I'm a fan of consensus decision-making processes and will generally - but not always - advocate for their use. I think that this is one of those "not always" cases. First, we do not run every single decision past the IETF, so I think that argument needs to be put to rest. Furthermore, we do not always allow the consensus decision to carry the day - for example, it's common to have broad support for undertaking a piece of work but not have that work go forward because of opposition from the I*. The answer to that seems to be that the position that prevails does so on technical merit, and I hope that's true (although I'm not convinced that it's always true). But, I think it's important to understand that when that happens, we're dismissing consensus views because we think that those views are wrong, not because there's consensus to do something different. In those cases, the I* are not simply respecting the consensus of the body, they're doing something different. Anyway, we're now in a situation in which a few people have expressed concern that an ombudsman could make bad situations worse. I think that's a valid concern - anybody who reads the Washington Post has probably seen an ombudsman make some really terrible decisions, and just recently we saw the IETF mailing list ombudsman make what I thought was a good decision on a just awful basis. But given that, my own feeling is that the benefit of having an anti-harassment policy outweighs the possible cost of having one which might occasionally lead to bad/incorrect outcomes. Someone else might feel differently, and they would not be wrong, they're just operating within a different framework and have different values. It's extremely difficult to come to consensus in that situation, particularly when you've got thousands of participants and a notable lack of interest on the part of entirely too many in saying "I don't agree, but I can live with that." And then there are issues around the establishment of bases for a decision. It is not unreasonable for people to ask whether or not there's an actual problem, and whether or not past incidents have been sufficiently severe to merit the creation of a policy. We've seen those requests. I think that it's very much in the interest of the people involved not to allow those to be relitigated in public. I think one of the things to understand about consensus decision- making is that it has notably poor scaling properties. I absolutely agree that we need to be better at managing how we reach decisions within a consensus framework, but I also think we need to be smarter about recognizing when it's the just plain wrong process to use. We let the IESG handle a lot of "management" decisions, and I'm good with letting them handle this one. Melinda