[Nix-dev] Again: Why don't these people have commit access

[Michael Raskin]

>Increasing the number of committers is certainly good. Of couse, as a
>committer, it's not that you have to always push your own complex stuff,
>instead create a PR. You for example you could certainly push commits that
>in most cases don't affect much people or that you are ready to revert,
>ecc. ecc.
>
>I was never asked for following any rule, except not triggering mass
>rebuilds, when got commit access. I tend at committing simple stuff that
>don't break, otherwise still create a PR.
>
>Your unfortunate case is that most of your PR are complex or controversial
>:P offlinehacker has the same problem, he has different needs and obviously
>want to push those needs. And many others.

I must note (from remembering Marc's commits back before Git migration)
that there are also many small updates that are simpler to keep in
branches unless there is commit access anyway...

(that is why given that MarcWeber is more qualified in many areas than
me, I would recommend @rbvermaa Rob just to make unsolicited GitHub 
offer of commit access).

>Also some PR are basically very hard to test for many people. For example
>testing new kernel options. Or because committers don't have knowledge on
>every field, and thus resilient to push.
>
>Just to say, it's not lack of trust, or being negative. It's lack of time
>(or platform) for proper testing and lack of time for proper understanding
>(linux is becoming too giant) the consequences of a change.

I think where we need an explicit commit policy is for reverts.

That would reduce negative surprises in some cases and it can also show
where we want to get job done 80% of time and revert the mistake 20% of
time.