maintainer timeout for FreeBSD commiters

[Chris Rees]

On 18 Jul 2012 11:33, "John Marino" <freebsdml at marino.st> wrote:
>
> On 7/18/2012 12:19, Chris Rees wrote:
>>
>> On 18 Jul 2012 07:44, "John Marino"<freebsdml at marino.st>  wrote:
>>>
>>> Yes, somebody would have to set that up but it would pay big dividend I
>>
>> think.
>>
>> It also does away with the QA aspect that committers currently provide.
>> I'd like to repeat that people sufficiently familiar with the ports
system
>> to QA patches generally ends up with a commit bit fairly quickly.
>>
>> Chris
>
>
>
> I wouldn't assume people that become this proficient necessarily want a
commit bit.
>
> The whole point of my proposal is give and take.
> Yes, you take away "QA" responsibility from an entire pool of committers
and make it the primary responsibility of this new class of maintainer on a
per port basis (and not nearly all ports either).  I was proposing that
your gains (much less PRs, more often maintained ports) far outweigh the
liabilities.  I would be selective who gets assigned to this new class.
 They should have a body of work that instills confidence that they can
handle QA.
>
> You don't get something for nothing and it's not hard to revoke the
privilege if a person can't handle it.

You are making a good point, but I'm trying to explain that the 'body of
work' for proposing a new developer is no greater than the standard you
suggest.

We do have developers who only commit to their own ports; while it's
generally hoped that they work on PRs too, it's not a requirement.  These
would fall under the category of 'super maintainers' if you like.

For further reading, Google 'Solutions for the PR load problem'.

Chris