> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On
> Behalf Of Oswald Buddenhagen
> Sent: Thursday, January 24, 2013 12:17 PM
> To: [email protected]
> Subject: Re: [Development] Nominating Iikka Eklund as approver
> 
> On Wed, Jan 23, 2013 at 01:32:09PM +0000, Koehne Kai wrote:
> > Parts of the packaging process is in gerrit (namely qtsdk/qtsdk.git),
> 
> > so he should be really  be able to stage things :)
> >
> uh, wtf?! nothing against iikka specifically (couldn't judge it), but this is 
> so
> totally *not* a reason to make someone an approver. in areas where no
> approvers typically operate (mostly playground projects), we create special-
> purpose groups where the involved people have the rights they need to get
> the work done. approver status is supposed to be earned by displaying
> reasonably universal competence within the project, not being the only guy
> available to check +2 in a particular area.

I guess giving Iikka local approver privileges for qtsdk.git would cut it, too. 
Anyhow, you might then start to wonder whether anyone who's an approver for 
qt-creator necessarily needs approver rights for qtwebkit, qt-creator etc ... I 
think it's reasonable to expect that people exercise their approver rights with 
care, and only use it were appropriate.
 
> fwiw, this is not the only nomination in recent times that makes me scratch
> my head ...

Yes, seems we've not yet a common understanding of the actual requirements for 
becoming an approver. My personal check list would be:

 - has the person the technical skill to be able to judge other people's 
contributions in his area of interest & expertise?
   - does he know about coding style, commit style etc?
   - does he follow the relevant mailing lists closely enough to be informed 
about changes & global announcements?
   - does he know which branches one should commit to?
 - has the person shown in the past that he's reasonable , and can one expect 
to exercise his rights with care?
 - is it expected that the person will continue to contribute? 
 - finally, does the project as such gain anything if he gets an approver? 
F.ex., does he review fixes were there's no (official or de-facto) maintainer 
that wants to review the same changes anyway?

All of this should be seen in the context of the area of interest & expertise. 
A person developing Qt Declarative for instance doesn't have to necessarily 
follow the qt-creator mailing list, nor know which branch to commit to there. A 
person working on docs doesn't have to know the details of the C++ coding 
style, but if you're approving doc changes you better know qdoc. And so on and 
so on.

I think all of the above can be checked for Iikka for the area of packaging, so 
I don't see a reason for him not becoming an approver.

Regards

Kai

> ______________________________________________
> Development mailing list
> [email protected]
> http://lists.qt-project.org/mailman/listinfo/development
_______________________________________________
Development mailing list
[email protected]
http://lists.qt-project.org/mailman/listinfo/development

Reply via email to