Summary from yesterdays (mini) FESCo meeting

[Dennis Gilmore]

On Friday 29 December 2006 15:24, Michael Schwendt wrote:
> On Fri, 29 Dec 2006 14:14:57 -0600, Callum Lerwick wrote:
> > I... am absolutely astounded by all this. For doing reviews, I keep this
> > template in a Tomboy note:
>
> -snip-
>
> > Which follows the review guidelines pretty closely. When I finalize a
> > review, I just copy and paste this template into a new note, go down the
> > ReviewGuidelines list, and type in an "Ok" or a "NEEDSWORK" for each
> > one.
> >
> > If the time required to copy and paste and type some OK's would add
> > significantly to your workload, I dare say you aren't putting in
> > adequate time, thought and effort into your reviews.
>
> Wrong way of thinking to begin with.
>
> It would be wrong for me to reduce my way of reviewing packages
> painstakingly to such a short'n'static list of MUST/SHOULD items.
I agree
> My personal list of MUST/SHOULD/HINTS items depends on the package I
> review. I adapt to what I see and to what seems necessary, and I believe
> several other contributors do the same. Neither do I want to process items
> which are irrelevant, because they don't apply to a package, nor do I want
> to create a list for all things I perceive as "okay". If my other messages
> in this thread have not been clear enough already, I can't help it.
I think thats really good,  if its not relevent skip it 

> Since I had started contributing reviews long ago, and long before some
> people invented metrics, I've never reached a point where I no longer
> found any new/bad things not covered by packaging/reviewing guidelines.
> I'd like to keep that style of doing reviews.
thats a good place to be.
> Anyway, the number of Fedora Extras contributors has reached a certain
> point where I think I won't spend a lot of energy into discussing things
> like this, especially since I haven't done many reviews for quite some
> time.
What i tried to achieve by starting all of this was something more than Just 
Approved in bugzilla.  for a few reasons.  
1) new reviewers look at existing reviews a s a guide on what to do.  So while 
the reviewer may have been through in the review people cant learn from it.
2) if questions crop up later about a package   its easy to find answers  
especially if the reviewer is no longer around. 

The output in the review needs to be flexible and not cluttered with noise.  
But it also needs to show that a review indeed took place
-- 
 ,-._|\    Dennis Gilmore, RHCE
/Aussie\   Proud Australian
\_.--._/   | Aurora | Fedora |
      v