On Tuesday 17 November 2009 14:24:27 Jorge Vargas wrote: > On Fri, Nov 6, 2009 at 3:01 AM, Diez B. Roggisch <[email protected]> wrote: > >> IRC is not an optimal medium for development discussion, IMHO. It is > >> not archived properly, unstructured and hard to search. Ideally, > >> development discussion and decisions should be documented in the docs > >> or the ticket system, but a mailing list is the next best thing. > > > > You are so right about this. For starters, I'm not even *allowed* to go > > to IRC at work. So from the about potentially 16-18h a day, I could only > > stay on IRC for about half of them. But then there is the issue of > > timezones, the fact that permanently following a chat in the hope of > > something interesting or relevant happening is massively distracting (to > > me at least), the interleaved discussions of others... > > > > I don't mind the occasional IRC conference. But even that I can't > > guarantee to attend. Real Live and such. > > > > And where are the results of those discussions you guys have? Can I read > > them? Can I comment on them? Can I *work* on them? No. > > I'm sorry all I said was that most of the support and bugfixes have > been moved there. All mayor changes had been discussed in the list.
Well, if people accept that IRC is the primary source of support - then mote it be. I hung out there the last couple of days in the evening, and nothing was really happening. But to be honest: even for support, I'd prefer if people would give it on the ML. Because there it is persistent, visible, and searchable, linkable. So IMHO those giving support there should consider moving that support over to the ML. Which I can't force anybody to do, but maybe the core devs and supporters can ponder this. > > I do agree with what you say and from my part I'm sorry we should > indeed use the ML more to report some bugs for everyone to see, > although other things are way better in a faster feedback channel. For > example the recent content-type bug will not be fixed if we didn't had > people on IRC discussing it. These are two different things. Of course a fast channel of communication is benefitial in a concrete situation such as fixing a bug - I've done that numerous times with Florent over IM. But for others to report and see if bugs are reported, IRC (or any other non-archived form of communcation) isn't the right thing. See above. > > > So I'm feeling I'm cut out of the development efforts. The public > > visibility of those is reduced, as no one can follow discussions and > > decisions by searching through ML or TRAC. Which might give the > > impression to others that the project isn't very much alive. > > I don't agree with this. In the end progress equals code and the more > checkins you see the more active a project is. Yes ML and trac help > with the impression but a project that is getting tons of tickets is > not more active than a project that is getting tons of checkins. Where is that flurry of checkins happening? I don't *doubt* it - I just don't find it. Which might mean that I'm to stupid for the whole bitbucket/HG-thing, but when I look at http://hg.turbogears.org/tg-21/ I see the last checkin being 4 months ago. If I'm supposed to look anywhere else, pray tell me. But please do *not* tell me that I've got to look at all forks. That they exist is of course perfectly fine - but frequent back-integration is crucial to prevent doubled or worse, conflicting work. Diez -- You received this message because you are subscribed to the Google Groups "TurboGears" group. To post to this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/turbogears?hl=.
