> And a serious error-detection system needs resources, says Elson. “We cannot expect it to work for free.”

Posted in a blog by Nature, whose entire business is based on the work of unpaid reviewers trying to find errors as best they can, is hilarious.