Why Conference Review Must End !

Exhibit A: Matt Welch's post on how PC deliberations happen.

Notes:

• Yes, I realize it's partly tongue-in-cheek. But it's not far from the truth !
• No, going to all-electronic meetings doesn't solve the problem. It merely replaces one set of group dynamics by another
• Yes, we can't hope to remove irrational biases in the review process. That's why all we can hope for is to force them to be exposed and questioned. A back-and-forth between author and reviewer can help do that. 
• And no, it's not true that "theory PCs are much better". 

I've been to NSF panels where the program manager does an excellent job of forcing people to state precisely what they mean by "interesting", "infeasible", "novel contribution" and other such weasel-words. When that happens, it's a bit easier to assess the contribution. One could imagine enlightened PC chairs doing this at paper review time, but there's really no time, given the number of papers that need to be processed in 2 days or so.