Sign languages

Mark Davis mark.davis at
Mon Feb 27 03:39:00 CET 2006

I have seen no evidence that a different syntax needs to be used for 
signed languages than for non-signed. Two different signed languages 
really ought to just have different 639 codes, just like two different 
oral languages. We may end up treating sgn like a macrolanguage once 
639-3 comes into play; that might be the best way to handle it.


Doug Ewell wrote:
> Michael Everson <everson at evertype dot com> wrote:
>>> Blech! So you want to have "sgn-GH" for 'Ghanian SL' and something
>>> like "sgn-GH-admrb" for 'Adoromobe SL', even though the two are
>>> probably quite distinct, if even related at all. That will work
>>> beautifully (NOT!) with left-prefix matching algorithms.
>> Such algorithms are not the reason we have language-tags, I am sure.
> Such algorithms will dramatically improve the usefulness of language 
> tags, by allowing applications to match and filter them in a standard 
> way that humans will understand.  The benefits are already 
> well-understood for non-manual languages.
> I get the feeling the only acceptable way to tag sign(ed) languages 
> will be to use Michael's scheme from 2001, which -- because of tags 
> like "sgn-MY-B" and "sgn-GH-EP" and "sgn-MX-YUC" and "sgn-eng-US" -- 
> will require a *completely different syntax* for tags that begin with 
> "sgn". That will be a knotty problem for those who choose to 
> participate in LTRU II.
> -- 
> Doug Ewell
> Fullerton, California, USA
> _______________________________________________
> Ietf-languages mailing list
> Ietf-languages at

More information about the Ietf-languages mailing list