On Fri, Jun 01, 2001 at 03:31:53AM +0200, Marco d'Itri wrote: > On Jun 01, Cesar Eduardo Barros <[EMAIL PROTECTED]> wrote: > > >Ask the IETF. They seem to like UTF8 a lot. > Because it's ASCII-compatible. This is not relevant. > > >Ask Linus too. The UTF8 support is in the kernel since, what, 2.0.x? > Because it's ASCII-compatibile. This is not relevant.
This IS relevant. We just CANNOT forget ASCII, as much as we would like to. And, when we already want something (mostly) upwards compatible with ASCII _and_ being the one and only unified encoding, why not to choose UTF-8, since it already IS a standard? > > UTF-8 maybe be useful for things like debconf templates as long as it's > able to recode on the fly to the $LC_CTYPE encoding, but don't you dare > fucking with the locales for languages and countries you don't know > about. All the locales should be in UTF-8. Point. <semirant> Otherwise you are just contributing to the great mess. Recently I wanted to adjust a computer for one particular user, who speaks Lithuanian, Slovak, Russian (in this order), but only a bit of english So my naive[*] guess wat to try something like export LANGUAGE=lt:sk:cs:ru Guess what? IT CAN NEVER WORK! lt, sk and ru locales use three different encodings. [*] I'd like to type naive properly, with i-diaeresis, but I just cannot, since it is not in ISO-8859-2 encoding my console is switched to I have a user who speaks slovak and italian... again no luck, she even cannot write mails properly with diacritics in both languages, since that requires to change the console font, chanke the keymap, AND TO EDIT /etc/Muttrc file. > Making sure applications can deal with UTF-8 is ok, things like recoding > the documentation are not. I stumbled across a package, uhm, more of them actually, one had documentation in Polish, one in Russian... now if I had not known before that Polish uses (fortunately) the same ISO-8859-2 encoding as Slovak, and that the Russian documentation is most[**] likely written in KOI8-R (and even more fortunately I've already set up russian fonts), I would have had a difficult time reading the documentation. [**] It happens to me more often than I would like: I download a program, with documentation in (guess) Russian. Switch my console to KOI-8, type less... uhm... gibberish. Use konwert to convert from CP1251 to koi8-r, pipe to less... still gibberish... try ISO-8859-5, still nothing.. Some head scratches, ls /usr/share/konwert/filters, pondering about what could be other russian encodings... trying CP866, still gibberish... giving up in disgust, maybe it was japanese or chinese, god knows Several days later found out it was in ECMA-cyrillic. > > Most people (with the possible exception of part of the CJK community) > do not want to use unicode yet, deal with it. > Most people do not care about the encoding, they just want to USE the computer productively. THEY CANNOT because of the mess in encodings. I resigned from diacritics in my name for the purpose of communicating (yes, there IS diacritics in my name). How would you feel if you HAD TO write your name Marco Ditri, just because there happened to be no common encoding where the apostrophe has the same numerical representation? I recently needed to set up database of some people... guess what, most of them had A LOT of Slovak diacritics in names, but some of them were French with french diacritics... UTF-8 was about the only way, even if it was a bit painful (and the later need to add cyrillic names went perfectly without glitches, thanks to UTF-8) Americans have it easy. People using languages falling into ISO Latin group category have it a bit difficult, but acceptable... as long as they do not want to communicate with people from some OTHER ISO Lating groups. Then is turns to be hell. </semirant> -- ----------------------------------------------------------- | Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ | | __..--^^^--..__ garabik @ melkor.dnp.fmph.uniba.sk | ----------------------------------------------------------- Antivirus alert: file .signature infected by signature virus. Hi! I'm a signature virus! Copy me into your signature file to help me spread!
