Wiktionary:Beer parlour/case-sensitivity vote

This page is no longer active. It is being kept for historical interest. No discussion is needed to revive this page; simply remove the {{inactive}} tag and bring it up to date.

This is a Wiktionary policy, guideline or common practices page. Specifically it is a policy think tank, working to develop a formal policy. Policies – Entries: CFI - EL - NORM - NPOV - QUOTE - REDIR - DELETE. Languages: LT - AXX. Others: BLOCK - BOTS - VOTES.

Currently, this project has first-letter case-sensitivity turned off (Update: this was turned on in 2005). Consequently, it treats "kind" and "Kind" as the same article title. This question began on Meta.

Should the English Wiktionary switch to first-letter case sensitivity?

(i.e. should "kind" and "Kind" be considered separate pages by the software?)

Secondary questions:

Should we run a script renaming many many pages?

Should people entering "tom" be taken to a blank page even if a page "Tom" exists? — Hippietrail 17:06, 23 Jun 2004 (UTC)

Arguments for

• It makes for better spelling, which is nice to have in a dictionary: Capitalization of the first word should be optional for words whose proper title/sentence case doesn't use it, as in Bantu languages, or words like mL and pH. —Muke Tever 16:43, 23 Jun 2004 (UTC)
  • But it could also be done by a new directive similar to NOTOC and REDIRECT, e.g. CASEFORM pH. A fix for initial caps implemented along these lines would also be useful for other wikis where turning off initial caps is out of the question (see for example en:Wikipedia's List of pages with capital letters that should be lowercase). —Muke Tever 02:56, 28 Jun 2004 (UTC)
    • Please expand your idea. I don't understand what you're suggesting, so I will reserve opinion for now. Eclecticology 07:23, 28 Jun 2004 (UTC)
      • Well, currently there are a couple of in-page commands like __NOTOC__ ("suppress automatic TOC creation") and #REDIRECT [[page]] ("redirect to page instead of displaying this page"). I don't know how non-trivial it would be to implement, but I am suggesting a similar command to alter the display name of the article. A valid (input) argument for the command would be a case variant of the title, possibly replacing keywords with characters not allowed in page titles, so e.g. w:C plus plus could display in the headline as "C++", and w:PH as "pH"; but you wouldn't allow w:Cat to display "Our feline Overlords". (Really wishful: also allow WikiHiero titles.)
      • Pros: useful on all wikis; solves Bantu and abbreviation problems, as well as pages currently uncreatable (like "C++").
      • Cons: doesn't solve the kind/Kind problem (though I still don't consider that a problem); slightly counterintuitive (the title "C++" is no longer parallel to its URL)
      • Inherent flaws: is not currently implemented (a developer would have to code it, unlike decapitalization which already exists); may be non-trivial (or impossible?); doesn't help when an anomalous capitalization has to share with an ordinary one, e.g. symbols like mL / ML — Wikipedia can use disambiguation, but disambigs aren't much use on the wiktionary.—Muke Tever 14:12, 28 Jun 2004 (UTC)

how about creating a way for editors to flag/unflag specific articles as "case-sensitive", and a way to automatically set this flag for all new articles? .sj:. 21:34, 3 Sep 2004 (UTC)

It allows the other wiktionaries to have case sensitive spelling. This vote holds back progress on the other wiktionaries. GerardM 17:43, 24 Aug 2004 (UTC)

Arguments against

It makes an unnecessary distinction, and will only cause confusion and problems.

• • This is not very clear, someone should clarify this. — Timwi 14:33, 23 Jun 2004 (UTC)
  • Exactly the opposite is true. Eclecticology 07:07, 24 Jun 2004 (UTC)
  • Every other letter in an article name is case sensitive so there would be little more confusion than already exists. Why leave the first letter arbitrarily case insensitive? 64.253.110.0 12:17, 13 Sep 2004 (UTC)

• The user proposing this has done so because he is also planning on moving all extant pages to lower-case. [1] This is a major format issue that mysteriously is not being put to vote, and should be resolved before this step is taken. —Muke Tever 16:43, 23 Jun 2004 (UTC)
  • I just reviewed the link, and your statement is a complete distrortion. Eclecticology 07:07, 24 Jun 2004 (UTC)
    • Sorry, the page says that he would "run a script that renames every page to be lower-case, except for those pages that contain their own title in capitalised spelling" — I expected that it meant what it said. —Muke Tever 14:34, 24 Jun 2004 (UTC)
      • Your latest response notes the important "exception", while your first comment did not. The voting proposal was in three parts, and if we accept voting a a valid way of dealing with these issues it really should have been put as three separate votes. I regard allowing lower case first letters to be the principal issue. What kind of script is used to adjust for that change is a secondary issue to me. Timwi's proposal on this sounds OK to me, but I have no problem dropping it if the main point can be accopmplished. Eclecticology 18:48, 24 Jun 2004 (UTC)
        • Sorry again, my first comment wasn't intended to besmirch the accuracy of the script—even were the script magically able to separate all capitalized and uncapitalized entries without supervision or cleanup afterwards, my purpose is to question whether it should be run at all. —Muke Tever 14:05, 25 Jun 2004 (UTC)
      • I also consider the operation of the search function to be secondary. Although I strongly believe that the first letter of a title should be case sensitive, I also believe that search entries should be case in-sensitive. Eclecticology 18:56, 24 Jun 2004 (UTC)

What is the reason for having case-sensitivity? The onus is on supporting its implementation, as it adds complexity and distinction that will be problematic and create a lot of work. If this proposal is just regarding the capitalization of the first letter, why is the problem not solved by simply removing case-sensitivity altogether? If appropriate, the capitalization of a title can be changed, but the spelling could still be unique and there would not be multiple articles with the same spelling with just different capitalization.

The only normal words where capitalization matters are proper nouns, which are listed in their appropriate section for the page of that spelling. Unique capitalization can be indicated by the current practice as noted in the template: that is, posting the word in its proper spelling and capitalization under the "Proper noun" section as is needed. As for abnormal definitions, like abbreviations, it also only serves to add complexity without serving any usefulness. Abbreviations are presented in the page of normal spelling and point to the appropriate definition that the abbreviation represents. That is, they are not in separate disambiguation pages if they have the same spelling as a normal word. Having case-sensitive pages is not particularly useful with abbreviations because they are almost exclusively references to other terms. They easily fit in one page, and different capitalizations can be indicated in that one page. More importantly, many abbreviations are appropriate whatever their capitalization. Are we to have a page on "A.M.", and then redirects to that page from all of "a.m.", "am", and "AM"? - Centrx 21:05, 23 Jun 2004 (UTC) - revised wording and removed point about searching 22:15, 28 Jun 2004 (UTC)

That prose is tortuous. All German nouns are always capitalized. Adding the word at the beginning of each section was just one way of trying to work around the problem of the failure of the headings to be case sensitive. If needed we can put "see also" notes to your various representations of "am". Case sensitivity in searches is a completely different issue from case sensitivity in titles. Searches should show separate listings for each capitalization variant. This happened in Wikipedia because it was understood that despite the policy for using text style headings some people would continue to use headline style. Eclecticology 07:07, 24 Jun 2004 (UTC)

If we don't make different articles for words that are spelled the same yet mean totally different things, then why should we make different articles with capitalization? If having separate pages for different capitalization is important, then having separate pages for different meanings that have the same spelling is appropriate as well. Also, if the capitalization is important for the German Wiktionary, why not set case-sensitivity on for the German Wiktionary and others that need it, and off for the English Wiktionary and others that don't? - Centrx 22:22, 28 Jun 2004 (UTC)

Nobody except you is suggesting having separate pages for different meanings of the same word. We do have German words on this Wiktionary. Eclecticology 07:46, 29 Jun 2004 (UTC)

I doubt its true I'm the only one who thinks that on the Wiktionary. Nevertheless, it is a suggestion that stems from several other paper and online dictionaries. Their example indicates its appropriateness. Could you point me to a credible English dictionary (for instance, not Wordnet, which I have seen many times to have false and inaccurate definitions and has the misleading and false practice of putting a bunch of words into a synonym group and giving them all the exact same meaning and usage examples) that puts things in the same definition based on spelling rather than meaning? This is not how it's done in the OED (checked 1989 and online), Webster (several editions and years), American Heritage (the ones on dictionary.com), Cambridge (on dictionary.cambridge.org), and Encarta (from encarta.msn.com). I don't think the Wiktionary should do things this way, but the reasons I have been provided for it apply well to keeping different capitalization on the same page. In the interest of constistency, either case-sensitivity-uniqueness with different pages for different meanings should be the policy, or neither.

Having everything on the same page is deeply flawed when combined with the format of the current template, which indicates that the top-level heading under each language should just be the word "Etymology" repeated over and over for each substantially different meaning. Anyway, the capitalization can be integrated into a single page, with headings like ==Noun: Kind== and ==Adjective: kind==. That's no different than the current practice of having different meanings on the same page distinguished by headings and it saves us from the ridiculous situation, similar to that I mention below, of having similar meanings on different pages yet meanings of greater difference on the same page. - Centrx 23:16, 29 Jun 2004 (UTC)

There is no need to be repeating the same etymology for different meanings. A new etymology is only needed when we are dealing with a word that is written the same but has a different origin as in calf. You frequently seem to argue against yourself. You say that having everything on one page is "deeply flawed" and having them on different pages is "ridiculous". It would not surprise me to find dictionaries on the internet with "false or inaccurate" definitions. Although I have probably referred to most of the works that you cite, I still find the paper dictionaries and other references to be more reliable. I use my paper New Oxford as a first reference when I have uncertainties, but I also have a growing collection of these books from various ages. The 1913 Webster is a good source for beginning articles, as long as one remembers that it is 91 years old, and that some of its entries need serious upgrading. Eclecticology 04:02, 30 Jun 2004 (UTC)

First of all, I am being consistent and especially precise. Re: etymology: I am not saying that you repeat the same etymology for different meanings. I am saying that the current template indicates that, when there are different etymologies (what I have been sometimes referring to as different "substantial meanings", as there are two levels of difference in meaning we are discussing at the moment) in the same spelling (page), those different etymologies should be in different sections (top-level below language) and each of those sections should be named "==Etymology==" (by which I do not mean fill in the blank with the particular etymology). As in calf, they might also be listed like "==Etymology 1==" and "==Etymology 2==", which are absolutely meaningless headings. One cannot even navigate to the definition they are looking for based on those headings. Such is a poor way of distinction.

Re: "ridiculous" and "deeply flawed": I am not saying that having different meanings on different pages is "ridiculous" and I am not quite saying that having everything on the same page is "deeply flawed". What I am saying is ridulous is: having meanings of a certain difference on different pages while at the same time having meanings of a greater difference on the same page. An illustration more by procedure: Say there are two meanings with spellings "blarg" and "Blarg" with a difference of meaning of value 10 (for simplicity, I will use numbers, differences in meaning are theoretically quantifiable but in practice of course qualitative. Nevertheless, the principle is valid: one can easily envision examples where there is tremendous difference in two meanings compared with two meanings where there is a miniscule difference). With case-unique implementation, those two words would go on different pages. Yet, we will find two other meanings with the same spelling "erkle" of a difference in meaning of value 50 and those two meanings will go on the same page. That is a ridiculous situation, not something so bland as "having things on different pages". As for deeply flawed, and this is not a direct response to your statement because your misunderstanding was about what was ridiculous, I was saying that the combination of having different meanings on the same page and using the current template is deeply flawed. Simply having different meanings on the same page is not as flawed.

Re: dictionaries: huh? Your statement regarding the dictionaries has nothing to do with what I was asking and relates only to my parenthetical statement that was nearly irrelevant to the matter at hand. All the dictionaries to which I was referring, with the possible exception of Encarta, as putting different substantial meanings (with different etymologies) in different definitions, are print dictionaries, the online versions of which are faithful transcriptions. The 1989 OED is a renowned print dictionary and its online version is a bona fide transcription thereof that is maintained as an official Oxford English Dictionary. It contains an updated draft version that will become the forthcoming edition, as well as an untouched 1989 edition. The other dictionaries I mentioned that were online are all directly from print dictionaries and are organized based on those dictionaries. In addition to the 1989 OED which I have examined at my local library, the following are print dictionaries which I have personally verified in their original bindings on paper and do what I described above and will restate again below: Webster's New World College Dictionary, Fourth Edition (1999), Webster's New International Dictionary (1930), and Merriam-Webster's Collegiate Dictionary (I forget the date as it is in another physical location, but I examined it today). With the online dictionaries that are faithful transcriptions of print dictionaries, it is quite clear that professional lexicographer's organize dictionaries in the same way in this respect. What is wrong with this way for the Wiktionary, and why would the same reason for it being wrong not also invalidate case-uniqueness? I will restate: Please find a credible dictionary that puts different substantial meanings (with different etymologies) in the same definition. Framed differently, please find a credible dictionary that does not have different definitions for different meanings, regardless of spelling. - Centrx 06:09, 30 Jun 2004 (UTC)

As far as I understand it, the switch will be flipped for all Wiktionaries or none at all. Since all Wiktionaries are describing all words from all languages, it wouldn't make sense to do otherwise. Capitalisation makes sense for any dictionary. Months and days are capitalized in English, but they aren't in French or Dutch. Adjectives for countries and languages are capitalized for English and Dutch, but they aren't for French or German. What goes on which page is an arbitrary decision. We have one set of rules for it now, we will have another set of rules when the switch is flipped. Or should I say if it is flipped? If this question had come up 6 to 8 months earlier, it would have been a no brainer and everybody would have agreed that it was a good decision to have case sensitivity. I guess that's why Timwi thought it would be a trivial question. Polyglot 06:56, 29 Jun 2004 (UTC)

What's important is the flexibility of being able to have two pages wherever the situation warrants. Limiting the first letter to a forced capital imposes an unnecesary restriction that makes things a lot more complicated. As a user I would want the title of the article about a word to appear just as the word would in normal usage. The current work-around of repeating the word with proper capitalization after the part of speech may be lost on those who only visit to look up a word, but do not know our convention. Eclecticology 07:46, 29 Jun 2004 (UTC)

The same can be said of putting different meanings on different pages even though they are spelled the same. It is important to have that flexibility (note that limiting the first word to a capital is a different issue from case-sensitivity-uniqueness. The necessity of the first letter being capitalized should be removed and there should be a separate vote for that). If there's a difference in meaning due to capitalization that would warrant a different page, then the capitalization can easily be integrated more prominently in current pages, as I mention above like ==Noun: Kind==. If different pages for a certain lettering are warranted persuant the current proposal, then there will be multiple entries with different capitalization and anyone who is going to notice capitalization in the title will notice this. Or, there can be a prefix like "Spelling: Kind" or "Capitalization: Kind", so that reader does not have to know our convention. - Centrx 23:16, 29 Jun 2004 (UTC)

Yes, so those words with capitalization can be put on pages that have appropriate capitalized titles. I'm not arguing against capitalizing titles, I'm arguing against having case-sensitivity where two pages with different capitalization, but the same letters, are unique pages. If you're saying that different capitalizations for the same meaning and letters should go on different pages, that would result in the ridiculous situation of having words with the same meaning but different capitalization being on different pages, yet words (spellings) with different meanings and the same capitalization going on the same page. - Centrx 23:16, 29 Jun 2004 (UTC)

No, I'm suggesting that separate pages would be created when a different capitalization implies a different meaning. There are of course many situations where a word with the same spelling and capitalization can have more than one unrelated meanings. Perhaps the day may come when that entry needs to be split, but that is a different problem. You seem to anticipate a lot of strange situations that assume people will go out of their way to create weirdness. Some are inevitable, but these are few enough that they are easily subject to corrective editing. In reality people tend to follow examples. What I am arguing for is to have the technical option to have any letter in a title capitalized, where two different capitalizations could indeed give different pages. If someone sets up an article for kiNd you can rest assured that every editor will look to see if it is legitimate. The search function should continue to be case insensitive as it is now, and it should produce a list of articles for all existing case possibilities. Again that gives you another opportunity to correct any inappropriate weirdness. Eclecticology 04:02, 30 Jun 2004 (UTC)

The comment above yours was in response to Polyglot's comment, as indicated by the indentation but misleading by the order. I don't know exactly what you are referring to, but I am anticipating peculiar things because this is a dictionary and hopefully will contain hundreds of thousands of words. "Strange" things may not be common but they exist throughout languages and they will be encountered numerous times. They are not subject to simply corrective editing if it is impossible to conform a definition to the standard format. In such cases we will either end up with the needless work of reformatting the entire dictionary or we will end up with numerous definitions that do not conform to the format of the dictionary, which is not the way of a formal work, least of all a dictionary. We must create a sound format that will hold the many variations in lexicography rather than waiting until we encounter problems that require extensive reformatting. To clarify, my argument is that it is simply not consistent or appropriate, formally or semantically, to have case-unique pages while at the same time putting substantially different meanings--which are separated in other dictionaries--in the same page. - Centrx 06:09, 30 Jun 2004 (UTC)

(On the other hand, as a user I would want to see titles capitalized properly, as poor typography weakens credibility.) The efficacy of the "workaround" can be measured: set a new user in front of a page, ask him to find information, and time him, checking also where he looks. Without testing we can only guess: does a user expect capitalization to be prescribed in a heading or title? Does a user notice the capitalization or lack of same in the "workaround" to be semantically meaningful? Is the outcome any different when the definition is divorced from its heading by a screenful of TOC display and definitions of the word in all languages from Aari to Enggano? —Muke Tever 14:24, 29 Jun 2004 (UTC)

With my above suggestion like ==Noun: Kind==, the capitalization is displayed in the TOC in the same heading that substantially different meanings are displayed. As for credibility and typography, having definitions grouped with other definitions solely due to their spelling (all meanings of the same spelling are in the same page) and capitalization (as by the current proposal), places typography above meaning in importance. Orthography above semantics likewise damages credibility. Or, in the case of needing to make the changes to all the Wiktionaries (languages) rather than only those that think it is appropriate, it's technical inability above lexicography. - Centrx 23:16, 29 Jun 2004 (UTC)

One of Muke's complaints above is that he did not like a llong TOC before getting to the article itself. That's fair enough. You can edit your user preferences so that the TOC doesn't appear. But if a person does that the format ==Noun: Kind== won't appear in the TOC. Eclecticology 04:02, 30 Jun 2004 (UTC)

My suggestion does not lengthen the TOC, and this argument is just as valid for having different pages for different substantial meanings. The same person who sets it so they don't see the TOC will not see the TOC for the various different meanings, but the same person will see the headings for the different meanings if they scroll through the page, just as they will see the capitalization information. For the same people who don't like wading through stuff to get to the article, why would they want to see a disambiguation/search page when they type in a word that has different capitalizations, or if it goes directly based on what you put in the search, why do they want to find themselves at the wrong page because they used the wrong case, and if users are expected to know the right case when they are querying, why do you expect that they need to be informed of it at the topmost level of page title? At the same time, if this is what's going to happen when case-uniqueness is implemented, why is it not appropriate to have the same disambiguation/search page display the words of various meanings (with the same spelling) that point to different pages. Indeed, this speeds up the finding of information, because one can find the definition they are looking for without wading through the unrelated rest of a page (and headings like "Etymology 1" are, again, not useful here). - Centrx 06:09, 30 Jun 2004 (UTC)

If I look up kind, I expect to find both the English adjective (no capital) and the German noun (capital) on the same page. I would favour no case-sensitivity at all for ease of searching and use. --217.43.99.99 17:10, 14 Sep 2004 (UTC)

Arguments that aren't really arguments

• Muke Tever wrote, having separately-cased forms of words on different pages might overemphasize the difference between some senses of a word, e.g. Cynosure vs. cynosure. — This is not an argument because in such cases you can still have the two things on the same page if you want. It doesn't follow that there should be a technical restriction that forces "kind" and "Kind" to be on the same page as well.
  • Okay, so for a parallel example in the language you're proposing this for, you would put rot "red" (adj.) and Rot "red" (neut. noun) on separate pages? —Muke
    • Straw man. There are times when having them on the same page will be preferred, and other times when it will be better to have them separate. It's good to have flexibility about such things. Eclecticology 07:47, 24 Jun 2004 (UTC)

      And there will be many times when people put them on the wrong page, or people needlessly move them, or people argue about which they should be one, or different people create them on each page. This is arguable good flexibility for good editors, but it's confusing or messy for newcomers. It will mean more work for editors. — Hippietrail 12:53, 24 Jun 2004 (UTC)

      • People will continue to place things at places based on reasons known only to them. this is standard for whatever system you choose. Eclecticology 19:53, 24 Jun 2004 (UTC)

      I wasn't setting up a straw man, I was trying to propose a more relevant example. —Muke Tever 14:34, 24 Jun 2004 (UTC)

• It will make harder to find a word, once you are on the article you should find this case-sensitive information, not before, even punctiations signs like ´¨^`~ should be ommited in searches(only in searches) just like Google does, but Article titles should be case-sensitive because they are the 'term' it'self, to reach the term 'new york' you should be able to reached like that and see the article title "New York", a dictionary is often used for corrections, don't expect users who already know that a city name should be written capitalized, to look for a city name.(i'm Levhita in en.wikipedia.org)
  • This is a concern that relates to the search functionality of Wiktionary, and not the technical restriction that disallows article titles like "kind" or "isiZulu", and hence it is irrelevant here. — Timwi 14:33, 23 Jun 2004 (UTC)

    It is relevant because searching for "bob" or "Bob" are currently equal. The proposed change will mean that people will have to change the way they search and will mislead many people into believing artlices don't exist. — Hippietrail 17:06, 23 Jun 2004 (UTC)

    Not at all. In search either "bob" or "Bob" should list both articles unless you designate that your search should be case sensitive. Eclecticology 07:47, 24 Jun 2004 (UTC)

    I can't test on Wiktionary because search is turned off, but on Wikipedia I found actually the opposite problem to what I expected. If the German article "Kind" exists but not the English article "kind", and somebody searches for "kind" then all versions will be found (kind, Kind, KIND, etc). If, however if she clicks "Go", she will be taken straight to the "Kind" article. She now has a chance to add English articles in the wrong place that she didn't have before. More work for editors. — Hippietrail 12:47, 24 Jun 2004 (UTC)

    "Go" should default to search when there is an ambiguity, or there should be an exact case option in the search function. I'm ecstatic to have the search function back, but it could still stand to be more sophisticated. Eclecticology 19:53, 24 Jun 2004 (UTC)

• You don't seem to understand what case sensitivity is. Wiktionary already has case sensitivity, what you really want is to turn off first-letter capitalisation. -- Tim Starling 03:01, 23 Jun 2004 (UTC)
  • I agree, the issues of case-sensitivity in titles, and initial capitals in titles are being confused here, so this vote as it stands is ambiguous. Muke
  • I perfectly well understood, and I gave the kind/Kind example to demonstrate it. — Timwi 14:33, 23 Jun 2004 (UTC)
    • No, the kind/Kind example is to demonstrate that you plan to make all existing entries lower-case, excepting German and proper nouns, which is a separate issue not related to this deunification of capitalized and noncapitalized page titles. —Muke Tever 16:43, 23 Jun 2004 (UTC)
      • That's not what was said. The default should be to have all titles start with a lower case letter unless there is some reason to have it capitalized. It merely provides the opportunity for "deunification". Eclecticology 07:47, 24 Jun 2004 (UTC)

• Capitalization of the first word should not be optional so that you can have every word put in lower-case based on disingenuous comparison to dictionaries, whose formatting of entries is much different from Wiktionary's — unlike dictionary entries, Wiktionary articles are much more comprehensive (the part of a Wiktionary article that a dictionary entry corresponds to is a very small section, where, incidentally, standard Wiktionary practice is to indicate the proper in-text capitalization). —Muke Tever 05:51, 23 Jun 2004 (UTC)
  • This is not an argument to keep the technological restriction, and nor is it even an argument to have title-cased titles. Why should we specifically be different from other dictionaries? And what significance does the length or comprehensiveness of Wiktionary articles have here? — Timwi 14:33, 23 Jun 2004 (UTC)

    "Being different from other dictionaries" is not the only thing which will change if this proposition proceeds. — Hippietrail 17:06, 23 Jun 2004 (UTC)

    • The length and comprehensiveness of Wiktionary articles is what makes them articles, and not just paragraphs on a page (as are entries in print monolingual dictionaries) or wordlists (as are entries in print bilingual dictionaries). —Muke Tever 16:43, 23 Jun 2004 (UTC)
    • Wiktionary articles are articles, and their titles are titles. Titles are put in title case. This doesn't make sense? —Muke Tever 16:43, 23 Jun 2004 (UTC)
      • The comprehensiveness of Wiktionary articles is not relevant. The last syllogism is a non-sequitur What do you mean by "title case"? Eclecticology 07:47, 24 Jun 2004 (UTC)
        • w:Sentence case describes it. I overstate my case somewhat, as it is sentence case that is the de jure standard across our wikis, not strictly title case — even those wikis [2][3] that incorporate this change only do so for languages that do not use capitals for title case. —Muke Tever 14:34, 24 Jun 2004 (UTC)
          • That's what I suspect, but I wanted to make sure. The Chicago Manual of Style uses the terms "headline style" and "sentence style". The paper dictionary to which I refer most often now is The New Oxford Dictionary of English (2001 edition). It starts all entries with a minuscule unless there is a particular reason to do otherwise. I see nothing disingenuous about that. Eclecticology 19:53, 24 Jun 2004 (UTC)
            • (Sorry, I'm only familiar with the terms from word processors' autocapitalization functions.) The disingenuity follows from trying to equate dictionary formatting with Wiktionary formatting. The New Oxford Dictionary of English, if it is like other print dictionaries, doesn't format each entry as a page to itself. We do. —Muke Tever 14:05, 25 Jun 2004 (UTC)
              • What difference does it make that we have each entry as a page? What is so insincere about following a model used by print dictionaries? Just because someone's views are different from your POV does not make them dishonest. Eclecticology 06:51, 29 Jun 2004 (UTC)
                • The difference having each entry as a page makes is that the semantics of an H1 tag are different from the semantics of a DT tag (the dictionary analog of which is the boldfaced headword). —Muke Tever 13:52, 29 Jun 2004 (UTC)
• These are coomputerese arguments, and not lexicographical ones.
  • • • • • • • • The "insincerity" about following a model used by print dictionaries is assuming it is a model all print dictionaries use and have used, which we must blindly follow. I checked eight online versions of print dictionaries (four I regularly use "*" and four more at random "!") and found two with headwords in ALL CAPS (Hobson Jobson* Devil's Dictionary!), three with Initial caps (like we have now: Webster 1913*, Chinook-English!, English-Greek!), and three with no special capitalization, as is being proposed: (Gaelic Etymology!, AHD4*, Merriam-Webster*). I would like to see arguments for detitling articles other than "it's what print dictionaries do" because even if that were necessarily the case, it doesn't automatically follow that it's what Wiktionary has to do. —Muke Tever 13:52, 29 Jun 2004 (UTC)
  • I have never said that all print dictionaries use the proposed system, so your ad hominem comment is unwarranted. What's interesting in your list is that the Merriam-Webster is a modern descendant of the previously mentioned 1913 Webster. I just went through a dozen reference works from different publishers. Of these the Petit Robert uses all caps, and the Velazquez English-Spanish uses initial caps; all the others use the proposed system, but that is just a numbers game. Both the Chicago and Oxford style manuals recommend lower case lists in indexes, and the Chicago does mention that capitalized entries are an older way of doing things. I know that this reference is for indexes rather than dictionnaries, but I don't know of any style manual that is specifically for dictionnaries. The strongest argument remains the flexibility that optional first letter capitalization provides. Eclecticology 05:02, 30 Jun 2004 (UTC)

• It is the Correct™ thing to do.
  • POV. There exist dictionaries with their entries in title-case— viz. Webster 1913—so this is an opinion, not a factual statement of Rectitude™. —Muke 05:12, 23 Jun 2004 (UTC)
    • You misunderstood. This vote is about removing a technical restriction. If we really want titles in title-case (which I doubt the majority would want), we could still have that without the software forcing a broken capitalisation upon us (the "isiZulu" example that was given elsewhere is a pretty good one, as is "pH" and the like). — Timwi 14:33, 23 Jun 2004 (UTC)
      • I didnt misunderstand. You initially stated — and failed to remark here, though it is salient in many of my replies — that you would "run a script that renames every page to be lower-case, except for those pages that contain their own title in capitalised spelling" [4] and you have been fudging together that issue with the issue of automatic first-letter capitalization under the guise of the latter as a "technical restriction" (pushing kind vs. Kind is an example of this). —Muke Tever 16:43, 23 Jun 2004 (UTC)
        • It is a technical restriction because it is imposed by the software no matter how we link to the word. As for "running a script", I never read that as anything more than trying to find an efficient way to make the necessary corrections. If he has been fudging at all it is as a result of trying to understand phantom problems. Eclecticology 07:47, 24 Jun 2004 (UTC)
          • I don't think these are "necessary corrections". It is a non sequitur to assume that just because automatic title-casing is turned off, all pages should have title casing removed — especially since the only standing argument for turning it off is to handle pages that have anomalous title case. —Muke Tever 14:34, 24 Jun 2004 (UTC)
        • Perhaps I should have said "consequent corrections". In any significant change like this the need for corrections is understandable. Moving an article often creates broken links that need to be fixed, independently of whether or not some of us agreed with the move. Conforming to dictionary practice still stands as an argument. Required first letter capitals was adopted on Wikipedia to address the need to have piping when a Wikified word began a sentence. Eclecticology 19:53, 24 Jun 2004 (UTC)

What is already in the development plan

For those who aren't aware of the long history of discussion of this, here's what is already planned in the way of software changes, as developers get time:

1. Case preservation in article titles. pH would be displayed as pH when you get to the article. At the moment it isn't.
2. Case-insensitive searches and links, even with case-preserving titles. So ph, pH and PH will all go to the same place, as would Diesel and diesel.

The reasons we don't have this today are:

1. It requires a change to the structure of the cur table in the database adding a second space for the name, so both the case-preserved name and the case-insensitive linking name are available. This also requires the site to be unavailable, so it'll be done at a time when there are other changes which need to have the site unavailable. Has to be two spaces because that's just the way databases work. There's already a performance-related change planned which will require a change to cur at some point in the next few months.
2. It requires the writing of more program code to handle converting to the case-insensitive search form for a link and searching for that then displaying the case-preserved name when you arrive.

Drawback to full case-sensitivity (as opposed to case preservation):

1. Diesel at the start of a sentence would be a broken link if the article diesel exists, because diesel and Diesel differ in capitalisation. Case preservation with case-insensitive searching and linking avoids this problem.

Case-preserving titles with case-insensitive links will happen eventually, as developers get the time and inclination to do it. Inclination because developers are volunteers as well, and choose which programming to do just as we all choose which articles to work on - based on our individual preferences and judgements about which things are most in need of attention.

From some more discussion today, I get the impression that the people who are likely to do the work don't want two different articles which differ only in capitalisation and see disambiguation and see also as solutions we're routinely used and more than sufficient for the task. If you disagree with that view, please marshall your arguments to explain why it's vital not handle this the way we usually handle disambiguation problems and, perhaps more important, why you believe it is more frequently the case that we have words which need two different pages based on title than it is the case that we have words which should go to the same place whether they are at the start or middle of a sentence. Jamesday 03:03, 31 Aug 2004 (UTC)

---

Thanks for the explanation from the technical perspective. It clearly dispels the wrong impression that this required nothing more than turning a switch off or on. It also makes it clear that there is more to this issue than the politics of style.

Although my own arguments have been based on practice on the English Wiktionary, I have avoided extensive comments about what's happening in the other languages. I have never felt that it would be proper for users of this Wiktionary to impose their views on the other projects. My earliest comments on this subject go back to Wiktionary's first month, and my views have remained consistent.

I do not primarily regard this as a disambiguation problem, even though there are a significant number of words where that will be a factor. The vast majority of words will have a single entry which may be lower or upper case.

Capitalization can be morphological when designating such things as proper nouns or German nouns, or it can be syntactic as at the beginning of a sentence or in headline style titles. In most cases the syntactic will override the morphological, but there can be unusual exceptions. It is perfectly acceptable to begin an English sentence with "pH" or "e. e. cummings". There can be a variation of these rules from one language to another, and we need to remember that a Wiktionary is about all languages expressed for the benefit of readers in one language.

Dictionaries are about morphology, not syntax. I have been not yet been able to find a formal standard for headwords in writing dictionnaries, which is not to say that it doesn't exist. The practice in a majority of dictionnaries is that headwords are in lower case unless there is a reason why a word should be uppercased. Some, notably Webster, do capitalize headwords. Yet others, such as Robert, put headwords in all capitals. The Chicago Manual of Style does, however, recommend that listings in the index of a book should be lowercased except for those words that would normally be capitalized in text. The user should have no doubt from the beginning about the appropriate convention for the word. The current workaround practice of repeating the word at the beginning of the definition section is unsatisfactory. The purpose of that practice may not be at all clear to the casual user of the Wiktionary. How often do any of us look at the preface or introduction of a regular paper dictionary when we want to look up a word?

In cases where the first letter of a headword is accented it needs to be pointed out that the accent can appear more clearly on a lower case letter. In some circumstances the use of accents on a capital letter is optional, and that can be a source of ambiguities.

As much as I may support the optional capitalization of headwords, I also agree that a search function should be able to find all entries of a word without regard to capitalization. The "Go" function should default to "search" when it discovers an ambiguity. Not only that but the search function should also be able to find words that differ only by accents or ligatures. The person who is unfamiliar with a word should be able to find it just as easily if he is unfamiliar with the proper accents. It will also help us in working with such issues as the Danish "Æ" (a separate letter), the French "Œ" (typically written this way but alphebetized as "OE"), the Dutch "IJ" or the Spanish "LL". Flexibility is important in adapting to the many different possibilities.

Finally, one of the stumbling blocks in the vote appears to have been Timwi's proposal for transition. Although I could go along with it, it still struck me as a secondary matter. For some voters it became a deciding factor in their vote. I believe that we should really concern ourselves with where we want this change to lead us. Unless we have made a clear decision to make the proposed changes, the transitional issue of how we get there is moot. Eclecticology 13:32, 31 Aug 2004 (UTC)

---

The need for wikipedia and wiktionary is different. Diesel will not be written in wiktionary as the word does not have a capital. The word English does need a capital in the English language, because of this case-preserving titles are necessary. Case insensitive searches are nice to have. As far as nl:wiktionary is concerned, this can be added later. The big problem is that other charactersets do capitalise as well and, I do not have these characters available to me. It really prevents me from writing correct wordarticles. Therefore it is a must have feature within a wiktionary with many foreign words added by people who do not regularly use that language. GerardM 10:18, 31 Aug 2004 (UTC)

Jamesday: I am glad to hear case-insensitivity is coming.

---

When reading the Jamesday stuff again, his argument re Diesel at the start of a sentence is not applicable to Wiktionary. Wiktionary is a list of words, maybe phrases that are only capitalised when they must be capitalised. In Wikipedia this would be a problem, however the request is to turn off first character capitalisation in Wikitonary only. GerardM 08:08, 17 Sep 2004 (UTC)

---

GerardM: I'm not quite sure how exactly capitalization of titles in other alphabets interferes with your editing. I deal with accented and uppercase Greek (among other things) on the la:wiktionary relatively often and I don't run across any trouble. By the way, diesel is an adjective derived from a proper name (Rudolf Diesel) so I wouldn't be surprised if it were capitalized in some senses (indeed, Wikipedia seems to use "Diesel" and "diesel" indiscriminately on Rudolf Diesel and Diesel cycle, though not on Diesel proper). Capitalization, where not morphologically dictated, is a fluid thing. —Muke Tever 16:23, 31 Aug 2004 (UTC)

Ec: Spanish LL shouldnt be a problem for anybody (or am I missing something?), though Catalan ĿL might. —Muke Tever 16:23, 31 Aug 2004 (UTC)

Perhaps so. My Velàzquez dictionary (one which does capitalize the first letters of headwords) treats "LL" and "CH" as sngle letters for alphabetization, but does not capitalize the second element. I can't comment on Catalan. That this issue should come up at all speaks to the need for flexibility. Eclecticology 20:17, 31 Aug 2004 (UTC)

What diesel does in wikipedia is irrelevant to wiktionary. A surname does not make an article for a word. When there is one word compulsary capitalised and one compulsary not capitalised, they are two words/two meanings.

No, the "word" diesel is in fact a surname, and used as an adjective is not compulsorily either upper or lower case; both are acceptable, with the one meaning "invented by Rudolf Diesel". The fuel-related meaning is a different matter, and is, yes, normally not capitalized, but I wouldn't be surprised if formerly this were less the case. This applies for many words derived from names. —Muke Tever 22:25, 31 Aug 2004 (UTC)

To me Russian, Georgian, Arabic, Hebrew, Korean, Tamil etc are all squirly bits. When I know the text to be copied from a source, I expect them to be correct. When I then create a new word based on the squirly bits, they should not modify on their first character. It is hard to recognise this when something changes when they should not. Nouns for languages are capitalised in some languages and not in others. I rely on consistence between the squirly bits and the UTF-8 generated title. GerardM 18:18, 31 Aug 2004 (UTC)

Of the six languages mentioned only Russian and Georgian have such a thing as capital letters. Eclecticology 20:17, 31 Aug 2004 (UTC)

Then propably Serbian, Moksha, Mari, Mongolian, Bulgarian as well they are included in the etc. You can read this as: your remark is beside the point. GerardM 20:32, 31 Aug 2004 (UTC)

Not really. Serbian, Mokshan, Mari, Mongolian, and Bulgarian all use the same script as Russian: Cyrillic. The majority of the world's scripts don't have anything like case or capitalization. The major exceptions are Latin, Greek, and Cyrillic, whose capitals are not difficult to learn to distinguish (especially Cyrillic, whose minuscules are essentially small caps). There's also Georgian and Armenian, yes, but Georgian's capitals are not even normally used. In any case, if you cannot read what you are adding, how do you know it is correct to begin with? If they are all just "squirly bits" to you, how do you know that you even have the proper lower-case letters? Even if your foreign text comes with a transliteration, that doesn't guarantee that the original text is accurate (cf. the history of მეგობარი] —Muke Tever 22:25, 31 Aug 2004 (UTC)

In any case, if it helps you rest better, Georgian titles are not automatically capitalized, even now. Thus all you really should have to worry about is Armenian. —Muke Tever 22:31, 31 Aug 2004 (UTC)

By using a "trusted source" I do not worry to much about having the right word for a translation. Farsi does have capitals, so Arabic must have them as well, however the rules of capitalisation are a tad different. Only to have to worry about Armenian, is enough of a reason to want to have first character capitalisation to be turned off. There is a growing amount of Armenian words on nl:wiktionary GerardM

Excuse me if I'm missing something. I can find the UTF-8 listings for the Arabic script and for the presentation forms, but nothing for the Farsi capitals. Since you are clearly familiar with UTF-8 perhaps you could let us know the range of UTF-8 number codes that cover these particular Farsi capital squirrelly bits. Eclecticology 17:19, 1 Sep 2004 (UTC)

آآ اا بب پپ تت ثث جج چچ حح خخ دد ذذ رر زز ژژ سس شش صص ضض طط ظظ عع فف قق کک گگ لل مم نن وو هه يي

The capital is the one to the left, some characters do not have capitals. Info from a native Farsi speaking person. GerardM 17:24, 1 Sep 2004 (UTC)

Farsi doesn't have capital letters! It has initial, medial, and final forms of the same letter. There is no concept of case. The form of the letter depends solely on whether it has other joinable letters on either side - just like cursive English writing but standardised. — Hippietrail 02:48, 2 Sep 2004 (UTC)

The actual votes

Votes in favor of complete case sensitivity

1. Timwi 00:19, 23 Jun 2004 (UTC)
2. Polyglot 06:28, 23 Jun 2004 (UTC)

   I wanted it (no initial caps) from the beginning, couldn't get it then, created a workaround, got used to that, first opposed the idea of a change, but now I'm strongly in favour of it again :-)

3. Yann 09:54, 23 Jun 2004 (UTC)
4. Arj 20:08, 23 Jun 2004 (UTC)
5. GerardM Is this vote only for en:wiktionary or is it for _all_ wiktionaries ??

   It is only for en. What happens on the others is up to the participants in them. -Ec

   Not entirely; wiktionaries like nl: do not get attention when operational assistance is required. nl: wants to move to UTF-8 badly. This is not done as "we do it in out own time, scratch your own itch". So, I do not think that a still small wiktionary like nl: can request a feature when it cannot even get operational assistance for existing procedures. GerardM 08:09, 24 Jun 2004 (UTC)

   Even worse, this fracas prevents other wiktionaries from moving over. The developppers do want all wiktionaries to be the same in this. GerardM 05:33, 30 Aug 2004 (UTC)

   See below for further comments

6. Eclecticology 07:52, 24 Jun 2004 (UTC) I've consistently supported this since Wiktionary's first day.
7. Jdforrester 04:16, 25 Jun 2004 (UTC)
8. Eric119 21:44, 30 Jun 2004 (UTC)
9. Millosh 02:59, 2 Jul 2004 (UTC)
10. SemperBlotto 15:37, 7 Feb 2005 (UTC)

NOTE: Two of the above votes are from individuals who are not registered or do not contribute in Wiktionary. - 30 June 2004

Votes in favor (with dependencies)

1. Jun-Dai 21:35, 1 Jun 2005 (UTC)

   I am only in favor of this if it involves adding an automatic link to the alternate form of the word (e.g., kind->Kind and Kind->kind) near the top of the page, which would, IMO, help contend with some of the confusion that would result from the new arrangement. One of the reasons that case sensitivity is not a problem in print dictionaries is that Kind and kind would be right next to each other on the page. In the Wiktionary, however, kind is no closer to Kind than it is to zoology. What's more, I don't think that case sensitivity is as much an issue as trying to find a way to break the various languages on to separate pages. English bank and German bank shouldn't be on the same page--they ought to be tabs on the bank page (which could have the English def. by default), or separated with some sort of disambiguation page.

   This is really the first vote on the matter from nearly a year ago. The cross links that you describe are essential, but with existing pages they would need to be prepared manually unless you feel they should be there when the link is to a non-existent page. I don't think we need separate pages for each language at this time, but this could change in the future if pages get too big. At this time we only have 13 pages in the main namespace that exceed 32k in length, and all of those are auxilliary or index pages. None are for single words. Eclecticology 02:07, 2 Jun 2005 (UTC)

Votes against

1. Raul654 00:35, 23 Jun 2004 (UTC)
2. Jrincayc 02:43, 23 Jun 2004 (UTC) I have already created enough redirects for things like Supply and Demand -> Supply and demand, and pluralization redirecets without needing to add even more. I can't think of a single good example of where the different case would make a different article.

   • How about a) script is less than perfect and leaves many articles uppercase, b) innocent user looks for word which already exists but only in uppercase form, c) user sees a blank page and assumes word has not been entered, d) user creates 2nd page for same word.
   • Where is this script so I can review it and see if will live up to the promises? — Hippietrail 17:06, 23 Jun 2004 (UTC)

3. Muke 05:12, 23 Jun 2004 (UTC) Strongly against case-sensitivity in titles.

   • Muke Tever 05:51, 23 Jun 2004 (UTC) Emendation of opinion: The page titles are just that, titles, and should be title case — or at least sentence case, which is the current practice. I firmly disbelieve that Wiktionary article titles should be lowercased although I don't think that the simple "first-letter-is-capitalized" rule is sophisticated enough to handle actual title/sentence-casing rules (cf. below) either.

4. Johnleemk 13:28, 23 Jun 2004 (UTC)
5. Patrick 14:10, 23 Jun 2004 (UTC) - Linking a word at the beginning of a sentence would often become cumbersome, requiring a piped link.
6. Centrx 20:47, 23 Jun 2004 (UTC) - See above.
7. Frazzydee 00:12, 24 Jun 2004 (UTC): I'm all for case-sensitivity in the rest of the title, but not for the first letter. I fear that people will create redundant articles.
8. \Mike 12:54, 24 Jun 2004 (UTC): I still haven't seen any good arguments pro...
9. Avriette I agree with the above argument that this is not a simple vote for or against capitalization. This is logically several arguments. At the least, whether Noun and noun should be different pages, whether "the script" should be run, whether all wiktionaries should accept the change, whether "the script" is accurate enough, and so on. I dislike decisions such as this which result in Action at a distance. I think before any decision is made, the true repurcussions should either be separated into separate votes, or the vote should be reworded and compartmentalized such that the decision made by the voters is that which they truly voted upon.
10. px 2004.07.09 17:37
11. Masterhomer 03:20, 15 Jul 2004 (UTC)
12. 68.192.172.119 07:30, 17 Jul 2004 (UTC)
13. 155.31.32.253 14:58, 17 Jul 2004 (UTC)
14. Ŭalabio 03:22, 21 Jul 2004 (UTC) - I vote for casepreserving with caseinsensitive searching. One will place capitalize pH properly, while one can still find pH be typing ph, Ph, PH, or pH.
15. 195.137.54.92 16:58, 4 Aug 2004 (UTC)
16. 68.38.129.228 07:33, 7 Aug 2004 (UTC)
17. Jamesday 03:12, 31 Aug 2004 (UTC) for reasons discussed in the current development plan section I added.
18. 217.43.52.203 21:26, 16 Sep 2004 (UTC) Case insensitivity is an entrenched feature of the internet. Having Wiktionary behave differently might be technically correct but will inconvenience users.

NOTE: Twelve of the above votes are from individuals who are not registered or do not contribute in Wiktionary. 05 Jul 2004

The software changes involved will be available for ALL projects, not just English Wictionary. Jamesday 03:12, 31 Aug 2004 (UTC)

Abstaining for now

• Hippietrail 03:28, 23 Jun 2004 (UTC)
  • This whole page is unclear. What are we voting "for" or "against"? Is it "for case-sensitivity" or is it "for turning off case-sensitivity"?
    • It says that clearly: "Vote for" means "kind" and "Kind" should be separate pages. — Timwi 14:33, 23 Jun 2004 (UTC)

      I've added a couple of the things which you will also get if you vote for this one thing. — Hippietrail 17:06, 23 Jun 2004 (UTC)

  • It fails to take into account that there are 3 possibilites: "fully case-sensitive", "fully case-insensitive", and the current scheme: "case-sensitive for all but the first letter".
    • "fully case-insensitive" is not an option we currently have in the software. It would take more programming. This vote is on whether or not to flip a simple switch. — Timwi 14:33, 23 Jun 2004 (UTC)

      Calling an upper_first() function ad opposed to an upper_string() function is small enough to ask the developers for rather than settling on a half-solution too early. — Hippietrail 17:06, 23 Jun 2004 (UTC)

      The switch may be simple, the results of the script and mistakes made by unaware users may not be. The two go hand-in-hand. — Hippietrail 17:06, 23 Jun 2004 (UTC)

      • I'm afraid it's not that simple. Currently, each article has a canonical title which is used for both display and indexing. Full case insensitivity would require separately stored titles for each of the uses. This is far from trivial. -- Tim Starling 01:45, 28 Jun 2004 (UTC)
  • The description is overly simplistic, telling us of one problem and one solution. There are more subtle issues.
    • Removing this technological restriction does not cause any "issues". Maybe having "kind" and "Kind" be separate articles might be an "issue" in some people's POV, but it's not. — Timwi 14:33, 23 Jun 2004 (UTC)
  • The very heading "Arguments that aren't really arguments" is POV.
    • (I changed "NPOV" to "POV" because I assume that's what you meant.) It's not POV because I'm justifying why I think they are not arguments. — Timwi 14:33, 23 Jun 2004 (UTC)
      • If you don't call that POV, I'll exercise the right to do the same to other arguments. —Muke Tever 16:43, 23 Jun 2004 (UTC)
  • The "Example Articles or Words where this should be done" section containing just question marks is POV.
    • You should take a closer look at the history. I'm not responsible for it. I've removed it again. — Timwi 14:33, 23 Jun 2004 (UTC)
  • This is much better discussed on the mailing list where I've just posted a lengthyish response.
    • It was already discussed months ago. Why does everyone want to have things discussed over and over again? — Timwi 14:33, 23 Jun 2004 (UTC)

      The current discussion already seems more than anything I noticed before, and I consider myself to be very active on Wiktionary. — Hippietrail 17:06, 23 Jun 2004 (UTC)

• Muke 05:12, 23 Jun 2004 (UTC) Abstaining on decapitalization of the initial letter
  • I think we can come up with better solutions — I like Hippietrail's idea of suppressing the title entirely or at least deemphasizing it
    • This is not currently technically possible. — Timwi 14:33, 23 Jun 2004 (UTC)

      Your solution is not necessarily better than pushing for a real solution, especially if the developers can find a simple way to implement them. — Hippietrail 17:06, 23 Jun 2004 (UTC)

  • Failing that, I think that page titles should actually be case-insensitive, with the title being displayed in title case.
    • This, too, is not currently technically possible. — Timwi 14:33, 23 Jun 2004 (UTC)

      Changing code is technically possible. Perhaps it is politically difficult or technically difficult, or would just take longer than you would like to wait. These issues should be addressed in your presentation. — Hippietrail 17:06, 23 Jun 2004 (UTC)

  • However, there may be needed a special directive for words whose title-case is not formed by capitalization of the initial (as in Bantu words like isiZulu, or acronyms, like NASA)—the first is an issue now, though as long as titles are still case-sensitive, the second is not a problem.

---

Wiktionary EN only???

When this was first proposed, it was said it was going to be implemented across all Wiktionaries. Wiktionary DE had asked for it, but we had to agree on Wiktionary EN, since the impact was going to be the biggest due to the amount of pages we already have. I think the other Wiktionaries and certainly NL have to implement this change as well, especially now that it is possible to get a clean start over there.

I would of course agree that it would be a good thing on all Wiktionaries. But I also strongly believe in the autonomy of each project. Actually when the idea was first proposed, there was only one Wiktionary. I do not speak German, so it should not be my right to tell the Germans how they should be running their Wiktionary. Eclecticology 09:32, 24 Jun 2004 (UTC)

Wiktionaries like nl: do not get attention when operational assistance is required. nl: wants to move to UTF-8 badly. This is not done as "we do it in out own time, scratch your own itch". So, I do not think that a still small wiktionary like nl: can request a feature when it cannot even get operational assistance for existing procedures. GerardM 08:09, 24 Jun 2004 (UTC)

Getting attention for operational assistance is beyond the scope of this discussion. Moving en:Wiktionary to UTF-8 came up very early in its life. It helped that Brion supported this. There are still a few problems arising from pre-conversion entries, but these are all quite minor. I absolutely agree that all the projects should be on UTF-8 (unless the people there don't want it). The operating parameters for nl:wiktionary were imported from nl:wikipedia which has not yet converted. Eclecticology 20:08, 24 Jun 2004 (UTC)

GerardM, stop being so unfair. Brion is working hard on the UTF-8 conversion. Just because you can't get it today doesn't mean you're being neglected. — Timwi 20:15, 24 Jun 2004 (UTC)

I am not being unfair, I do not know. I try to be informed; I am subscribed to the wiktionary-I, wikitech-I and wikipedia-I mailinglist in order to know what is going on. I go to meta frequently so I do my best to be _informed_. There is no feed back when I make a request or it is "go away" , everything is geared to be of use to en:whatever.

nl:wiktionary can go off line for some time (duration conversion script) without notice. We are now 832 pages, the majority of the nl:words have special characters. If post conversion problems are the future, it helps when the conversion is sooner as it will mean less post conversion blues. GerardM 08:53, 28 Jun 2004 (UTC)

I'd really much prefer it if all the Wiktionaries were converted, or none of them. Myriad minor distinctions between wikis make things hard to manage. Also it would obviate the need for a wasteful repeated discussion for each of the 150 wiktionaries. In fact, Fire asked me to switch of first letter capitalisation for de.wiktionary.org, but I refused, telling him to ask on the mailing list if it could be done for all wiktionaries. He did so, but apparently no-one noticed. -- Tim Starling 01:45, 28 Jun 2004 (UTC)

I totally agree that all Wiktionaries should be converted or none of them. I think there's a large chance that in the future the Wiktionaries will be linked together, and such minor differences will be cumbersome. --Georg Muntingh 16:37, 24 Aug 2004 (UTC)

There is no point in linking all wiktionaries into one when things like this cannot be resolved in the better technical way for a dictionary.

Apparently en:wiktionary decides for all wiktionaries

It has been said that all projects are independent. However in practise this is not the case. de:wiktionary and nl:wiktionary do not want capitalisation. For practical reasons it is better to change the capitalisation on the smaller wiktionaries. However from a system management point of view it has been decided that we are all going or we are not going.

To think that the democratic choise of en: should hamstring all other wiktionaries is a bit much. The thing is that many people do want en: to change. De: and nl: want to change, how is this going to be resolved ?? GerardM 17:38, 24 Aug 2004 (UTC)

De: and nl: want to change. Why? Are you sure this is the case? --217.43.99.99 17:12, 14 Sep 2004 (UTC)

It was the German Wiktionary who started the request less than one month after they got started. This is because capitalisation is important to them. Gerard has up to now contributed almost 99% of the Dutch Wiktionary, so yes, he is sure the Dutch Wiktionary want this.

Back when the English Wiktionary was started in December 2002, the main contributors also wanted different pages for entries with different capitalisations. We were told to make do with the software as it was designed for Wikipedia, and thus that's what we did.

Now it would be possible to do things properly and it's a pity that there is such an incredible opposition to this. The kind, Kind] problem will be resolved. Don't worry about that. There will be links between those pages. 07:49, 15 Sep 2004 (UTC)

I understand that there would be links between the pages, but if I am a user of the Wiktionary and searching for other uses of the word 'kind/Kind' (say) in various languages, I want to see all the possibilities on a single page. I think most users will tend to look up by default in lowercase since search engines generally do not distinguish (http://www.google.de/ ). I accept that in some languages it is technically correct to distinguish between capitalisations, but from a user's point of view - even, I would charge, in those languages - it's a detrimental change. That would be my objection. --217.43.52.203 21:23, 16 Sep 2004 (UTC)

The question of the en:wiktionary preventing other wiktionaries from changing to non capitalisation is history, many wiktionaries HAVE changed to non capitalisation. The Polish is contemplating it, de: fr: it: nl: cs: hi: have already moved. One word of advise, the longer you wait, the harder it is. GerardM 07:16, 1 Nov 2004 (UTC)

Sv: has moved as well. \Mike 07:59, 1 Nov 2004 (UTC)