[Ltru] Re: script tag for IPA

[Frank Ellermann <nobody at xyzzy.claranet.de>]

Martin Hosken wrote:
 
> The registrar of ISO 15924 has indicated that he has no
> intention of ever giving IPA a script code and that it is a
> variant of Latn. Perhaps you can get him to change his mind,
> but I doubt it.

For the Unicode Consortium (the ISO 15924/RA) I've no idea how
appeals work.  You probably need a rejected "Latp" registration
request, and arguments why that decision violates ISO 15924 (?)

> As Mark has stated

BTW, Mark certainly knows everything about any "Unicode appeal
process", if it exists.  Almost all contributors here are in
some way Unicode members, Randy (maybe) and I are exceptions.

If they don't know how to overrule their appointed registrar -
maybe he's simply right and IPA is no script or variant as
defined in the standard ?  I've just read the page about the
ISO 15924 history - the author of the rules would know what
they mean.

For RFC 4646bis I hope that we'll add a requirement that the
subtag reviewer MUST indicate that (s)he has read the RFC, and
is willing to implement it, and is appointed for a defined 
period of time, ideally by the IETF NomCom.  Or something in
that direction, making sure that the reviewer has still read
and write access on the review list - at the moment that isn't
relevant for the language list, but for other review lists it
can be an issue if they're silent for months (or even years).

> In discussions with the ISO 15924 registrar on this, he
> seemed open to the idea of extending the private use script
> code space.

Apparently there's also the possibility to reserve codes if
they are rejected by the JAC.  At the moment that won't help
you, because reserved codes can't be used in language tags -
same idea as EU wrt country codes.  "Officially", of course
you're always free to do what you want as long as you don't
claim to follow rules after you've clearly broken them... :-)

> In addition, I agree that since a script variant (in my 4
> character scheme) would always occur after a real script,
> there is no need to worry about codespace overlap.

It would break all 4646 parsers.  It would also break a main
feature of 4646, minus grandfathered cruft you can decompose
a tag into subtags, and you always know what each subtag is
if you keep the leading hyphen:  -1234 => must be a variant, 
-abcd => must be a script, -123 => must be an UN M.49 number,
-ab => must be a 3166 CC, -abc => extlang, abc => language,
ab => language, abcd => language (not yet possible), etc.

You can also recreate the original tag from its pieces to a
certain degree.  For extlang and variant you need the prefix
as noted in the registry, for generic variants you lose, but
maybe 4646bis kills generic variants by one s/SHOULD/MUST/.

Frank



_______________________________________________
Ltru mailing list
Ltru at ietf.org
https://www1.ietf.org/mailman/listinfo/ltru