Owen's not saying that only certain people should be allowed to add things. He's saying there needs to be actual *review* before things are added willynilly. That means someone who is a good enough coder (and therefore qualified) needs to look at the code and make sure not only that it's working, but that it's a Good Idea (tm), is implemented well, won't open us to massive security exploits, etc..
I don't necessarily agree entirely. I think we often do err on the side of perfection far too frequently, holding up progress because we're waiting for the ultimate perfect solution to come along. Sometimes hacking together a new feature that gets the job done and refactoring it later isn't a bad thing. I'm also not opposed to opening things up for a big hacking session once in a while... but the same review obviously has to happen at some point, it's just against a branch instead of individual patches. As with most things the key is finding the healthy medium. Going whole hog (Gmail is showing a Hog Hunting ad next to this email right now) in either direction will create more problems than it solves. On Thu, Feb 19, 2009 at 3:34 PM, Sean T Evans <[email protected]> wrote: > > Actually, in order to commit, a review by a commiter makes sense. And > not all commiters are qualified to sensibly review each commit. Me, > for example. > > Sean T Evans > > On Feb 19, 2009, at 2:27 PM, Arthus Erea <[email protected]> wrote: > > > > > I think this argument really encounters some dangerous thoughts. > > > > Not only are you saying that a review by a committer is needed, but > > you're saying that not all committers are allowed to provide such > > review. > > > > Translation: Unless I or one of my pals approves of something, it's > > not part of Habari. > > > > On Feb 19, 2009, at 2:29 PM, Owen Winkler wrote: > > > >> > >> Arthus Erea wrote: > >>> > >>> On a theoretical level, patches should be reviewed in a timely > >>> manner. > >> > >> When people suggest bug hunts in the future, there should be a > >> commitment to supply the review and feedback you've suggested. > >> > >> This should not be an off-hand promise to merge a mangled branch > >> with a > >> weekend's worth of applied patches into trunk, often including new > >> features that non-committers try to railroad, not just bug fixes. It > >> should not be a babysitting session by a PMC designer/novice coder > >> with > >> commit access who just commits patches he's supplied that seem > >> functional. This wouldn't result in an adequate review, for your > >> personal purposes or for Habari. > >> > >> If a commitment to performing true reviews does not exist, then the > >> bug > >> hunt shouldn't happen. > >> > >> The hunting of bugs and committing of patches go hand in hand, and so > >> the scheduling of an event of this kind should account for both > >> actions, > >> not just one. > >> > >> Owen > >> > >>> > > > > > > > > > > > --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/habari-dev -~----------~----~----~----~------~----~------~--~---
