Appeal to ISO 639 RA in support of Elfdalian

Shawn Steele Shawn.Steele at microsoft.com
Sat Apr 23 21:41:24 CEST 2016


> But I can't add a private subtag to any other platforms!

I’m not sure exactly what you mean.  If you pass x-elf and someone wants to recognize that, then they may.  If it’s well formed most systems will treat it similarly to any other tag they don’t recognize.  So any advantage over a real “elf” (or whatever it would be) tag is minimal for systems that don’t have wide adoption of the language yet.

> And I think we should put very little weight into what you prefer, as you're not even planning to use the code yourself!

I suspect I will encounter the code, particularly if you go further and ask for CLDR support or my users/developers have interest in the language :)  If you would like it to be available in the Windows Language profile list, I’m the person that will be touching that data to make it happen :)

I believe that I may not personally plan on using the code, however my job is to ensure that Microsoft systems behave for the codes on our systems.  So my job isn’t to use the code, but to make sure that you are able to use the code.

On Windows, if it is, indeed a distinct language, then making it a subtag of Swedish is going to likely cause all sorts of interesting challenges.  Also making it an unusual form of a BCP 47 tag, like the 5 letter code, is going to break all sorts of applications that took shortcuts and assumptions about the form of BCP 47 tags.  Deviating from 639 will also break assumptions similarly.

I presume that you would like your code to work across the entire system, and/or other platforms, web sites, etc..  It’s my experience that a new 3 letter ISO code can get traction pretty quickly.  Something like a 5 letter theoretically useful form, or a language tag that supplements 639 is going to break applications.  In some cases it will take a very long time to fix those applications, if ever.  Many of them aren’t going to see Elfdalian as a very interesting case and won’t bother trying to figure out how to make 5 letter codes work, or to extend their dependency on the ISO 639 tables.

I believe that the best way for the Elfdalian users to have broad support is with a real ISO code.  That may not be the fastest way to get a code for narrow use by some applications in the next 3 weeks, but longer term it is probably less painful.

> And I don't think that that the RA will look into this matter with such a childish perspective and say anything like  “see, you don’t really need it.”

I would like to think that standards bodies are above that sort of thing, however my experience has been otherwise :(

-Shawn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.alvestrand.no/pipermail/ietf-languages/attachments/20160423/986fedde/attachment.html>


More information about the Ietf-languages mailing list