This proposal is not good for Wiktionary. It goes contrary towards the failed unification attempts of before. I'm against this proposal. The voting on it should not be allowed. -- Bugoslav17:17, 8 June 2011 (UTC)Reply
We are accepting all natural languages. We are just disputing whether Bosnian, Serbian and Croatian are languages. Furthermore, that part of CFI is intended to mean 'if it is spoken on Earth as part of a non-constructed language, it may be included regardless'. It doesn't mean we should include all languages as distinct languages, it just means that all words should be covered regardless of language. —CodeCat18:04, 8 June 2011 (UTC)Reply
I do not agree with your interpretation. Read WT:CFI and you will see for yourself. Read what is written on top of that page.
»As an international dictionary, Wiktionary is intended to include “all words in all languages”.«
You still seem to think that your interpretation is the only possible one. But the term is put in scare quotes which indicate that the term is open to interpretation both on what a word is, and what a language is. A vote will decide how the Wiktionary community chooses to interpret it, which rules out any individual bias. If the outcome of the vote is unification, and you feel it goes against the core principles of the Wikimedia projects, you're free to argue your case there. —CodeCat18:14, 8 June 2011 (UTC)Reply
A cautionary word to start with
This proposal is about the Wiktionary categories. I'd much rather we stay on topic, discussing how Wiktionary should approach these languages, avoiding personal attacks and general unpleasantness.
Regarding Bugoslav's comments, Wiktionary has no standards or guidelines on what "language" means. Where does one language start and one language end? There is genuine debate over whether these are one language, or two, or three, or four! ISO 639, according to its code, says they are three languages. While we usually follow ISO 639, there are exceptions. As well as a number of languages with ISO 639 codes not being allowed, we have Wiktionary only codes (for example {{roa-jer}} - Jèrriais). Believe it or not, I have no real feelings on the matter. A superficial look into the subject makes me think they are one language. But to be honest I only want to uphold the consensus on Serbo-Croatian. My sincere advice to Bugoslav is not to insult other users' intelligence. There is real debate over whether these are one language or not. Pretending otherwise will not fly, and will likely only irritate other users. This is my advice, take it or leave it as you wish. Mglovesfun (talk) 20:34, 8 June 2011 (UTC)Reply
The code "sh" is deprecated (thus, presumably disencouraged) by ISO, in favor of using "hbs", but the two-letter version is the one chosen and widely used on Wiktionary nonetheless. --Daniel22:02, 8 June 2011 (UTC)Reply
Two-letter codes are not allocated anymore by ISO, so it's perfectly fine to use sh (as well as any other unallocated two-letter code should the need arise) for our purposes. --Ivan Štambuk06:41, 13 June 2011 (UTC)Reply
True. If I remember correctly, ISO also has a policy of never recycling deprecated codes (i.e., never assigning them to something else), so we can always use them freely. --Daniel07:07, 13 June 2011 (UTC)Reply
Again, just to make things clear to you, Bugoslav: I'm sure you will see if you consider the proposal logically, setting all points of view aside, that nowhere in it is there any suggestion of exclusion of any word or language; its sole object is shared categorization and presentation format. The premise is: 1. It is generally undisputed (in the Balkans and elsewhere) that “Serbo-Croatian” (or BCS(M), Central South Slavic Diasystem/dialect continuum or whatever else you want to call it) constitutes a genetic node. In any case, the international standard ISO 639-3, supported by Croatia, Bosnia & Herzegovina, and Serbia, defines “Serbo-Croatian” as a “macrolanguage” containing the “languages” Bosnian, Croatian and Serbian. 2. For the purposes of Wiktionary, entities that could be considered different “languages” may be (and sometimes are) lumped together as a single “language” (there is, as Mglovesfun rightly says above, no absolute standard here on what constitutes a “language”, and this has to be decided on a case-by-case basis, and only for the purposes of Wiktionary, without any wider scope). This lumping approach is the case with here with Hebrew (everything from Biblical to modern vernacular), Latin (pre-Classical to modern, plus Late Latin), Ancient Greek (Epic, Classical [Attic, Ionian, etc.], Koine, Byzantine, and, it appears, Modern Katharevousa), Dutch (covering, among others, Flemish [vls] and Zealandic [zea]), Min Nan (Amoy, Teochew, etc.), and more. In this case, as in these others (perhaps even more so), it is (as I see it) practical to lump together rather than keep separate sections and categories, as it makes it easier for users to find relevant information in one place (also notes about different and overlapping regional usage), decreases clutter, makes it easier for Wiktionarians to keep entries in sync and up to date, etc., etc. To date, I have not seen anything about this approach which is in some way inferior or more inconvenient than keeping the separate sections. – Krun22:30, 8 June 2011 (UTC)Reply
(unindent) Actually I didn't know Serbo-Croatian still had a valid ISO 639-3 code! I knew that {{sh}} was no longer active, it never occurred to me that the ISO 63-3 equivalent still was! Mglovesfun (talk) 22:48, 8 June 2011 (UTC)Reply
I don't speak Serbo-Croatian/Serbian/Croatian/whatever. I just see good arguments in favor of the proposal. So, I support it too. --Daniel04:26, 9 June 2011 (UTC)Reply
As Wiktionarians (vast majority) we work in the field of practical lexicography, that is, in the art or craft of compiling, writing and editing dictionaries.
Wiktionary is not a project about General linguistics, but about creating a dictionary and a collaborative community around it. That's all there is to it!
Wiktionarians cannot and should not decide on whether or nor something is or isn't a language. For God's sake, we have → ☺ (Category:Emoticons).
There are enough sources which deal with those languages as separate but closely connected languages. Some, like SIL International clearly list the three as individual languages (within some larger structure, group, whatever), but still as individual languages. Some sources, predominantly before 1980s, describe the group as "one language with two varieties". And some, like Škiljan wrote about 4 centres, Zagreb, Belgrade, Sarajevo and Titograd. He considered that those 4 centres have evolved (1988) into 4 different varieties (sociolinguistically).
In Croatia, from December 15, 1997 the field of lexicography is no longer considered a part of philology, but instead as part of information sciences. Please consider that fact.
Lexicography is more connected with the information sciences and the presentation of knowledge than theoretical linguistics. Sorry for the long reply. Much obliged. -- Bugoslav03:06, 12 June 2011 (UTC)Reply
You said this is a discussion, not a vote. Actually, Wiktionary discussions are a little of both things, as people can still just throw a "support" or choose to elaborate their opinions.
However, SIL and Dubravko Škiljan don't own Wiktionary, and (I think) aren't Wiktionarians. So they don't even "vote" here. Merely associating a proposal with their opinion can be futile.
Wiktionary is in itself an authority; we can make our decisions by ourselves. We recognize certain authorities, and others we don't, depending on certain criteria, contexts, arguments and consensus.
In particular, many people know that Serbian/Croatian/Bosnian/Montenegrin exist, and even know how authorities treat them (particularly the different ISO codes, which are serious business here) but still see "Serbo-Croatian" as a better language header.
It does not mean that "Serbian" does not exist, or that we don't acknowledge its existence. We do! We even have an entry for it: Serbian. --Daniel03:55, 12 June 2011 (UTC)Reply
FWIW other editors who haven't participated in this discussion but are taking part in the merge include Ivan Štambuk, Krun and JorisvS. Mglovesfun (talk) 11:31, 12 June 2011 (UTC)Reply
(somewhat offtopic) Bugoslav, I wonder why have you chosen to point out only my comment and considered it a vote. Some previous commentators (the ones which commented before me) were giving their support also, but you chose to single out only my comment. Do you have some grudge against me? I hope you noticed that I was one of the editors who supported your unblocking? I'd really like you to elaborate on this. Also giving support isn't necessarily a vote, IMNSHO. What do you think?
(more on the topic) So even if it is that in Croatia lexicography is no longer considered as part of philology or moreover that lexicography is not a part of philology at all or even not a part of linguistics for that matter, would you then be bold enough to state that grammars of "Croatian language" and "Serbian language" (or "Bosnian" and "Montenegrin" for that part also) differ that much, that they should be considered different languages? If so, it might be helpful to elaborate on those differences in grammar and hopefully to give some numerical estimate of level of difference among those languages based purely on grammar differences. Of course some explanation for the numerical estimates given could also be useful. I think replying to my request could help the discussion and even its eventual conclusions. Regards, --BiblbroXдискашн01:03, 13 June 2011 (UTC)Reply
I may regret this, but what does WT:NPOV have to do with this? If you're asserting that they're three (or four) languages, that is in itself a point of view. I don't see how saying they are three (or four) languages is 'neutral' while saying they're one is 'non-neutral'. On second thoughts, this is quite a good question and I await a reply. Mglovesfun (talk) 20:58, 13 June 2011 (UTC)Reply
