From: Beman Dawes (bdawes_at_[hidden])
Date: 2004-02-25 20:27:08
At 09:31 PM 2/24/2004, Thomas Witt wrote:
>here is my next try at an improved formal review procedure. I did away
>with the additional mailing list. The outlined process tries to reduce
>the burden on all involved as far as possible.
I've been thinking about the review scheduling process, and was about to
propose we make it a bit more "open". The basic idea being that we keep
enough current information available in CVS so that authors can round-up
review managers themselves, and that review managers can handling the
scheduling without the Review Wizard doing anything. The Review Wizard's
job then becomes one of keeping an eye on the process to set policy, smooth
out bumps, assist when needed, and make sure there are enough qualified
Review Managers. This should eliminate the Review Wizard as a bottleneck
for setting up most individual reviews.
The basis for such a self-managing process would be two CVS files:
* A queue of people willing to act as review managers, ordered so the
people who have managed reviews least recently are at the top of the list.
Only the Review Wizard should add people to the queue, but once on the
queue Review Managers are responsible for keeping their position updated
(by moving their name to the bottom whenever they accept a review to
* The review schedule, including all future reviews that have review
managers, and with dates if the review manager has approved a date for the
review. Maintained on a first-come-first-served basis by the review
The procedure would be similar to Thomas' proposal, except who does what
changes a bit:
>* The author posts a prereview request to the list. Subject line must
>contain "[prereview request]".
Part of the request is a request for a volunteer as Review Manager
>* The author posts review request to review wizard. Subject line must
>contain "[review request]" and
> the message should contain a link to the prereview request.
I think this message will no longer needed.
> The review request must be seconded by two boost members. To second a
>review request does
> not imply a positive review. A person who seconded a review request
>can act as a review
> manager. If nobody has volunteered so far the review wizard helps
>finding a review manager.
If no one volunteers as review manager, let's make it the proposer's job to
find a review manager from the top of the review manager's queue. The
Review Wizard can privately twist arms if needed, but hopefully that won't
be needed for most review requests.
>* The review manager checks whether the library is suitable and ready
Yes, as always.
>* The review manager sends a schedule request to the review wizard. The
>subject line must
> contain "[schedule request]".
That step can be eliminated if the review manager schedules the review
>* The review wizard schedules the review. Reviews will be scheduled on a
>first come first serve basis,
> based on the schedule request.
The review manager schedules the review and updates the CVS reviews
>* The review manager announces the review.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk