[Ltru] Possibly a way forward? (was extlang & deprecation (was draft updated...))

["Mark Davis" <mark.davis at icu-project.org>]

I ran the following by a few people, and it may offer a way out of our current situation,  by taking the following steps.

1. Tie "Preferred Value" more closely to canonicalization.
   
2. Decouple "Preferred Value" and "Deprecated"; a subtag can have a Preferred Value without being Deprecated.
   
3. Make it very clear that when there is a Preferred Value or when there is Deprecated, the tag or subtag is still valid, and perfectly acceptable.
   
4. In particular, make it clear that the X-Y values are perfectly acceptable, and as valid as the Y forms.
   
5. Add that some implementations will find it more convenient to have their own canonical form. Give examples of "iw" for Java, and of X-Y forms.
   

Can anyone not live with the current draft plus these changes?

Mark

On Tue, Jul 1, 2008 at 6:31 PM, Doug Ewell <doug at ewellic.org> wrote:

John Cowan <cowan at ccil dot org> wrote:

That's not the definition of Deprecate:

Dictionary definitions of "deprecate" aren't relevant, because it is a term of art among programmers and standardizers.  From Wikipedia:

     In computer software standards and documentation, the term
     deprecation is applied to software features that are superseded
     and should be avoided. Although deprecated features remain
     in the current version, their use may raise warning messages
     recommending alternate practices, and deprecation may indicate
     that the feature will be removed in the future. Features are
     deprecated -- rather than being removed -- in order to provide
     backward compatibility and give programmers using the feature
     time to bring their code into compliance with the new standard.
             http://en.wikipedia.org/wiki/Deprecation

Actually, while this disagreement over the use of the word "deprecated" may seem academic, I think it cuts to the very heart of the X-Y versus Y controversy.  People on this list -- even the same person at different times; I'll dig up quotes if asked -- have said that (1) deprecated subtags are perfectly fine to use, because they'll just get normalized away, AND that (2) deprecated subtags are something to be avoided at all costs.

We can't have it both ways, and expect those on the opposite side of this battle to accept their preferred forms being "deprecated" on the basis of argument 1 ("it's still OK to use them") when we still have draft text, and arguments on the list, that their use will be considered non-conformant.

As I've said before, we can get so caught up in our own terminology that we forget that the general tagging populace may not understand these terms the way we understand them -- and without a glossary, we can't fault them for devising their own definitions.  I agree with John that the use of "deprecated" in the programming world is different from the mainstream dictionary definition, which evokes visions of burning at the stake, but we don't even agree among ourselves exactly what it does mean.

Until we do that, and unless we decide formally that "deprecated" has the gentler meaning of "OK to use, but will be normalized away," the battle lines between the "zh-cmn" crowd and the "cmn" crows will continue to be drawn.

--
Doug Ewell  *  Arvada, Colorado, USA  *  RFC 4645  *  UTN #14
http://www.ewellic.org
http://www1.ietf.org/html.charters/ltru-charter.html
http://www.alvestrand.no/mailman/listinfo/ietf-languages  ˆ

_______________________________________________
Ltru mailing list
Ltru at ietf.org
https://www.ietf.org/mailman/listinfo/ltru

_______________________________________________
Ltru mailing list
Ltru at ietf.org
https://www.ietf.org/mailman/listinfo/ltru