[Telepathy] The telepathy-reviewers mailing list.

[mikhail.zabaluev at nokia.com]

Hi,

> -----Original Message-----
> From: telepathy-
> bounces+mikhail.zabaluev=nokia.com at lists.freedesktop.org
> [mailto:telepathy-
> bounces+mikhail.zabaluev=nokia.com at lists.freedesktop.org] On Behalf Of
> ext Sjoerd Simons
> Sent: Saturday, March 26, 2011 2:32 PM
> To: Olivier Crête
> Cc: telepathy at lists.freedesktop.org
> Subject: Re: [Telepathy] The telepathy-reviewers mailing list.
> 
> > I don't think it's a good idea to have random people review stuff for
> > random modules. Since our modules are relatively small, I think there
> > should be a owner/maintainer for each module so we have a minimum of
> > accountability. Also, the maintainer should always have an overview
> of
> > the changes to his module, so that he can maintain a good level of
> > consistency
> >
> > Also, some modules (like tp-farsight/farstream) require specialized
> > knowledge so in practice only one or two people could review them
> > anyway.
> 
> I'm not sure how you get from Will saying that people should use their
> own judgement to what patches they can review to ``have random people
> review stuff for random modules''. Note that this is also encoded in
> the
> criteria themselves.
> 
> The "Criteria for Reviewers" section has the following:
>   They are also expected to show good judgement in whether or not they
>   review a patch at all, or defer to another reviewer.
> 
> Futhermore it has:
>   They should also be in touch with other reviewers and aware of who
> are
>   the experts in various areas.
> 
> The first bit is crucial for a reviewer imho, it basically says ``Know
> your own limits''.
> 
> Do note that using this procedure doesn't any way change how things are
> currently done wrt. reviewing. It's just there to have a more clear
> definition of when someone gets/can expect a certain status.
> 
> 
> To come back to the original question, I'm totally fine with having a
> grand unified list of reviewers and comitters and would very much be
> against anything else. At some point in time we had a wiki page with a
> matrix of projects and reviewers, which is just a massive pain to
> maintain and will never reflect reality.
> 
> The reality is always that some patches can be reviewed by essentially
> any comitter and others can be best reviewed by one particular person.
> That one person can be different for different areas of a project!

I support Sjoerd's points. In our team in Nokia, I encourage everybody to use the common mailing list for reviews, with a stated premise that anybody can voice their concerns with any patch. The preferred reviewers are indicated with Cc in the message or personal assignment in ReviewBoard. In practice, people cluster around projects they know well anyway, so the "good judgment" approach works (except, it's usually me who raises the occasional hell about things people thought were OK :)).

Best regards,
  Mikhail