Following the thread on dev@ and the thread on users@ asking our users
what we should do with 1.5, I propose we branch the 1.5.x release
branch at the end of this month.
For those who didn't follow the discussions, I just read through the
users@ thread, and this is what I think sums it up:
- As can be expected, there are still a lot of people who want merge
- However, relatively few said that 1.5.0 should absolutely feature
merge tracking, no compromises. And those who did say that basically
just said they would skip 1.5 and bump to 1.6 when it comes out.
- A good portion of replies express interest in getting the features
and bugfixes we have right now out immediately, rather than push it
all off until we get merge tracking done.
- A few people remarked that getting the features we have now "out
there" is also good for merge tracking stabilization, as we will have
time to address bugs in these features before we release the merge
tracking behemoth, and could therefore focus more on merge tracking
Personally speaking, I like the stabilization argument (MT is a big
enough code change as it is, let's try to get other code changes out
of the way), as well as catering to those who need stuff other than
merge tracking as soon as possible. I believe that, even if that means
we get 1.6 in 9 months rather than 1.5 in 6 months, it will be worth
it in terms of not having to manage feedback on MT and feedback on
many unrelated features.
So, with a release manager cap on, I propose to branch at the end of
March. That gives everyone a little over two weeks to wrap up features
before we branch. After that, tentative timeline would be a couple of
days to merge the late-coming patches, then an RC1 release. After
that, regular soak cycle until we get it out the door.
In that time, the CHANGES file needs to be updated. I'm willing to
tackle that, if the plan sounds good to you folks.
To unsubscribe, e-mail: email@example.com
For additional commands, e-mail: firstname.lastname@example.org
Received on Mon Mar 12 19:43:08 2007