[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: About Bug#250789



* Joey Hess [2004-06-11 14:44:16-0400]
> (Please use the bts, I'm not the only one working on this stuff, and can
> easily miss mails. I'm also the worst possible person of base-config's
> several maintainers to talk about this stuff to, since I don't
> understand encodings, fonts, or localisation at all, really.)

Apologizes to increase your work load.  From now on, I'll be more 
careful on this issue.

> Recai Oktas wrote:
> > As you may recall, despite the fact that it setups console correctly on 
> > startup, termwrap leaves the console in a bad state while exiting, and 
> > requires the user to reboot or issue the '/etc/init.d/console-screen.sh 
> > restart' command.  I can't figure out why the exit code exists (that is, 
> > the block at the end which starts as 'case $ENCODING in').  This code 
> > block should be simply removed, if there is no spesific reason behind 
> > it.
> 
> I'm currently trying to get a usable base-config into testing to use for
> the tc1 release, and since recent, insufficiently tested, font changes
> have led to the current mess, the last thing I want to do is make any
> more changes in this area right now. I'll look at it again after tc1 is
> released. I suspect that the code was added for a reason though.

You misunderstood my mail.  I don't want a change in tc1 (may be except 
the new languagechooser), just want to call your attention to the 
problem.  I am tired of such i18n idiosyncracies, particularly for 
Turkish.  I haven't ever seen an _almost_ out-of-the-box localized Linux 
distribution, and want to make sure that this time, we don't experience 
the same problems again for Debian.  These feelings are based on some 
painful experiences which a lot of English spoken people has not been 
aware of.  I don't want to encounter with the same l10n questions 
directed in debian-user-turkish again and again.  (Sorry, if you find 
these statements insulting.  This is not my intention.)

> > After looking the screenshots, another change which seems to be 
> > mandatory is that, we should change the 'iso02.f16' to 'lat2-sun16' in 
> > the following line for iso02 (FYI, I've reverted back my changes to 
> > languagechooser for iso02):
> > 
> > ISO-8859-2)
> >         # Load ISO-8859-2 charset mapping into console
> >         try_load_charset "$ENCODING" iso02 iso02.f16
> 
> There is no chance of making languagechoser changes for tc1. If
> necessary we can do this kind of change in a tc2. If this is a release
> critical issue for rc1, you need to explain *why* it would be worth the
> bother of throwing away tc1 and spending weeks retesting with a tc2;
> saying it's "mandatory" is not good enough.

If you mean the 's/iso02.f16/lat2-sun16/' change, isn't this clear just 
for the same reasons of 's/iso01.f16/lat0-sun16/' change for English in 
the latest base-config?  I supplied some screenshots for a sample 
language (Polish) which IMO, clearly explains why such change would be 
worth and _mandatory_.  I try to do my best and perform some tests 
before suggesting a change, though I'm not a console guru.  Anyway, 
thanks for your time and efforts.

Best regards,

-- 
roktas

Attachment: signature.asc
Description: Digital signature


Reply to: