I hate to be the one stating the very obvious, but git is already used
as the revision control system of the Linux kernel. The Linux kernel
is a major open source project, which has large commercial sponsors,
and a great many developers. In short, git is not a flash in the pan -
it is probably the pre-eminent open source RCS. What makes you think
that you're special?  :-)

I doubt that if you cornered *any* of the Postgres devs they'd be
willing to defend the use of CVS on any grounds other than
institutional inertia (that was certainly my experience!). This
institutional inertia is perhaps understandable - a major
infrastructure project like Postgres ought to be very conservative.
I'd speculate that Postgres devs are inclined to think "it may not
matter all *that* much what RCS is used if everyone already works
around CVS's considerable limitations, so why spend time and effort on
lobbying for git?". They would have a point. Besides, why should
Postgres be the vanguard? It makes more sense for its satellite
projects like Slony to lead the way, in my humble opinion. It's
extremely unlikely that by adopting git, the Slony devs will have made
the wrong call, and will fall out of lockstep with PG - I haven't
encountered any voices in the wider Postgres community that are
lobbying for anything other than git, so there probably aren't any, as
has been noted on this thread. Besides, even if that does happen, does
that actually matter all that much?

Regards,
Peter Geoghegan