Date: Sun, 16 May 1999 12:12:03 +0100 From: Nik Clayton <nik@nothing-going-on.demon.co.uk> To: Wolfram Schneider <wosch@panke.de.freebsd.org> Cc: "Andrey A. Chernov" <ache@nagual.pp.ru>, Kazuo Horikawa <horikawa@jp.freebsd.org>, nik@nothing-going-on.demon.co.uk, ru-freebsd-doc@freebsd.ru, doc@FreeBSD.ORG, freebsd-translate@ngo.org.uk, hanai@jp.freebsd.org Subject: Re: [RU-DOC] FDP Directory Reorganization Message-ID: <19990516121203.C62097@catkin.nothing-going-on.org> In-Reply-To: <19990516000424.48809@panke.de.freebsd.org>; from Wolfram Schneider on Sun, May 16, 1999 at 12:04:24AM %2B0200 References: <19990514204302.B43389@catkin.nothing-going-on.org> <19990515222516E.k-horik@yk.rim.or.jp> <19990515174717.A656@nagual.pp.ru> <19990516000424.48809@panke.de.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
I'll address a couple of points here;
On Sun, May 16, 1999 at 12:04:24AM +0200, Wolfram Schneider wrote:
> On 1999-05-15 17:47:17 +0400, Andrey A. Chernov wrote:
> > But I don't think it really happens for documentation, so it seems that
> > <lang>/<encoding> scheme is right. Maybe even <lang>.<encoding> if people
> > choose only one encoding for documentation, but *not* simple <lang>
> > (assuming some unknown default encoding) which means that program can't
> > sense it.
>
> I prefer <lang>.<encoding>
>
> or just <lang> for translations in iso8859-1 character set.
People seem to be agreeing that <lang>.<encoding> is the way to go, and
I'm happy to follow that, particularly based on the contents of
/usr/share/locale.
There will *not* be a default encoding. We've made that mistake once.
iso8859-1 should not be a special case just because it's the default
English encoding.
Special cases are bad, they lead to work arounds for that special case
that inevitable need to be changed when the thing you're special casing
for changes.
Suppose iso8859-1 was the default, and all the Makefiles, CGI scripts,
and so on default to it. And then we switch to Unicode a few years down
the line. We'd have to go through however many Makefiles, CGI scripts, and
whatever have grown up since then to teach them about the new default.
Far better to explicitly state the language and the encoding up front. Even
for languages where there is only (currently) one encoding used.
The only unaddressed question I think is, do we need the country information
in the name as well? I don't think so, as en_GB.ISO_8859-1 and
en_US.ISO_8859-1 will be the same text.
However, does this assumption hold for all the other languages that we'll
be dealing with? What's prompted all this is that some of the Chinese
translators want to make documentation available in both EUC and BIG5
encoding.
The /usr/share/locale directory has zh_CN.EUC and zh_TW.BIG5. Notice that
the country selector is different. Do we need to follow this, or can we
collapse it to zh.EUC and zh.BIG5?
If we can't collapse it then, for consistency, all the other directory
names will need to include the country selector as well.
N
--
There's some milk in the fridge about to go off. . . and there it goes.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-doc" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990516121203.C62097>