Should theory conferences have explicit anti-harassment policies?
Unfortunately, I've heard (third-hand) of enough instances of the behaviors described by this policy at some of the research conferences I've gone to, to make me think it might be a good idea that they also institute policies like this. But I've tried looking around at the web sites of some recent theory conferences and I don't see any such thing there.
As for the likely response of "we don't need a policy because the victims can just take action themselves, and if they don't then it wasn't a serious enough problem to require an official policy", see this other blog post for why that sort of blame-the-victim non-policy is inadequate.
Update: Andy D. adds his support and provides some good explanations for why it could be helpful and wouldn't likely have much of a downside.