What is the problem with ports PR reaction delays?
freebsd.contact at marino.st
Sat Jan 25 00:23:09 UTC 2014
On 1/25/2014 01:03, Big Lebowski wrote:
> On Sat, Jan 25, 2014 at 12:41 AM, John Marino <freebsd.contact at marino.st>wrote:
>> On 1/25/2014 00:30, Big Lebowski wrote:
>>> Hi everyone,
>>> I wanted to ask about the growing time of reaction to ports PR's - what
>>> the problem? It seems to me, as a ports contributor, that this time is
>>> growing, not shrinking, and there's no formal/automated procedures that
>>> would help in managing the issue.
>>> What is the problem?
>> Seriously? You did all that research and didn't notice this?
>> query all open ports PRS: 1705 problems total.
>> Short answer: Sheer number of PRs.
> I did, in fact, and this is not the problem, this is the symptom of the
No, it's not. PRs are getting closed, all the time, every day. But PRs
are coming in faster than they are getting closed. A couple of days
ago, the number of open PRs was down to 1660.
You can say the solution is to increase the rate of PR closing. I could
counter with not allowing any new PRs to be submitted. Both would solve
the problem, and my way is easier to implement. :)
>> I would have thought the situation is kind of obvious. There are lots
>> of PRs, nobody is paid to process them, there are no teams dedicated to
>> "new ports"
> Perhaps there could be? Or some commiters who would assess if the PR can be
> handled easily or it needs a more experienced eye to look at it?
Timeless question: How does one force a volunteer to do anything?
Answer: You can't.
>> , lots of PRs are stuck because their non-freebsd.org
>> maintainers are MIA or abandoned their emails and nobody is going to
> Shouldnt those be invalidated and closed to reduce the workload illusion
> that tends to set off people from doing work?
Now, who is going to do it?
And after you select him or her, how do you force them to do it?
>> touch a PR in "feedback", etc, etc. Finally, some committers are pretty
>> sloppy at closing complete/obsolete PRs.
>> When people get paid to do it, I think the situation will improve. :)
> Great, then perhaps FreeBSD deserves dedicated, paid commiter to handle
> that situation? After all, we're donating money to the project, and if
> there's general consensus this is a problem that needs to be fixed and this
> is the only way to do it, then perhaps something can be done about it?
I don't control FreeBSD Foundation funds in any way.
Having a full time paid person doing PR triage would put a huge dent in
the backlog. But we're just wishing, right? There aren't many paid
positions and those don't do clerical work like trouble tickets.
More information about the freebsd-ports