On Sat, Jan 12, 2002 at 03:06:09AM -0600, [EMAIL PROTECTED] wrote:
> That's sometimes true. On the other hand, if you accept workarounds,
> you may end up supporting them from now til forever. Whereas if you 
> refuse to implement workarounds, and force people to fix their systems,
> sometimes it will be easier to just fix the systems, and you see the
> buggy behavior disappear. 
> 
> Bad HTML that browsers bork on gets fixed. Bad HTML that browsers accept
> doesn't. Apache implemented the HTTP (2?) spec in a way that AOL didn't
> work with. Apache, being right, didn't back down, and AOL fixed their
> servers and browsers. GCC and Linux developers regularly have spats about
> illegal C in the kernel. The Linux developers regularly fix such C so the
> kernel compiles with later compilers.

I'm in complete agreement with you on all of this.  The problem is that
if we try to push the "real" fixes for the yen/backslash problem on
people, they simply won't take them.

If we tell CP932 users, "your 0x5C is a yen symbol, so translate it to a
Unicode yen symbol", what will they do?  Probably say "no, that'll break
almost all applications", just like our applications would break if we
changed ISO-8859-1 backslashes to Unicode yen symbols.

We could tell them, "most encodings are ASCII-compatible from 0x20 to
0x7E, but yours isn't; change 0x5C to a backslash for consistency",
we're making a pretty heavy request: to change an encoding that's been
in active use for a long time.

> In this case, the OGG spec may not make much difference (who uses Yen signs 
> and Backslashes in MP3 tags?). But given the choice between complex behavior
> and simpler, more correct behavior that won't matter much in the end, I'd go 
> with the simple correct behavior. Then it's going to be implemented more 
> consistently, and you never have to try and deprecate it once it's a moot 
> point.

For the ogg tags, I'll probably suggest that it convert CP932 0x5C to a
correct Unicode yen symbol, and end up with CP932 clients that can't
enter backslash symbols at all (instead of ones that can't enter yen
symbols at all.)  This is assuming I can get them to agree to
standardizing the nonstandard translation tables (for the purposes of
Ogg.)

-- 
Glenn Maynard
--
Linux-UTF8:   i18n of Linux on all levels
Archive:      http://mail.nl.linux.org/linux-utf8/

Reply via email to