From keld@dkuug.dk Tue Jan 21 22:08:56 1992 Received: by dkuug.dk (5.64+/8+bit/IDA-1.2.8) id AA09077; Tue, 21 Jan 92 22:08:56 +0100 Date: Tue, 21 Jan 92 22:08:56 +0100 From: Keld J|rn Simonsen Message-Id: <9201212108.AA09077@dkuug.dk> To: Teruhiko.Kurosaka%Eng@sun.uucp, i18n@dkuug.dk, wg14@dkuug.dk Subject: Re: (XoJIG 440) Re: (XoJIG 438) Re: Support for symbolic character names X-Charset: ASCII X-Char-Esc: 29 > Walt Daniels writes: > |So 90% of the "symbolic" names are just codepoint numbers. Why not just > |use the numbers for the rest of them and eliminate the European bias! > I agree with Walt. I am not concerned with the fact that the > characters are named in English. My complain is that the way East > Asian characters are "named" is not useful. I can easily guess what > ":" would be called but I have no idea what symbolic name of "Kuro", the first > character of my family name, is. I'd have to look it up in the JIS > standard book anyway. So why don't we just use the codepont itself? > -kuro (JISX0208_39x75-:) Well they use the codepoint for Japanese, Chinese and Korean chinese characters, so that should be OK with you? If other scripts can make another naming for these scripts, that the users like better, would that be a problem for you? I think that not only the Latin script can be handled quite conveniently by other means than just specifying the codepoint in some special character set. By the way, latin script is not just used in Europe, but also in North and South America, Australia, Africa and some places in Asia. And many other scripts that Latin can be given good mnemonics. So to say that using non-numbered mnemonics is Euro-centric is way off the target, IMHO. Using numbers from 10646 would be a step forward, IMHO. Still it would be like going from random addressing, to standard addressing, like having a variable always at the same address in a programmming language, but still needing to write it as a number. Going to more mnemonic namings would be like introducing variable names in a programming language, and having the same mnemonics meaning the same characters, would be like introducing standard headers and variables and functions. I think it is a needed feature for C to be able to handle characters in a more portable way that can be done currently. I do not think WG14 should do the naming of a lot of international characters, as this is not really within the current JTC1 project, and more closely related to the WG20 group, but I think it would be good for WG14 and within the scope of the JTC1 project to define how C should use such a naming from WG20 - or the WG15 concepts developed for specifying strings in a character set independent way - and the corresponding mechanisms to specify a character set. POSIX is going to happen, and it would be good if we could employ the information already available. C already has been a pioneer in giving good support for extended character sets and this is one of the issues we should address in enhancing the features of C to handle extended characters. Keld