Tagging transliterations from a specific script

[Doug Ewell]

Michael Everson <everson at evertype dot com> wrote:

>> I'm still not convinced that this "transliterated from script X"
>> scenario is so common, or ever going to be so common, that we need to
>> worry about generalizing the solution.
>
> Do you object to my having suggested that we do worry about it? I
> thought I was being prudent.

I don't object to that suggestion on its face.

There is a certain cost to creating an extension that might or might not
be justified, given the expected use.  Somebody has to write an RFC to
define the extension (going through the usual IETF process, involving
Internet-Drafts and Last Call and Waiting for AD Go-Ahead and AUTH48)
and either set up a mailing list for discussion or, as in the case of
the 'u' extension, designate an existing mailing list.  There is
certainly more infrastructure work needed to do this than to register
one or two variants.

Peter has already stated his opinion: "I'm not convinced we need an
extension for transliterations."  I think his alternative was to
deprecate 'alalc97' and break it into multiple variants: ALA-LC from
Arabic, from Cyrillic, etc. which to me seems unnecessary for most (or
even all) applications of 'alalc97'.  But I still believe that if the
only case for this is Tatar during a certain time frame -- and according
to Avram, even this case might point to a different solution -- the
overhead of an extension might be excessive, like hiring a contractor to
hang a picture on your wall.

--
Doug Ewell | Thornton, Colorado, USA | http://www.ewellic.org
RFC 5645, 4645, UTN #14 | ietf-languages @ is dot gd slash 2kf0s ­