Ability for maintainers to update own ports

[Charles Swiger]

On Nov 10, 2003, at 3:25 PM, Eric Anholt wrote:
> One thing I've wondered about, though, is the process for people 
> getting
> commit bits.  Is it always on the initiative of an existing committer
> (as it was in my case) or would it be considered acceptable for people
> who would like to become committers and maintain their ports directly 
> to
> propose themselves on a list for someone to pick them up as a new
> committer?

There may be some advantage to having an existing committer act as a 
guide and mentor for the person gaining commit privileges; certainly, 
that was the way Apple handled giving CVS commit access to the Darwin 
tree for non-Apple people.

With regard to the origin of this thread, it might be useful to 
understand what the actual scope of the problem is: how long does it 
take, on average, for a new port submission to be committed?  My 
limited experience suggests an interval between one week for a "highly 
interesting" port (dvd+rw-tools, in my case) to three or four months 
for other ports, but real data pulled out of the bug database would be 
more reliable than personal impressions.

It might also be useful to set realistic targets for response times and 
mention them in the Porter's Handbook, so that people who submit ports 
know what to expect in terms of a timeframe.  I'd really like to be 
able to suggest something like "you should have a committer review your 
submission within one month" and have that be doable, but I can't say 
whether that (or any particular fixed interval) would be practical.

-- 
-Chuck