Dropping maintainership of my ports
lists at eitanadler.com
Wed Apr 27 13:11:00 UTC 2011
> So, if the maintainers of the small leaf ports would be able
> to commit their work themselves, it would free the ports committers
> with the large ports projects on their hands to work on those ?
> Would this work ?
If you look through this list's archives I actually proposed this idea
myself once ;-) The problem is that (a) often maintainers don't do the
work 100% well. I am not knocking everyone - but just because you
_say_ you will do the work doesn't mean you _will_ do it. Furthermore
becoming a committer implies a certain level of trust beyond just "you
can submit patches" as you get shell access to the FreeBSD machines
and other such things.
> I use three boxes (for 8.1 i386, amd64 and 9-current amd64) to
> test. I do not use a tinderbox, as I assume considerable complexity
> to set one up.
You may want to mention that you use said three machines in the PR -
it would be of service. Different committers have differing policies
for what they will commit. Some will accept that, others will want
tinderbox logs, and others will always test every patch themselves.
> Does using a tinderbox make a large difference ?
Using a tinderbox ensures that a "clean" build works (ie that there
are no missing dependencies).
> I agree, the infrastructure is massive!
23,000 ports held together with Makefiles - I'm still surprised it works ;)
More information about the freebsd-ports