Getting 3.0b3 out the door

[Dan Mosedale]

I've just written up a blog post announcing a decision that
thunderbird-drivers has made to ship 3.0b3 sooner rather than later:

<http://weblogs.mozillazine.org/dmose/archives/2009/06/getting_30b3_shipped.html>

So, what does the rest of 3.0b3 look like?

Of the regressions that have happened since beta 2, some subset are
important enough that we couldn't ship beta 3 without fixing them. I
just did a triage run through all the b3 bugs and slipped the ones that
we fairly clearly could ship without. This brought the number of
remaining blockers down to 20 from 35. I skimmed the blocking? list and
didn't see anything that was both sufficiently serious and sufficiently
reproducible that we'd hold b3 for it.

I expect that we'll be able to push another 5 or 6 bugs to b4 tomorrow,
and there are still a few more yet on the list that we can probably push
off if the fixes aren't sufficiently straightforward. This would leave
us at ~12 bugs (assuming, of course, that we don't discover new ones or
ones that we've forgotten). I only see one must-have bug on the list
that need string changes (bug 490097).

This leads me to suggest the following dates for b3, which we should
discuss at the Thunderbird meetings on Tuesday:

Slushy string freeze: Thurs, July 2 (very soon!)
Slushy code freeze: Tuesday, July 7
Firm string/code freeze: Tuesday, July 14
l10n-mozilla-1.9.1 freeze: Thursday, July 16
Target ship date: Tuesday, July 21

I think it's conceivable that we could be more aggressive than this and
get it out sooner, but I'd like to see what the bug list looks like
after we go over it at the driver meeting before deciding whether to
pursue that.

Dan