- From: Albert Lunde <Albert-Lunde@nwu.edu>
- Date: Sat, 7 Dec 1996 23:11:09 -0600 (CST)
- To: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, www-international@w3.org
I think we should be very cautious about overloading the MIME charset or Accept-charset to mean more than just a character encoding. (I'll like to put in a word for reviving "Charset considered Harmful as an informational RFC" ;) In the context of HTML, the issues of what characters can be usefully used or rendered as gypths seem more closely allied to SGML than the MIME transport layer. (Though I don't know any SGML mechanism i to solve it all.) Remember that I could send all these funky characters using numeric character references, and a charset of US-ASCII. A true multi-lingual document is likely to have a mix of languages and scripts. That's why we have LANG attributes, to help the software out in choosing appropriate fonts, hyphnation methods, spelling dictionaries, etc. We haven't really addressed all the possible cases where one language uses multiple scripts/writing systems. (Hindi/Urdu _may_ be one such example I'm just guessing here.) I'd like to see more people implement the i18n spec before we fight about what comes next, however...
Received on Saturday, 7 December 1996 21:17:21 UTC