On Thu, Sep 20, 2007 at 10:01:37AM +0200, Abdelrazak Younes wrote: > José Matos wrote: > >On Thursday 20 September 2007 07:33:25 Abdelrazak Younes wrote: > >>And even then, if you didn't shy away, or if you didn't loose interest, > >>you'll have to ask many times. You'll have to show that you truly > >>deserve the great honor to become a 'real' LyX developer. > > > > Are you saying that have commit permission will change this? > > Yes. It's like saying "You are welcome to contribute. We have a certain > set of rules that you ought to respect (coding style, ask for reviewing, > etc) but we trust you initially to be cautious with commit." > > Having the key to house is a duty that encourage them to try to do the > right thing from the very beginning. It is also a way to say that we > value their contribution and that we are not an arrogant bunch. > > If someone doesn't respect our rules, we just have to explain to him > what he/she did wrong. If he/she keeps not doing right, we just revert > his contribution. In practice, I am pretty sure this will never happen.
I am pretty sure that it takes a while to get accustomed to the atmosphere, even for otherwise respected programmers as each project works slightly differently. I've seen quite a bit of code slipping in the previous year or two that did not adhere to about the only strict rules we tend to enforce (code formatting). I take this as a sign that people are not accustomed to it. Even if one does not agree with the rules, it is simple to stick to them. > > I am puzzled. > > > > We always expect new developers to be cautious with commits, what is the > >difference then? > > The difference is that we don't scare off wannabe developers. I know for > sure that we have lost some actual, valuable contributors because this > secretive, restrictive policy. Not giving commit rights is not a way to scare off fresh people. There are OS projects that have just one person with commit rights that work nice with "outside" contributors. > > In a sense the organisation of a free software like is not much > > different from other voluntary organisations. I would be foolish not to > >learn from their experience. > > How come that other successful free software projects are not that > secretive and restrictive then? Couldn't we learn from them? You are > blocked by the mental model imposed by the centralized SVN way. Note > that SVN is just a tool we can be very liberal with this and other > projects do just that (KDE). Look at gcc and gdb. Or Tcl/Tk. Or Noteedit. Or ... Neither will grant you commit right just because you happend to come up with a patch. > > I have been working with other voluntary organisations for more than 10 > >years, from that experience I know that there are people with a sudden > >desire to contribute the funny thing is that after some time that interest > >fades away and no longer hear from them. > > Or they may come back. And that is very fine. No, usually the interest lasts only a few days, weeks at most. Having a timeframe of a few months acting as filter works rather well. Andre'
