> I already have ShipIt::VC::Mercurial.  There's no obvious next step.  Now I 
> give up.

I've had like issues with many CPAN maintainers, and build systems. I
for one *abhor* using svn, hg, and darcs. There is no good reason to
it -- I just don't have the experience with it or need it typically.
It boils down to being a gatekeeper for a casual git users like
myself. But, I diverge. I find this to be a meta-issue of CPAN --
which I'm quickly viewing as bad for workflow. My new MO is to fork on
gitpan (if the repo isn't natively Git). Then to publish on CPAN if I
feel like it. This only sounds offensive at first sight, but I think
it is a lesser crime then letting patches die in a dark-net because
you can't invest the time to get them upstream and the author can't
invest the time to throw you a bone.

My goal in writing this isn't to vent, it is to convince you that
forking DateTime is not evil. It's all we have. Get over the stigma
and make all crimes lesser to letting your patch die off of CPAN. Fork
DateTime into DateTime::SanePractices. Git really has done wonders for
making code public. Anyone can fork on gitpan and apply a patch at the
very least. The mentality is even something that I'm fundamentally
more comfortable with: forking is a good thing.

I think I have patches to a few of Dave's things that I never got
upstream for one reason or another. Many of my forks on gitpan are
used locally.
http://github.com/EvanCarroll/Fey-Loader

-- 
Evan Carroll - e...@dealermade.com
System Lord of the Internets - Dealermade
web: http://www.dealermade.com
ph: 888.403.9143

Reply via email to