Well, the problem is that they are, in fact, a serious part of scientific discourse. Peer-reviewed publications are the output; the process of science also requ

Well, the problem is that they are, in fact, a serious part of scientific discourse. Peer-reviewed publications are the output; the process of science also requires conversations in offices and labs and conferences, preprints, letters between researchers, etc. - and always has required it. You need to figure out what's interesting enough to research, what angles you may not have thought of yet, what ideas seem fruitful but lack someone pursuing them, etc. Blog comments (and mailing lists and newsgroups and Twitter) are a new way of accomplishing this at a larger scale, but they're not fundamentally anti-scientific - the conversations at Solvay weren't peer-reviewed either. The major claim of the article is that Peters is pushing his ideas in that realm - not the realm of peer-reviewed articles - and taking mindshare away from legitimate and worthwhile ideas.