Re: Reviewing changes

[Thomas Lord]

One advantage of a formal review process
is that (if designed well) it creates a 
record of the review -- there is accountability.

In Karl's system, if someone asks "were all
of the changes in this release reviewed?" karl
can say "Well, all the diffs were posted to a
mailing list."

In a formal system, one can say, "yes, here
is a list of who signed off on each change" and
perhaps even "here is the checklist they filled
in and the comments they gave in answer to more
general questions that are part of a review".

Would such accountability and such discipline of
practice improve quality?   It would seem to 
depend on the nature of the particular review 
process, and of the people in the project, 
wouldn't it?

-t



On Thu, 2008-11-20 at 00:12 +0200, Eli Zaretskii wrote:
> > From: Karl Fogel <address@hidden>
> > Cc: address@hidden,  address@hidden
> > Date: Wed, 19 Nov 2008 17:06:07 -0500
> > 
> > We don't need to have a fully-specified, formal review system to benefit
> > from more frequent informal reviews.  Many projects get by on just
> > having the diff+log appear in the same email -- then the review "system"
> > is simply people reading their email.  It works quite well.
> 
> If there's no agreement to have a review process, I can simply ignore
> your review.
> 
>