On Mon, Sep 12, 2005 at 02:50:48AM -0400, Nathan Walp wrote:
> On Mon, Sep 12, 2005 at 12:12:05AM -0500, Mark Doliner wrote:
> > Hi kids! We've been making a lot of progress recently towards making CVS HEAD
> > more stable and usable (nice job everyone), and I think it's time to set some
> > target dates for Gaim 2.0.0.
> > October 31st - String freeze and feature freeze. By this date all major
> > changes should be done. The next month will be bug fixing, making sure
> > everything that works in Gaim 1.5.0 still works in 2.0.0, and maybe even some
> > documentation (organizing the ChangeLog, for example).
> The point Richard made about the order of string- and feature- freezing
> with regards to the beta is valid.
> Actually, I think we need to allow for a feature and string un-freeze
> post-beta. The reason: prefs. There are a handfull of prefs that are
> going to be put back by popular demand. No, they're not going to get
> used to it. Yes, there's going to be a line of people outside my cube
> demanding I fix it all. No, I don't want that.
Correct. We removed a TON more preferences as we moved towards our
current state, with the idea that doing so would give us a feeling
for which ones are actually needed and which ones we can do without.
As we release betas to wider audiences, I think that it is almost
certain that some preferences will have to be re-introduced. It also
seems very likely that we'll get feed back along the lines of "This
doesn't make sense" and might need to change a string to clarify it.
I'd like to feature freeze as suggested, but string freeze only after
we start to get feedback.
However, I also want to give the translators more than a 5 days to do
their work, I was thinking on the order of 10 days to 2 weeks.
> in reality, we could probably skip the beta entirely, if we can set up
> some nightly builds. In reality, I spend most of my time actually using
> Gaim at work, on win32, in a situation where i'd rather not have a copy
> of the code around. If there were win32 nightlies, I'd use it there,
> and it would get tested more. These can be "private" for now, but at
> some point when we think we've got the majority of the major kinks
> worked out, we can "throw the switch" and announce these nightlies to
> a slightly larger audience.
Warren has requested that we send him a beta. we also had a user pop
up here on gaim-devel, late october when we are *ready* for the type
of feedback he tried to give, I'd like to take him up on that offer.
Other than that, I think our own daily use, perhaps coupled with the
daily use of those we know well, should be fairly good testing.