1. Agreed this is a problem, but it's not unsolvable. A pilot project of peer review as a service could ferret out these issues and come up with solutions. Her

1. Agreed this is a problem, but it's not unsolvable. A pilot project of peer review as a service could ferret out these issues and come up with solutions. Here's a couple of ideas: (a) let authors give feedback on reviews for editors to consider when assigning this reviewer to future manuscripts, (b) require public disclosure of who reviewed the paper (this presents other problems, but should at least be considered), (c) editors can bring on 4th or 5th reviewers as needed if authors/editors/reviewers feel it is warranted (after seeing all 3 initial reviews) 2. Time is worth more than money. Remember it's not only the PI but also the postdocs and grad students who are not productive while they work on review/resubmission related work. I bet many labs would be glad to pay this amount of money for a single review process. The goal would be for single-review-per-manuscript to be able to replace journal reviews in most cases. Sure, some top flight journals may require additional review--and that's fine--but those should eventually be outliers, not the norm. I don't agree about the perception issue for a reputable source which is providing a tangible product. The authors end up with a peer reviewed manuscript that is citable. Call it biorXivGold or something like that and you're done. ^^ This red team/blue team (peer review) on this idea is a microcosm of the best part of science :)