Article View: comp.std.internat
Article #325Re: An alternative I18N paradigm
From: Walt Daniels
Date: Tue, 29 Dec 1992 15:26
Date: Tue, 29 Dec 1992 15:26
28 lines
2301 bytes
2301 bytes
I agree that having **only** an international locale would be a
disservice to many people. However such a locale would perform a very
valuable function. Namely establishing a base from which many locales
can be derived by a small set of changes. For instance a Spanish
locale is almost identical to a French locale except for some changes
to sorting and the spelling of the names of the months. I am assuming
that an international locale would have ISO 10646 as its codeset. Both
the French and Spanish would probably agree with the Japanese sorting
rules for instance, as well as the various numeric sets of characters
that are used in a number of languages. A full locale description
appropriate to ISO 10646 will be very large. Most local locales will
be small changes to it.
There is another useful role that an international locale can play.
The current definitions of locale do not carefully distinguish between
two roles that locales play. One role is as a tag describing the
contents of a file. The other role is as a user preference of how s/he
wants to view the data. The codeset part of a locale is always a tag.
The LC_MESSAGE category is always a user preference. Some of the other
categories are sometimes a tag and sometimes a preference. These two
roles must be separated. One way of treating an international locale
is that it is always the stored form of the file. All other locales
should be interpretted as user preferences.
--
Walt Daniels (IBM)<dan@watson.ibm.com> 914 784-6736
(Internet)<dan@watson.ibm.com>
Technical Details
Message-ID:
<DAN.92Dec29102634@dan.watson.ibm.com>Path:
arc1-ca.pugleaf.net!sparky!uunet!newsgate.watson.ibm.com!yktnews!admin!yktnews!danReferences:
<24479@alice.att.com> <1hkff3EINN5uv@uni-erlangen.de> <1hncs1INN1qq@corax.udac.uu.se>Show Full Headers
Complete Article Headers
Newsgroups: comp.std.internat Path: arc1-ca.pugleaf.net!sparky!uunet!newsgate.watson.ibm.com!yktnews!admin!yktnews!dan From: dan@watson.ibm.com (Walt Daniels) Subject: Re: An alternative I18N paradigm Sender: news@watson.ibm.com (NNTP News Poster) Message-ID: <DAN.92Dec29102634@dan.watson.ibm.com> In-Reply-To: andersa@Riga.DoCS.UU.SE's message of 28 Dec 92 17:17:53 GMT Date: Tue, 29 Dec 1992 15:26:34 GMT Disclaimer: This posting represents the poster's views, not necessarily those of IBM References: <24479@alice.att.com> <1hkff3EINN5uv@uni-erlangen.de> <1hncs1INN1qq@corax.udac.uu.se> Nntp-Posting-Host: dan.watson.ibm.com Organization: IBM T.J. Watson Research Center, Hawthorne, New York Lines: 28