Changes to review policy

R Kent James kent at caspia.com
Thu Apr 28 16:44:36 UTC 2016


On 4/27/2016 10:33 PM, Jim Porter wrote:
> Therefore, I'd like to propose the following: any change to
> Thunderbird's defaults should have a super-review before landing.

Let's ask these questions:

1.    Is the Correspondents column in fact a debacle that in hindsight
the majority of key developers agree was a mistake as implemented?

2.    Would a super-review process have prevented the debacle?

3.    Do we have sufficient reviewers that we can add complexity to the
review process without needlessly slowing down an already slow process?

I'm afraid that the answer to all three questions is no.

Re 1), as the guy who takes responsibility for the shipped release code,
I had to review the case of the correspondents column a couple of months
ago, asking whether it needed to be reverted in some way prior to TB
45.0 upgrades. Looking in particular at the discussion in bug 1152706,
there was not sufficient consensus that there needed to be a change to
push for a reversion prior to upgrades. Even now the support is not clear.

Re 2), the original bug 36489 had plenty of comments from all of the key
players. This is not the case of some inexperienced module peer
approving something that wiser "super-reviewers" would stop. As long as
I have been involved with Thunderbird, there has been a general
consensus among the key UI people in Thunderbird that new features need
to be pushed to users, even if that disrupts the workflow of a
significant number of users. I have fought this on occasion, and always
lost the battle.

Re 3), lack of reviewers for key changes is already a big impediment to
progress. It takes me two years to get major changes through the
existing process. I've generally promoted more lightweight review
processes, like our general policy that any reviewer can review anywhere
they feel qualified. I don't believe that we have the resources to add
complexity to an already difficult review process.

So while I think that we have a problem in sometimes forcing unwanted
changes on users, I don't think that the proposed super-review is the
solution.

:rkent



More information about the tb-planning mailing list