Problems deciding if az- should have multiple registrations...

Addison Phillips [wM] aphillips at
Mon Apr 14 16:40:15 CEST 2003

Peter wrote:
> > I imagine that there are systems with locales that look like:
> >
> > az.ISO8859_1 at latin
> > az-AZ.ISO8859_1 at latin
> Is this an actual example, or only hypothetical? My impression was that
> Mark was concerned about interoperation with .Net, and I've already
> commented in another message that .Net makes a three-way distinction, and
> it does not include anything comparable to your example here.
This is a hypothetical.

Actually, I note that IE makes a script distinction ["Azeri (Latin)" and
"Azeri (Cyrillic)"] in its languages dialog, but these both result in the
RFC3066 tag "az". The GUI implies a distinction that cannot be represented
in a standard tag.

My concern about this hypothetical is that it implies that if I *do* find a
system that makes the arbitrarily deeper (generative) distinction, I won't
be able to set the behavior on that system correctly. You can already see
this with the IE example.

I also know for a fact that many more common locales are fully generated. My
Solaris 2.8 system has:

th and th_TH
ko and ko_KR
ja and ja_JP
it and it_IT
sv and sv_SE

This is addition to the usual compement of en, es, and fr locales. But these
latter have actual language or locale differences to distinguish them. Of
the foregoing, only Korean is really a candidate for indeterminant behavior.
But there is a very real distinction between (for example) the sv and sv_SE
locales or the ja and ja_JP locales. Not the least of which is: if I want a
Unicode (UTF-8) locale, I need the long form and not the short one in both
cases, at least on my machine.

> > 1. *Are* we in agreement that RFC3066bis needs writing?
> My vote is yes.

Me too, but I'm waiting for the Official Poll from Harald before wading much
deeper into the discussion.

Best Regards,


Addison P. Phillips
Director, Globalization Architecture
webMethods, Inc.

+1 408.962.5487 (phone)  +1 408.210.3569 (mobile)
Internationalization is an architecture.
It is not a feature.

Chair, W3C-I18N-WG Web Services Task Force
To participate see

More information about the Ietf-languages mailing list