Arjan Oosting wrote:
Op di, 09-08-2005 te 16:23 -0700, schreef Erik Steffl:
BTW I think it makes a lot of sense to use experimental for most of
the initial testing
That is wat unstable is for.
well, what is experimental for then? And what would you offer to
desktop users?
and only release to unstable when it looks like the
package is a release candidate.
That is testing, it should be possible to release testing anytime as the
next stable.
You're complaining that testing is too old, and unstable unstable, but
no, here's what I was saying:
unstable: mostly OK, think it would be useful if people took it a bit
more seriously, i.e. not breaking it for extended periods of time
without a good reason and if they break it at least explain what's going
on (see jackd bug #318098)
testing: bugs take forever to get fixed so overall it's worse than
stable. Security was a huge problem, might be getting better (saw the
link about the effort but don't know what the status is)
stable: too old, most of the time, not suitable for desktop (in general)
it takes some time to fix bugs and make sure a package is in a
releasable state, so yeah the software in testing might be a little bit
old, but that can't be helped. Moving development from unstable to
testing won't fix this, then unstable gets 'old' and we are back were we
started.
what are you talking about? Nobody was suggesting anything like that.
I was merely saying that for lot of desktop users unstable is the only
sort of acceptable option so people might want to take it seriously,
like not breaking it for extended periods of time with the attitude
"it's just unstable". If developers needs to try something crazy then
there's experimental, as far as I can tell.
erik
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
