In 1967 Nature adopted 'single-blind' peer review to evaluate manuscripts — a model that hides the identities of the reviewers, who often feel more at liberty to speak honestly and critically this way. Last year, Nature Geoscience and Nature Climate Change started trials of double-blind peer review, where the authors' identities are also hidden, and several journals now offer this facility. Nature Physics will be next.

Peer review is not perfect. Publishers are always thinking of ways to improve this process, and author anonymity may redress any perceived biases against gender, race, age or geographical location. Or perhaps not. During the trials, surveyed authors generally agreed that double-blind peer review was a good idea in principle, but ultimately, take-up was low, peaking at 20% for the two journals (Nature Nano. 9, 871–872; 2014).

Such short trials are inconclusive, but the impression is that double-blind peer review is a viable option, which is why all the Nature journals will offer it as an alternative from next week. Authors will have to opt in and prepare their manuscripts accordingly.

In physics, the popularity of the preprint archive would seem at odds with a double-blind peer review system, and authors may wish to speak about upcoming work at conferences. Ultimately, the choice will be up to the authors.