From owner-sc22wg5@dkuug.dk Fri May 23 06:50:54 2003 Received: (from majordom@localhost) by dkuug.dk (8.12.8p1/8.9.2) id h4N4osLH037728 for sc22wg5-domo; Fri, 23 May 2003 06:50:54 +0200 (CEST) (envelope-from owner-sc22wg5@dkuug.dk) X-Authentication-Warning: ptah.dkuug.dk: majordom set sender to owner-sc22wg5@dkuug.dk using -f Received: from mail500.nifty.com (mail500.nifty.com [202.248.37.208]) by dkuug.dk (8.12.8p1/8.9.2) with ESMTP id h4N4oHEc037723 for ; Fri, 23 May 2003 06:50:49 +0200 (CEST) (envelope-from takata@edogawa-u.ac.jp) Received: from takata-n8.edogawa-u.ac.jp (airh032124124.mobile.ppp.infoweb.ne.jp [219.104.146.134])by mail500.nifty.com with ESMTP id h4N4otrx015861 for ; Fri, 23 May 2003 13:50:57 +0900 Message-Id: <5.1.1.8.2.20030523093136.052aa3f0@mail.edogawa-u.ac.jp> X-Sender: gaf01617@mail.edogawa-u.ac.jp X-Mailer: QUALCOMM Windows Eudora Version 5.1.1-J Date: Fri, 23 May 2003 13:50:54 +0900 To: sc22wg5@dkuug.dk From: TAKATA Masayuki Subject: Re: (SC22WG5.2701) latex question In-Reply-To: <200304090000.h3900qGu020497@dkuug.dk> References: <200304082249.h38MnaUp019765@dkuug.dk> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-2022-jp" Content-Transfer-Encoding: 7bit Sender: owner-sc22wg5@dkuug.dk Precedence: bulk Keith and others: I ignored this question for a while because I was too busy and there was no reason that I had to quickly respond, sorry. At 03/04/08 16:59 -0700, Keith Bierman wrote: >Another somewhat interesting question is whether the characters >you have are katakana, hiragana or "kanji". Japanese is written >with all three, two are "sylabets" and the third (kanji) is >idiograms. > >I would have guessed that the sample you have would be katakana. The "without her nothing is possible" example is written in the usual way, i.e. mixture of kanji and hiragana (no katakana, in this particular example) with an equivalent of a full stop. >Actually, I would have thought that Fortran processors with Nihongo >support would have selected katakana characters to support first. >So, if I'm right, I would *hope* that what you have is katakana ;> As far as I know, there is no Nihongo Fortran 90/95 at the moment. People went back to F77's way, i.e. sequence of bytes, using EUC-JP or Shift_JIS coding, which allows to mix 1- and 2-byte characters together in a 1-byte character type without escape sequences. In old days, when kanji was too heavy, katakana-only implementations were acceptable for many people. But it is not, now. Japanese children are first taught hiragana, then katakana. They have about 50 characters each. Both could be used to transcribe any pronounciation in Japanese. There are about 2000 daily-used kanji characters. Only very young children's books are written without using kanji, which is essential for distinguishing the too-many homophones in the language. On the other hand, you can not write Japanese without using hiragana and katakana. In this sense, there is only one set of characters for Japanese language, which includes all three. Hiragana and katakana are always treated separately, because they were invented in Japan, borrowing their form from kanji characters, which are basically Chinese characters. >It would be possible (perhaps even interesting) to have a Fortran >implementation which used the KIND facility to provide all three >sets of characters. However, I have no idea if anyone has done so. Not yet. But now UCS (ISO/IEC 10646) is dominating the world and we have selected_char_kind. Things will change soon, I believe. Regards, Makki -- (Mr) Takata, Masayuki: Associate Professor Edogawa University, Nagareyama, Chiba 270-0198 Japan phone:+81-4-7152-0661ext546 fax:+81-4-7154-2490 http://www.edogawa-u.ac.jp/~takata/