From keld@dkuug.dk Fri Sep 4 23:17:39 1992 Received: by dkuug.dk (5.64+/8+bit/IDA-1.2.8) id AA23480; Fri, 4 Sep 92 23:17:39 +0200 Date: Fri, 4 Sep 92 23:17:39 +0200 From: Keld J|rn Simonsen Message-Id: <9209042117.AA23480@dkuug.dk> To: i18n@dkuug.dk Subject: Re: request for feedback on character set identification proposal Cc: iso10646@jhuvm.bitnet X-Charset: ASCII X-Char-Esc: 29 Some comments on character set naming. I believe the paper from Ju:rgen Bettels is a good starting point for a very necessesary task. I have done quite some work on this also, viz. the JTC1/SC2/WG2 and JTC1/SC22/WG20 papers I have delivered and the RFC 1345 I was the author of. The scope is about right, as far as I can tell, I have tried to cover the same (almost) in the above mentioned papers. Some additional remarks: I believe SC22 will need for programming language support character set names like "ASCII", "ISO8859-1" "IBM850" "UCS2" - that is: tokens that could be used in programming languages as a way to charactarize strings etc. Several names (aliases) would be convenient for this, eg "ISO8859-1", "ISO-IR-100", "IBM819" which could all mean the same encoding. Is that intended to be covered by the WG3 project? Also I believe that names covering more than one registration in the ISO 2375 registry will be needed. Eg "ISO8859-1" covers registration 6 and 100 and a C0 and C1 control set, in total 4 2375 registered sets. For convenience it should be possible to refer this under one name. The same goes for eg. the different Japanese encodings, where explicit shifts between character sets (SO/SI) may occur. In SC22, POSIX is doing specifications of character sets in "charmaps" and a registry of charmaps are on the agenda of SC22/WG15 and SC22/WG20. Charmap format is specified in ISO/IEC DIS 9945-2:1992 This should be consistent with the new work from WG3. It would be very convenient if the new registry could be provided on-line, eg as a Directory (X.500) service, and via FTAM and FTP. Keld Simonsen