Hi, > Maybe most importantly, distributed revision control places any > possible contributor on equal footing with those with commit access; > this is one important step in making contributors feel valued.
I think this is a very important point, but subtle. I realize that's a dangerous combination, but I'm going to have a go at exposition. I think it is true that the distributed model _tends_ to make contributors feel more welcome, but it's not to do with permissions, it's to do with the process. The process is much more important than the permissions. If we want new contributors to feel welcome, we need a clear, explicit process, that everyone agrees to, and follows. I don't mean something enforced by permissions, but something followed, by convention, and with care, by all the developers. That provides a clear and healthy basis for people to join. In that situation, and in that situation only, new developers do not worry about whether they are clever or important or well-known enough to contribute code. That does tend to follow from the distributed model, because it is fundamentally built on the 'show me the code' model of development. Not surprisingly. I completely agree with Anne that we will work it out when we switch, and the details of process should not delay us. But, this is just a vote for some careful thought - and discussion - and agreement - on what sort of atmosphere we want to convey as a community. That atmosphere comes directly from our development model - or rather - the development model is the clearest indicator of what kind of colleagues we are. Are we careful? Are we serious? Are we thoughtful? Are we open? Are we clear? Do we value learning and teaching? Are we coding for the long-term? See you, Matthew _______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion