handling of "ship it"

[Leo Franchi]

On Wednesday, November 03, 2010 04:54:31 PM Lydia Pintscher wrote:
> Heya folks
> 
> In the last days I've seen a few occurences of the following:
> 
> * patch gets submitted to reviewboard
> * patch gets very quick review
> * ship it is clicked by one person or more
> * patch gets commited
> * person most familiar with the code than the person who said ship it
> finds time to check reviews and says that the patch should not have
> been commited to trunk
> * unhappyness of various degrees follows
> 
> Clearly we have a problem here. How are we going to fix it? Suggestions?

I think given the fact that we have a large codebase and people who are 
experts on their own parts, we can do a few things. 

* Make important comments on the ReviewBoard request. That way *all* 
developers are on the same page. I've been guilty about this too.*

1) Identify active developers who are very familiar with a certain segment of 
Amarok. Active means "reads the ML, responds to review requests" 
   For example: Bart: PlaylistProvider/Podcast

2) For review requests that are significant and affect of these portions of 
Amarok: 
      Give the maintainer 2 weeks to say Ship It *or* Don't Ship.

3) If after 2 weeks the maintainer hasn't explicitly rejected or approved it, 
other developers can do so as well. 

What's this for? So someone who has concerns with codebase that they are very 
familiar with is able to give his/her opinion before the patch gets merged. 
We've had things like these bite us before, let's all try to avoid it 
happening in the future.

thoughts?

cheers,
leo

-- 
leo at kdab.com                                 KDAB (USA), LLC
lfranchi at kde.org                             The KDE Project