
Mailing List Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [tlug] Re: Piping stderr?
- Date: 01 Jul 2002 18:55:44 +0900
- From: "Stephen J. Turnbull" <stephen@example.com>
- Subject: Re: [tlug] Re: Piping stderr?
- References: <3D109EC0.4070703@example.com><87n0trz38q.fsf@example.com><s3t3cvj6oad.fsf@example.com><87r8j1w8ol.fsf@example.com><877kktb463.wl@example.com><87vg895g7k.wl@example.com><871yaxnjei.fsf@example.com><87adpkj70k.wl@example.com><87u1nrao03.fsf@example.com><87znxjhlw4.wl@example.com><87ptyf91b3.fsf@example.com><87wusnhfg8.wl@example.com><876607wlpj.fsf@example.com><87eleuz973.wl@example.com><87d6ues4tn.fsf@example.com><87n0thf6m2.wl@example.com><873cv9owxa.fsf@example.com><87lm91f11n.wl@example.com><87u1nooofx.fsf@example.com><878z50fe79.wl@example.com><877kkkm6dx.fsf@example.com><871yasqcb5.wl@example.com><87y9d0kmpm.fsf@example.com><87znxgos8v.wl@example.com><87k7ojlx87.fsf@example.com><87fzz7lv2i.fsf@example.com><87lm8zu4il.wl@example.com><874rfnlj62.fsf@example.com><87u1nkcbk4.wl@example.com><87hejjhnry.fsf@example.com><87elenrg5o.wl@example.com>
- Organization: The XEmacs Project
- User-agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.4 (Informed Management (RC2))
>>>>> "Jiro" == Jiro SEKIBA <jir@example.com> writes:
Jiro> On this list, there are a lot of people who are not
Jiro> interested in I18N and don't know well about it.
Oh, actually I think almost all of them are interested. They don't
necessarily know all the pages in section 3 of the manual, of course.
Jiro> And actually I couldn't help doubting your mb*/wc* knowledge
Jiro> when you wrote "libcsi" idea.
Sorry. I simply assumed that "CSI" actually provided support for
programmers trying to write internationalized programs. This would
require a lot more functionality than is available in glibc.
Jiro> If you use on the SJIS locale, yes, error will occur.
Jiro> That's helpless ith CSI. But you can avoid by stripping that
Jiro> locale.
So much for code set independence. Now the poor programmer has to
know in advance what locales are "dangerous."
You see, I'm not talking about _me_, I know how I'm going to handle
this. I'm talking about somebody who's trying to write her first I18N
program. Pity that the poor lady is a Turk working in Germany, and
she never heard of Japan until the World Cup. Shift JIS? Nobody
could possibly really use something that broken, could they? How
could a sane, non-Japanese programmer imagine that?
And why should she have to learn?
Jiro> You prefer UTF-8 hard-coding, even it has incompatibly with
Jiro> legacy codesets.
No, I prefer _real_ internationalization, where the programmer doesn't
need to know about the locales where her program is going to be used
to write safe, correct programs. This may impose a certain amount of
restriction on the programmers or users, but that can eventually be
generalized in a safe way if needed. Such need will rarely be urgent
for shells and scripting languages (remember them? that's what we're
talking about).
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
My nostalgia for Icon makes me forget about any of the bad things. I don't
have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py
Home |
Main Index |
Thread Index