Re: [gentoo-dev] Requiring two sets of eyes for all eclass commits

[Roy Bamford]

On 2010.04.24 18:40, Petteri Räty wrote:
> 17:34 < Betelgeuse> robbat2|na: how easy to it to prevent commits to
> CVS
> if the commit message doesn't match a certain pattern?
> 17:36 <@robbat2|na> go and checkout the CVSROOT and there should be
> an
> example there
> 17:37 < Betelgeuse> robbat2|na: Ok so doable then. Thanks.
>
> What do you think about not allowing commits to eclasses without
> mentioning an another developer who has reviewed and approved the
> diff
> in the commit message? There's enough people on gentoo-dev for urgent
> stuff too.
>
> Regards,
> Petteri
>
>
In industry, the practice is called peer review. Its generally thought 
to be a GoodThing as its part of the process of trapping errors as 
early as possible in the process, where they have lowest cost.

We cannot easily attribute cost in terms of money, so think about it in 
developer and user hours wasted as errors 'escape'.

Industry also recognises the need that any process needs to be tailored 
to the circumstance so the peer review process is not enforced. Project 
groups are permitted to assess the risk of screwing up against the cost 
of a fix. (That's overly simplistic).

In short, following industry best practice, the peer review process 
should be strongly encouraged but we should stop short of using tools 
to enforce it.

-- 
Regards,

Roy Bamford
(Neddyseagoon) a member of
gentoo-ops
forum-mods
trustees