I think for an open discussion it is important to know who is part of the discussion. As an author you are putting a paper out there, and you do your best to ma

I think for an open discussion it is important to know who is part of the discussion. As an author you are putting a paper out there, and you do your best to make sure the paper is original, relevant and correct. As a commenter, you should also try to do your best, and the best way to assure that in my opinion is to attache prestige not only to papers, but also to comments. This includes negative prestige for making bad comments. That's why I would say a real name policy is important.