Bug#99324: Default charset should be UTF-8
I used an advanced feature of my MUA to respond to more posts at once..
hope nodoby minds
On Sun, Jun 03, 2001 at 10:27:22PM +0200, Marco d'Itri wrote:
> On Jun 01, Radovan Garabik <garabik@melkor.dnp.fmph.uniba.sk> wrote:
>
> >> because they are broken or because the charset is unlabeled.
> >so we first make them work with ISO-8859-*, then work on making applications
> >work with UTF-8, then work on making those terminals display UTF-8?
> >I can see a shortcut here...
> No, we first make sure our applications can all work correctly with
> UTF-8 too, then the communities who cares about it will change their
> default locales from the national encoding to UTF-8.
>
"we first make sure our applications can all work correctly with
8 bit characters, then we use 8 bit encodings"
hands up who thinks it should have been like this when the transition
from 7-bit ascii to 8-bit national codes took place.
we would be waiting still today...
(and the 8 bit national codes ARE an improvement over 7-bit ascii,
even if not ideal)
or even better:
"we first make sure our applications are y2k compliant, then
we switch to year 2000"
in other words, you can hide away from the problem, until it
hits you hard - and with the explosion of internet and international
communication, it is already starting to hit hard.
> >All the i18n stuff in glibc is a bit flawed... it assumes you
> >NEVER want to change the default locale while the program is running,
> >and it assumes everybody has correct terminal.
> So, are you proposing we switch our default encoding to one which is not
> well supported by libc?
I did not tell that. I told glibc locale support is flawed, designed by
people who did not see the need for _multilanguage_, as opposed to
_non-english language_ support, irrelevant of encoding.
(glibc does support utf-8 somewhat, it just does not like
mixing locales)
>
> >locales are in UTF-8 unconditionaly
> If gettext can recode it's not relevant.
It is, since you have a mess of encodings otherwise
ru_RU, ru_RU.KOI8-R, ru_RU.ISO-8859-5, does not this
look a bit "schifoso"? It really caused some nausea to me
when I needed to set a russian locale
(two debian packages, both with russian locales, both in
different encodings, plus glibc bug preventing to use any of them)
And you might want to include a name, a different character or
whatever in a gettextized string, which just does not
have a representation in current locale
(make russian gettextised string out of this:
"We saw Toto in his last film..."
(I'd like to put accento grave where it belongs, but it just is not
in my 8-bit national charset...)
)
On Sun, Jun 03, 2001 at 10:39:30PM +0200, Marco d'Itri wrote:
> On Jun 01, Roland Mas <lolando@debian.org> wrote:
>
> >now we miss the euro sign too. The transition to Latin-9
> >(ISO-8859-15, with the euro and the missing characters) is *already*
> >causing *major* nerve breakage amongst people.
> Right. It's already bad when broken applications don't work well with an
> encoding different from the one used by most people in may country,
> guess what would happen with a multibyte encoding.
> (I'd have switched to latin-9 if there would not be so many broken
> applications.)
so you propose to fix applications to work with different 8-bit codes
first, then switch to those codes (latin-9), then fix applications to
work with utf8, then switch to utf8? Again, I can see a shortcut here...
>
> > And how exactly do you expect to get the developers to care about
> >Unicode if no-one is putting some pressure on them? They already
> This is a free software project. You don't put pressure on other people,
there has to be some pressure to keep things standartised to prevent anarchy.
Why is there debian policy if not?
> or these people will tell you to fuck off.
great, my next package will have description in EBDIC
and nobody try to put any pressure at me to change it to something else
> If you want something to be done, you do it.
>
What is the purpose of this flamewar? :-)
Trying to do something.
On Sun, Jun 03, 2001 at 10:43:10PM +0200, Marco d'Itri wrote:
> On Jun 03, Radovan Garabik <garabik@melkor.dnp.fmph.uniba.sk> wrote:
>
> >2) Require using utf8 in debian control files (debian/changelog, debian/control,
> > Packages). This is not such a great change as it seems, since it will mean
> > only replacement of a few characters in a few packages (currently using
> > iso-8895-1). Woody+1, maybe woody too.
> >3) Require using utf8 in documentation. Definitely should go into woody+1.
> No fucking way. First fix the tools (i.e. you complete most of the other
> points), then we'll talk about changing the encoding.
then, why are you using latin-1 console?
there are many programs NOT dealing with 8-bit codes properly
(chfn, as I rather painfully found out)
I suggest you fix those programs and only then switch to 8-bit console
On Sun, Jun 03, 2001 at 10:29:35PM +0200, Marco d'Itri wrote:
> On Jun 01, Arto Jantunen <viiru@debian.org> wrote:
>
> >> I'm not arguing about this. I agree that in a perfect world everybody
> >> would be using unicode, encoded as UTF-8 or UTF-16. My point is that
> >> there is too much broken software to switch now to UTF-8.
> >What we are talking about here is a plan to work towards fixing all the
> >broken software and eventually moving everything to UTF-8. Not pulling
> >the plug on everyone right this moment.
> Sounds good. Then please fix all the broken software and after this is
> done we'll talk again about switching to unicode.
s/all the broken software/enough software to be able to work with utf-8/
and we are getting closer and closer there, and since we live on the edge
(distributions/unstable, right?) the time has probably come to do something
with this
--
-----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__ garabik @ melkor.dnp.fmph.uniba.sk |
-----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!
Reply to: