The ever-growing number of submissions to many journals has necessarily increased the number of scientists serving as reviewers. Although the majority of these perform their duty honourably and provide valuable feedback to the authors, some produce bad or even damaging reviews, which may not be filtered by the editors.

I believe anonymity is important for the peer-review process, but some power could also be granted to the authors in order to balance the equation. The flexibility of online systems could be employed to establish a feedback mechanism that may help journals weed out rogue reviewers.

One can imagine a scenario in which all authors would be asked to complete an online questionnaire about the reviews of their manuscript. The questionnaires could be anonymous, but should allow the journal to cross-reference the feedback with the name of each reviewer. Once sufficient data have accumulated, the journal will be able to identify reviewers who are serial offenders and decide not to approach them again.

Gathering feedback from the authors and using that to improve the peer-review process is a simple way of humanizing an increasingly electronic process.