Christopher Blizzard wrote:
> The tree did manage to maintain the same quality of the time right
> before the release of the 0.9.1 milestone. However, we think that it
> wasn't because of the fact that drivers were approving specific bugs.
> With only a few specific exceptions we approved almost every bug that
> came across our collective plate. There are a few possible
> explanations for the continuing stability:
It seemed to me that rubber-stamp super-reviews were less frequent.
Perhaps drivers was holding super-reviewers to some level of
accountability? (Sometimes explicitly, maybe often implictly.)
Getting a= hurts (probably you -- drivers@ -- more than me!), but I
think it did make a positive difference. When I do sr=, I'm not
assessing risk, and don't have a vision of what's important or not.
``Just the code, ma'am.'' I think that a= serves a useful purpose to
that end. (E.g., ``wow, this is a big change with questionable value;
let's make sure it's extremely thoroughly tested before whacking the
tree with it. Remind me why we need this again?'')
I'll be a bit sorry to see it go. My $0.02.
chris
