On Sat, Feb 16, 2002 at 09:30:33PM -0700, Joel Baker wrote: > On Sat, Feb 16, 2002 at 08:19:03PM -0800, Thomas Bushnell, BSG wrote: > > Joel Baker <lucifer@lightbearer.com> writes: > > > Claims that this is impossible don't hold much weight with me, when people > > > all around us *are doing it*. It is clearly, and observably, possible. Say > > > that it's more work than you think is worthwhile, if that's what you mean, > > > but don't claim it's impossible. > > Oh, I do think it's possible. I'm not trying to argue against your > > suggestions, I'm just saying that they don't account for changes that > > are actually big changes, by assuming that everything will fit. > > I think you might want to (if you want to revise your proposal) > > include more discussion about how to handle particular types of large > > change. > This is what I tried to address by giving an example of how things *can* > fit. Nobody has, to date, come up with something that couldn't fit within > a 4-month release window (IE, the size necessary for 3 releases a year). > I will accept any example people choose to give, and do my best to show > how to make it reasonable. If I can't, well then, you'll have a point, > and the proposal would need modification. Unicode support at all levels. I would very much like to see good Unicode support in woody+1, but I don't see how this would be possible with a 4-month release cycle. Granted, there are some things that can be done in stages, but much of that has already been done: woody has tools to Unicode-enable the console, x-terminals that support utf8 when passed the right switch, and binary-incompatible, unicode versions of various console rendering libs. The problem comes when upstream (or Debian) wants the new version of these libs to ship with UTF8 enabled by /default/; lots of apps would need recompiling, and worse, extensive testing to shake out single-byte assumptions. The alternative to doing this in one release cycle is to drag out the lifespan of Unicode-specific libs for far long. Granted, if we could get the Debian release cycle down to 4 months, it would be widely acclaimed as a miracle, and at that point spreading Unicode out over 2 releases wouldn't be so bad; but barring miracles, how would you handle this issue? Steve Langasek postmodern programmer
Attachment:
pgpk_3CVC2qLd.pgp
Description: PGP signature