could refer to any of three different aspects of LoPbN:
    1. There are at least a few stretches like {{Lopbn-l8|A|s|t|u|v|w|x|y|z}} where lks for each of those 8 name-heads were created 28 months ago, along with the pages, each with a straightforward index to aid in navigating the (then) depth-2 tree. Five of those 8 pages got 10 to 40 names at that time, two got two names each, and the Ax page got its first name only 6 months later. I have renamed pages and given them new scopes, but removed no pages, based on two principles: i don't know how to be sure which may have to be put back as the list grows, and i have plenty to keep me busy on LoPbN without fixing what ain't broke. Although there are limitations implicit in my next two points, there's no reason i can see why {{Lopbn-lrngz|A|u}} or maybe even {{Lopbn-lrngz|A|s}} shouldn't replace their corresponding existing pages, if C/C is right in thinking something is now broke.
    2. I have split a number of pages that were below the 32 kB recommended limit.
      This is in accord with the following rule of thumb:
      The Table of Contents should fit comfortably on a reasonable-sized window, and each section should do the same.
      (BTW, using a LoPbN page of any size with its ToC turned off is, as far as i can see, always a bad decision.) That practice increases clicks, but a few clicks can save the user a much greater effort in scrolling. Maybe someone can quantify this from a human-factors point of view, but lacking that i have followed the principle that binary searches, where they can work, are better than linear searches. And yes, that limits the size of many pages.
    3. I have created a number of micro-pages and even name-free ones (besides the ones that lead down to pages deeper in the tree) like Mb in the {{Lopbn-l2|M|a|b}} {{Lopbn-lrng|M|c|d}} stretch. Mb may be the first i did, and provides an easy example. Mb has only three names; why can't it be folded into Ma or Mc-Md?
      • OK, first, take a new {{Lopbn-lrng|M|a|b}} page (created by overwriting a rdr). Ma is only an index to over 15 pages of names, one or two levels deeper in the tree, including e.g. the Mat and Maru-Marz ones. Putting Mb back with Ma would mean breaking up the index at Ma, in order to permit some kind of format that communicates the downward lks to more Ma... pages are more akin the the Mb names, than to the other intra-tree lks (lateral ones to Me, Mo and whatever, and the obliquely upward lks toward names beginning with other letters of the alphabet. If anyone has a workable scheme other than mine of "a page can have either names or lks to deeper in the tree, but not both", they can to bring forward a specific scheme; otherwise, objecting to the consequences of this scheme is pointless.
      • What about a {{Lopbn-lrng|M|b|d}} page (even tho it was Ma-Mb that i split Mb off from)?
        • Headings with one equal-sign at each end are enormous in some skins; i resign myself to some but prefer to avoid using them, other than one at the top like the Mc one. Merging in Mb would be enough to make me subordinate all hdgs one level. As i write this, the hdgs hav three real levels and a dummy top level entry for all the Mc names; adding the Mb names, it would be 4 real levels, with the single and six-fold levels in reserve. Not a real problem.
        • There are 20 headings on the page; when i finish writing this, i will divide the McCo-McCr section three ways, and McG, McK, and McN-McS each two ways; subdividing the McCo names will create sections the same depth as the existing subsections of McCa, so the depth of the section tree would still be comfortable. The additional Mb section is not a significant change, but the already needed subdividing will raise the number of hdgs to 28 anyway, probably enough in itself to require splitting the Mc names. (... even if having 11 headings at the top level, some of them ranges, weren't enough to confuse the reptilian part of the brain that probably does a lot of the processing of a ToC.)
      • I suppose i shouldn't assume it's obvious that Mb can't share a page with, say Md and Mf. It can't bcz they still have to be in alpha order, and if they are, even casual readers will become editors who say "Hey, there's not even any Mc names either; i know some" and add them between the Mb and Md ones.

--Jerzyt 00:43, 28 October 2005 (UTC)