[Ksummit-discuss] [CORE TOPIC] Recruitment (Reviewers, Testers, Maintainers, Hobbyists)

[Darren Hart]

On Fri, Jul 10, 2015 at 12:32:34PM +0000, Jason Cooper wrote:
> What we're also looking at here is adjusting the wetware / software
> boundary.  The goal of checkpatch is the same: offload maintainer brain
> time onto automated software.  But it's a double-edged sword.  Now we
> see checkpatch-only patches.  Was it a net-gain for maintainer relief?
> Or simply a shift in the type of annoyances?
> 
> I don't know.  The lesson learned from checkpatch and other tools
> shouldn't be lost on any effort we try today.  That's not to say we
> shouldn't try.  Rather, we should try new processes and tooling, *and*
> establish sane goals for the same.  "After 1 year of autotest at vger,
> we'll poll the maintainers and ask if they noticed a difference."  This
> implies some sort of tag system so maintainers can trivially tell when a
> patch has cleared autotest at vger.  Maybe:
> 
> Autotest-passed: https://autotest.kernel.org/2015/07/10/<cookie>
> 
> Depending on the proposed system, the goal can be more quantifiable, if
> needed.

I was going to suggest pretty much the same thing.

> 
> > Josh suggests that we should provide a service that people could push
> > code to and 'get automated feedback on what needs fixing'... but isn't
> > that what checkpatch was for? OK, a local tool can't cross-compile it
> > for you on every platform we support, but it can do a lot of stuff
> > short of that.
> > 
> > I do like the idea of a 'test' mailing list which receives patches and
> > checks them for corruption though.
> 
> Agree.  And in sensible cases, it can suggest a fix.  e.g. "Missing
> S-o-b, please read the DCO [link]. If you agree, add your S-o-b to the
> end of the commit message.  'git commit -s ...' will add this for you."
> 

I presume this would go back to the list and the submitter. I can see arguments
for and against including anyone else on Cc from the original patch. What did
you have in mind?

> wrt recruitment, It would be advisable for a few maintainers who are
> interested to be subscribed to autotest at vger to keep an eye on things
> and intervene if necessary.
> 
> thx,
> 
> Jason.
> 

-- 
Darren Hart
Intel Open Source Technology Center