On 9/21/2010 3:50 PM, Ariel Rabkin wrote: > My sense is that there's virtually never feedback on patches prior to commit. > There is, however, often feedback on design choices before the patch is > posted. > > I assume even with commit-then-review there would still be review for > major change proposals before implementation.
That's what we've done at httpd. Often a sandbox is created in svn to help facilitate collaboration on a particular implementation experiment, and then merged back to trunk once there is consensus around the major new idea.
