On Thu, May 14, 2009 at 7:50 AM, Jonathan Ellis <[email protected]> wrote:
> On Thu, May 14, 2009 at 9:45 AM, Matthieu Riou <[email protected]> > wrote: > > Not to mention those who are in another timezone or those who would like > to > > contribute on their spare time (like a day here and there). If the real > > development happens outside the ML, you won't even know those people > exist. > > I can appreciate the real benefits of the ML but neither of these is > among those. We have developers from US, EU, and I believe asian time > zones on IRC and we overlap enough that it works out fine. > > If anything, IRC is _better_ suited than a ML for accomodating people > who want to pop in and fix something because of its much lower > latency. > I think none of us said IRC is completely bad and should be avoided at all cost. All we said is that important decisions taken on IRC should be further discussed on the mailing-list. And for good reasons. What if I found a bug on Cassandra that I didn't report yet but only learn it's too late with the release email? What if I have no clue that you're doing this instead of that and start working on a patch for that? What if I'm foillowing the ML to see in which direction the project is going? What if I can only contribute on Saturdays? And the list goes on... Again, IRC is good for technical (and non technical) chit-chat and formulating quick agreement. But it's really bad as a complete ML replacement, more than one open source project got really hurt this way with no new blood showing up. Matthieu > > -Jonathan >
