Talk:ISO 639-2

What's the difference between "bibliographic use" and "terminological use"?--218.102.231.133 03:22, 31 December 2005 (UTC) Reply

The difference reflects the way that ISO 639-2 was put together. For about twenty years before ISO 639-2, libraries used a three-letter code to reference the language in which a book or article was written. Most of these codes were based on the English name for the language, like ger for German. The two-letter code ISO 639 (which later became ISO 639-1) was often based on the native name for a language, like de for German. It was felt that the larger ISO 639-2 standard should reflect the policy of part one rather than that of the library codes. However, lobbying from libraries, who would be a major user of the standard, forced a compromise. On the twenty-two codes where there is significant difference between library codes and ISO 639-1 codes, ISO 639-2 provides two codes, each reflecting one of these codes. The library codes from the basis of the bibliographic code, and ISO 639-1 formsthe basis of the terminological code. For most uses, where there is no existing coding system, the terminological codes are prefered. The draft ISO 639-3 standard uses terminological codes where there are two ISO 639-2 codes. To sum up, German has ISO 639-1 de, ISO 639-2/B ger, ISO 639-2/T deu and ISO 639-3 deu. --Gareth Hughes 14:34, 31 December 2005 (UTC) Reply

Thanks--218.102.231.133 02:30, 1 January 2006 (UTC) Reply

Bibliographic codes are an anglocentric atavism and practically deprecated. RFC 3066 says: [2.3 Choice of language tag] 3. When a language has no ISO 639-1 2-character code, and the ISO 639-2/T (Terminology) code and the ISO 639-2/B (Bibliographic) code differ, you MUST use the Terminology code. NOTE: At present, all languages for which there is a difference have 2-character codes, and the displeasure of developers about the existence of 2 code sets has been adequately communicated to ISO. So this situation will hopefully not arise.--84.188.172.180 18:42, 4 February 2006 (UTC) Reply

Hello, i have a question. In the "Collective language codes" section there's the line "col Shilluk language".

But Shilluk is a language not "languages" (Shilluk language), so it can't be a collective language. And its code is "shk", not "col", and Shilluk has only a 639-3 code, see: List of ISO 639-2 codes, Documentation for ISO 639 identifier: shk, at SIL, ISO 639 Code Tables /S, at SIL

"col" = Columbia-Wenatchi, and also has just a 639-3 code (Columbia-Moses language), so it's also not a collective language (see also: ISO 639 Code Tables /C, at SIL, Documentation for ISO 639 identifier: col, at SIL).

And both (shk/Shilluk & col/ColWen) are missing from the official code lists of ISO 639-2 & 639-5 (see: Codes arranged alphabetically by alpha-3/ISO 639-2 Code, ISO 639-5 codes ordered by Identifier; also: List of ISO 639-2 codes & List of ISO 639-5 codes)

So i think it is maybe a mistake. Or not? Thanks for the answer. — Preceding unsigned comment added by ELVe (talk • contribs) 08:27, 30 September 2012 (UTC) Reply

What ISO are of Proto-Baltic, Proto-Slavic, Proto-Germanic?--Ed1974LT (talk) 18:30, 20 August 2016 (UTC) Reply

I need to find an ISO CODE TWO-LETTER of the ENGLISH LANGUAGE 177.142.244.53 (talk) 13:27, 22 March 2025 (UTC) Reply

If you're looking for two letter codes, you want ISO 639-1, not ISO 639-2. The first example in the ISO 639-1 article is en for English. DRMcCreedy (talk) 16:07, 22 March 2025 (UTC) Reply

This is a bit of a ridiculous example of formal WP Policy legal marking of missing sources and/or citations, and I like to understand how that came to be, so I followed article's history.

Because situation might be a bit confusing, I'd like to let other editors have an idea what happened, and what I intend to do before I do it (I'm neither "an old salt" at WP, nor a native speaker (or writer as is the case), so (by proposal given by one source template) I initiate a discussion here.

1. When user:Tobias Conradi created this article in 2005 (changing it from existing redirect) he included external links section with three links to Library of Congress, which seem to have been and to still be the registrar and primary authority on language code standards in his first edit. (Seem - because I have not gone searching for Ivy league published peer reviewed sources to formally confirm it - I just went to the ISO 639-2/RA Homepage link and - since I saw nothing suspicious of phishing - otherwise I would proceed critically (checking IP adresses regarding DNS and about authority of this page's authors' at ISO) - I concluded that the article was reliable sourced already from creation.
2. Looking into histories (both of this article and of MAchine Readable Cataloguing (MARC) of librarian methodology I found that Library of Congress was (and still is in 2005) between primary agents in development of those methods and tools since late XX century, which got standardized (in US) Nationally, and internationally (ISO). Conceptually, the article was at least adequately sourced from creation.
3. By WP policies, formally, the article was not sourced (no inline citations, no references section; external links don't count - WP:REF: "Wikipedia's verifiability policy requires inline citations for any material challenged or likely to be challenged, and for all quotations, anywhere in article space. ").

Formally, all lists of codes and languages in that (and related) articles could be called (unmarked) quotes, and, by WP policy, require inline citations.

1. Conceptually (regarding contents being verifiable), all was good and well till June 2017, when @Ringbang reformated a list of codes and languages to columns, and moved nav boxes below external links section, and, without checking the contents, renamed the link * to Official website (by formatting it with a template, probably unintentionally).
2. Probably at preview he spotted no reference section and no references template, and tagged the article with template as no sources, which was technically true (even more after his renaming of the Library of Congress department's for (MARC standards) ISO 639-2/RA Homepage to "Official Website", instead of e.g. Library of Congress Standards International Organization for Standardization (ISO) 639-2 Registration Authority Home page...)

Edit: He just intended to move/use the Official Website link from Wiki data, but failed to see the minor difference to see between WP page name ISO 639-2 which was the name of the standard, and original name of the link ISO 639-2/RA, which was link to the registration authority's official page, and obfuscated the relation

Next try with inserted <ref name="ISO MA TOR"/> after lead, to support the statement about Registration authority there, by calling named citation defined below {{reflist}}, which gave

Cite error: The named reference ISO MA TOR was invoked but never defined (see the help page).

but works fine here.^[1]

Tried to bypass (not solve) the problem by exchanging the named citations so definition goes above and use comes below. Doesn't work, below first {{reflist}} I get ref name not defined.

Tried to patch (not solve) the problem by putting same citation (same name and contents) both above and below first {{reflist}} so definition goes above and use comes below. Kind of works:

• Both citations visible and correct, both number one (by luck in this case)
• Second {{reflist}} needed above navigation errors
• Cite Error check shows errors: "Multiple references contain the same content", which stays,
• I remove "Multiple references are using the same name" by changing second call to non-named ref, so {{kind of done }}, or {{kind of patched }}.

1. After @Rich Farmbrough added data with an inline citation in August 2020 , he spotted it located under navigation tags and added Reference section and <References> maintenance tag above it.

A day or two later @Rfl0216 spotted a single citation with Unreferenced tag on top, changed it to single source tag (with then current date). He (like Ringbang before missed there were sources, but not in the expected section.

There seem to also have been overlap (they seem to have both added reference section and tag/template - and it seems one removed the other's then duplicate section, but that ended well anyway)

Beside a lot of adding and updating of contents, and some vandalism and it's reversals, this is what I found relevant about verifiability of this article's contents.

What I intend to do:

1. Move links, that support the article in general, from external links to source section (to be created) below existing reference section.

Edit: By content I would need to e.g. support the statement, that Library of Congress Network Development and MARC Standards Office residing in external links sedtion, is a/the designated ISO 639-2 Language Coding Agency and ISO 639-2 Registration Authority^[1] , which I have verified and can show here with no errors (but I get maintenance message(s) notification I didn't find yet; the real problem is technical, since this citation is shown after navigation templates, not with others at the reflist template call above External links section. Patching with second relist template call above navigation templates. {{kind of patched }}, see above

it seems that citation must precede references tag, which here, to succeed, would need to move external links section before references section (which works in preview, but is against WP:MOS).

1. Modify misnamed links (if any still exist) Done
2. remove single source tag template from the article (Now at least one Reference to ILO, LOC and Microsoft) Kind of Done (commented, hopefully someone more versed looks at it after me)

Marjan Tomki SI (talk) 13:41, 23 July 2025 (UTC) Reply