On Mon, Aug 23, 2004 at 03:12:51AM -0500, Branden Robinson wrote: > I certainly agree. The thrust of my comments was to make sure NMs > understand that licensing issues are often difficult, and that if one isn't > prepared to wrestle with them oneself, one needs to place more trust in > one's peers who do.
This is an important point which I fully agree with (more below). > I am dismayed and exasperated by the recent trend of bashing the > debian-legal list collectively, particularly when it comes from people who > don't participate in its discussions. Maybe there is some sort of "Real > Hackers Don't Need Debian-Legal" elitism going on; maybe it's just good > old-fashioned fear of what one doesn't understand. I get the feeling that the "elitism" as you call it is more related to the above. Perhaps there is some feeling that understanding licensing issues isn't difficult (which would be a misplaced feeling) but more importantly I think that those who have placed their trust in their peers feel that their trust might be violated. It's certaintly how I feel in some respects and it's how I read many of the other critics of -legal, both past and present, as well. That said, I think you're on to the solution by making sure people are aware that these issues are difficult. > At any rate, I'm not saying we need to make the P&P process turn our NMs > into legal experts. I *am* saying we need to educate them that legal > issues, even in Free Software, are sufficiently complex that expertise is > actually required. Armchair quarterbacking from a position of ignorance > is antisocial and corrosive to our organization. Agreed, and this approach would get rid of my reservations about simply pointing NM's to -legal for everything. My worry there was that we would either be turning prospective DD's in to robots or that we would be pointing them to so-called experts who haven't studied the issues but are more than happy to enforce their views on to others. I'm trying to come up with an effective outline for good NM questions to accomplish this. Here's what I've got so far, and I'd love more input, especially in terms of actual licenses that people with more experience in this area think are exemplary of the problems inherent in license analysis. - Each question is based on some actual license. This license is meant to highlite one specific issue, either codified explicitly in the DFSG or to be inferred from it. - Each question includes at least one link to the mailing list archive where the license was discussed, so that the NM can read what has been said about the license. The license should be picked so that the discussion isn't too long (the QPL would be a very bad choice, for example) and possibly with particularly good posts. If anyone remembers any very good posts that clearly explain and/or define a point, these would also potentially be noted in the question, or brought up by the AM after the question is answered. - The questions should be a mix of both obviously non-free, obviously free, and less obviously in either camp. For the last category, it should be stressed by the AM at some point, either in the question or the answer, that the boundaries are not so clear cut and that these can be very difficult issues. - I think around 4 to 7 licenses would be a good number to have. There were 4 when I went through NM, and I have to look them over again (along with my answers) so I don't want to comment on them right now. Anyhow, that's what little I've got so far. I'm hoping to attack this in a more concrete manner, with actual licenses and writing after release. - David Nusinow