The double-open review model seems natural when treating code and data. This is perhaps an effect of the social norms in open-source-software communities. Open peer review of the manuscript itself still raises concerns, however. While there's some evidence that open reviews are of higher quality and more courteous \citep{Walsh_2000}, early career researchers may shy away from the idea of having to write a critical review of a senior researcher's work, fearing career retribution. A recent study of attitudes and experiences among researchers \citep{Ross_Hellauer_2017} found that while 60% of respondents believed open reports would increase review quality, they were less keen on open identities, and 74% believed reviewers should be allowed to opt in to making their identity open.