The theory biosphere is at “double blind review” once again: here, here, here, here and here. There are good arguments on both sides, but I fall on the side that opposes it. To be fair, I didn’t get to read the entire discussion, and I will therefore spare you my arguments. I do want to discuss a claim that seems to be made that taking the identity of an author into account in the review process is the definition of bias. If in bias we mean discrimination then I disagree. Being more forgiving to faults in the presentation of a student is in my opinion not discriminatory as there is a higher chance that the faults would be corrected when pointed out by the PC (and as we want to empower newcomers). Being more careful with an author that has a history of bugs is in my mind fair game as well. These and others are based on many real-life cases, My favorite was an anonymous submission of two related papers to the same conference that did not cross reference each other and were caught by chance. If these were two independent groups of authors the decision would have rightfully been different. I believe that we should educate, but trust the PC. On the other hand, we should give the PC members tools to protect themselves against their own unconscious biases and I therefore always supported methods that make the author names less prominent (say, by putting it only on the back of the paper or other ideas discussed by Boaz)
This is highly related to a paper (coauthored with Cynthia Dwork, Moritz Hardt, Toni Pitassi and Rich Zemel) that argues that fairness cannot be obtained through blindness but only through awareness.