On Wed, Jul 21, 2010 at 6:51 PM, Roan Kattouw <roan.katt...@gmail.com> wrote: > I would say so, yes. The more people 'lightly review' something, the > less likely it is to have obvious-ish flaws. However, most of the time > the name of the person that reviewed it is the most significant piece > of information, assuming you know the reviewers well.
Maybe, but I don't think we need actual UI for indicating this. What we'd really want to know is whether at least one trusted person has carefully looked at the code and really thinks it's correct. IMO, we should have a system that encourages people to commit to a change's correctness, in a way that they can be held accountable (via people paying less attention to their review) if they miss things too often. Having a secondary review level that doesn't really mean much because you only superficially reviewed the code will encourage people to use that level of review so that they can't really be blamed if the commit is actually bad, rather than taking responsibility for it. _______________________________________________ Wikitech-l mailing list Wikitech-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-l
