Hello I think this is actually a very bad suggestion, which would only do more to fragment and further decrease developer interaction. For the past decade the primary means of undertaking development has been the ML. I don't see any of the major issues being discussed on the forum. Your suggestion is that we should end using our main development communication method, to a system that most of the main developers don't use at all, and most avoid like the plague. Really, what sense is there in that?
Most developers are on the ML. It still remains the main development communication path. Just last week loony cyborg wanted the project to buy some piece of software. I wasn't too sure how to answer him so what did I suggest? Go to the ML. There wasn't much response, but its still elicited some response. The very fact we're discussing this issue here on the ML, rather than the forums, might be the best proof of its central importance to this program. I understand that forums do have general advantages over email chains, mostly visual/spacial organization, but this has not limited our development in the past. In my mind, the biggest problem with your proposal is the actual transition. You're going to have to convince everybody to use the forum. That takes effort at a time when, as you suggest, enthusiasm is at a low and the general views of the forum is even lower. I also disagree that it is a good place for discussion. The forum would only increase the useless noise level in serious development threads, for the most part with people who have very little invested in the actual development of the program.Just imagine trying to have the iPhone discussion several years back but with far more rancour and accusations: it would be impossible to get a useful debate out of that forums. Finally, it also removes a lot of former developers and interested parties from pertinent discussions. Many of the older devs might not be active IRC participants, but they do read emails because they are supposed to be an digest of views. Lets say that someone wants to undertake a change in the UI. There are seve ral former developers who I know still read the ML and can add their perspective, as they have in the past. Many are not going to migrate over to the forums, so you've just removed them from the ongoing development discussion. That's an exceptionally bad outcome in my mind. So that should not be the approach we take. I think what we should try to do is to try to better use the ML, mostly because its our system for over a decade and it has served us quite well for most of that time. Actually the problem in my minds is that we're actually not using it enough. We've become a bit loose in our development process where some of the decisions were made on IRC with limited consultation, or that people are not taking the time to look through the emails and critique them as necessary. I've been guilty of that myself. Frankly, that's not going to be solved by having us move the discussion to another format, especially one that is so disliked by a good portion of the development community. Rather developers should be encouraged to be more diligent in bringing stuff to the ML and make an better effort in encouraging dialogue there. R On 2014-02-02, at 5:43 PM, "Ignacio R. Morelle" <[email protected]> wrote: > On Sunday 02 February 2014 16:08:54 Nils Kneuper wrote: >> People agreed that it would make sense to e.g. have some kind of "devblog". > > I feel like this overlaps a lot with the "Culture of discussion" thread. > > If developers wanted to promote and discuss their changes with the userbase, > the forums (and in particular one forum section [1]) have existed for _years_ > to serve that purpose. There are some people who have made use of this option > [2] [3] [4] [5], but it's not the norm. Not nowadays, anyway. > > 1: http://r.wesnoth.org/f2 > 2: http://r.wesnoth.org/t39430 > 3: http://r.wesnoth.org/t38748 > 4: http://r.wesnoth.org/t39570 > 5: http://r.wesnoth.org/t39840 > > Having developers post about _development_ in the forums aids a lot in both > showing that activity _is_ taking place (since not everyone uses IRC or > mailing lists in the 2010s) and allowing _feedback_ to take place in both > directions in a more relaxed environment than the bug tracker. > > Incidentally, most [6] of these post-FOSDEM discussion threads are a prime > example of choosing the wrong communication channel for important subjects. > How many users read the *developers* *mailing* list? Are *all* developers > even subscribed to this mailing list? What about prospective contributors? > And > which is ultimately more visible from www.wesnoth.org: the forums board, or > the mailing list? There must be something wrong with the antiquated ML > approach if every single thread I post here has barely any activity or > visibility, regardless of its importance. Furthermore, there is at least one > existing, _recent_ case of a developer making an extremely significant > decision on their own because their attempt at starting a mailing list > discussion was met with virtually unanimous silence. > > 6: Not all. The CoD thread, the asheviere thread, and this one are better > suited for the developers' ML for obvious reasons. > > Similarly, over the past years there have been severe communication issues > with regard to GSoC, and this has not gone unnoticed by the community [7]. > > 7: http://r.wesnoth.org/t38889 > > At least in previous occasions there were organization acceptance > announcements in the front page (not the most visible location for _existing_ > users, I should say), but last year there wasn't even that. Looking at how > other projects deal with GSoC, they are generally more open with regard to > GSoC events and results and encourage students to interact with the userbase > through blog posts and such, providing progress reports and talking about > existing or future design decisions. The Battle for Wesnoth Project may not > be > a purely technical project like a desktop environment framework or an > operating system, but the aspect still exists and is extremely relevant for > other members of our heterogeneous development team, the WML and Lua content > creators in our user community, innovative artists exploring the limits of > the > game engine, prospective mainline contributors, and enthusiasts in general. > > Right now we are in a position where many crucial areas of mainline > development are stagnating due to various reasons, and one of them is that > there isn't much interest from the userbase in contributing to Wesnoth. The > areas in which we need the most help are barely publicized (not unlike those > that are being actively worked on, though), and for most forumers it just so > happens that some random user will suddenly gain a red username and Developer > rank once in a blue moon, without an explanation at all. I have begun an > effort to alleviate this issue by publicizing new developers [8] and > unaffiliated contributors [9] in release announcements (which I also started > handling in order to make them more enticing and less opaque to the > community), but what we need the most in this regard is clean and unified > *documentation*, rather than pieces and scraps scattered across the wiki that > are not directly reachable from the front page or navigation bar _in one > click_. > > 8: http://r.wesnoth.org/t39622 > 9: http://r.wesnoth.org/t39850 > > (On a related note, having more active contributors always paves the road for > expanding the project's trusted circle and getting more Wesnoth.org staff > members, which is something that may become an issue if any of the four > people > with administrative access to baldras get hit by a bus, given that none of > us seem particularly active at once -- yes, this includes me.) > > Thus, I think that aside from starting a "devblog" (using what platform?) we > should look into making better use of the tools we already have: the front > page (casual observers!), the wiki (prospective/current contributors and > content creators!), and the forums (everyone). > > (On the same line of thought of the devblog, some months ago I thought about > starting a series of informal interviews with members of our development team > and posting them in the forums to raise awareness amongst players of the > _humans_ who invest so many hours of their spare time to instill life into > Wesnoth, and the technical aspects they focus on. Of course, that hasn't gone > anywhere past the planning phase yet.) > > -- > Regards > Ignacio R. Morelle <shadowm> > > _______________________________________________ > Wesnoth-dev mailing list > [email protected] > https://mail.gna.org/listinfo/wesnoth-dev _______________________________________________ Wesnoth-dev mailing list [email protected] https://mail.gna.org/listinfo/wesnoth-dev
