On 12/11/2010 10:01 PM, Scott Merrill wrote:
Absent a strong identification of specific problems relating to our
use of SVN, I find this line of inquiry to be a bit strange. It'd be
like saying, with a straight face, "Let's consider rewriting Habari in
Ruby." Sure, it could be done, but we have a substantial investment in
what we have in place now, so what's the specific thing we're hoping
to improve by changing?
Sorry that we accidentally skipped over that step.
There are a number of benefits, but the primary one for me comes
anecdotally, wherein contributions to the core code are more easily
submitted and included.
For example, some maintainer for another project would like to throw a
blog online. When questioning friends, he is presented with Habari,
which fits his needs. He's not really interested in joining the Habari
development team or learning the Habari ticket/diff ingestion workflow,
but has several useful changes that could be submitted.
If Habari was in git, the tools for suggesting and reviewing patches are
built into the SCM. It may be easier for us to review changes also,
since it's apparently easier with git to get his whole version to test
without having to apply patches to specific svn revisions.
Git also offers the traditional advantages of distributed SCM, in that I
could work on a feature to completion locally with the benefit of local
SCM, and when the feature was complete enough, could push it to the
central repo where everyone else could use it. The advantage is that
features added by single developers that aren't ready for distribution
wouldn't be introduced to everyone in their incomplete state, but those
developers would be able to maintain their own versioning locally on
their work. This is not possible with svn alone.
This latter one isn't quite as important in my mind as the former, which
has turned away at least two developers, to my knowledge. Granted, I'm
unlikely to attach significance to the words of anyone who dismisses a
project because of its choice of SCM, but I get the impression that more
serious developers will expect the barrier for contribution to be lower
as we move into the future, and we could be missing out on their
contributions if we aren't using one of the newer distributed SCM tools.
Like others in the thread have mentioned, I think it would be worthwhile
to launch a pilot project - maybe a single plugin or theme - using git
as a test just to see what's what. If it lives up to everyone's
expectations, we could then consider it for core use. If not, or we
find it's not yet mature enough, or it's just too difficult for our
major contributors to change our workflow, I don't have a problem
continuing on with svn. But it seems like a worthy effort to figure out
whether it is a viable alternative to what we're doing.
Owen
--
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at http://groups.google.com/group/habari-dev
