[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Merge tracking - a proposal for impl. AS

From: <kfogel_at_collab.net>
Date: 2004-09-08 17:45:45 CEST

Jacob Refstrup <jacob.refstrup@hp.com> writes:
> This is a proposal for how to implement merge tracking
> with Ancestor Sets (AS) with improved "blame" capability
> (being able to blame who merged and who originated the
> code).
>
> [...]

Thanks for the thoughtful post, Jacob.

I can't give you much of a technical response right now, as I -- and
at least a few other developers I know -- are deliberately avoiding
raising the merge tracking topic at the moment. We're trying to get
SVN 1.1.0 out the door, and at the same time address a few other,
smaller-but-still-important issues, in particular reserved checkouts.

Obviously, on a volunteer project, I can't presume to speak for all
the developers. But my guess is that others have similar priorities
right now. However, I've saved your proposal. Even though it's only
a high level overview, it seemed well-thought-out and will help when
we *do* get to merge tracking.

Merge tracking is something many of us are thinking about in the
background. I think that when we finally do raise the topic, we'll
probably start with a goal-oriented approach: define exactly what is
meant by "merge tracking", specify exactly what problems it will
solve. Only then can we talk about implementation. Your proposal
mixes these two stages a bit. Because merge tracking is such a broad
topic and means different (though overlapping) things to different
people, there is a high likelihood of misunderstanding and people
talking at cross-purposes, unless we first clearly define the problem
to be solved.

However, I'm *not* proposing that we start that process now :-). We
need to get 1.1.0 out the door, fix a number of bugs, finish reserved
checkouts (a.k.a. "locking"), and take care of a few other things
first, IMHO. Merge tracking's day will come, don't worry. If it's
any comfort, some of us are even starting to budget time for it in the
medium- to long-term future. But the stuff that needs to be done in
the short-term is extremely compelling, and can't be delayed.

Best,
-Karl Fogel

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Wed Sep 8 19:29:33 2004

This is an archived mail posted to the Subversion Dev mailing list.