Variations on open peer review - which are best?
In 2016 I surveyed mathematicians about attitudes to journals, and as part of that I asked what they thought about various nontraditional practices. The possible journal policies below each gathered support between 15% and 45%. Open peer review means that feedback on earlier versions of a published article is made public by the time the version of record is published by the journal. I am not talking here about post-publication review.
- Open peer review with signed (not anonymous) referee reports.
- Open peer review with opt-in possible by authors/referees.
- Open peer review with opt-out possible by authors/referees.
- Referee reports are anonymous but useful information from them is sometimes made public by editors.
I forgot to ask about this, which seems to be used by some journals such as SciPost Physics, and is stronger than option 4:
-
- Open peer review with anonymous referee reports.
Now I would like feedback from people here about which ones they would recommend, and any other variations that look promising. Currently I think option 5 seems good: referees should be known to the journal but readers don't need to know who they are. The history of a paper can be helpful to know. It would be even better if we had open review at each stage of the process. This might make people think twice before submitting inappropriate papers, or writing inappropriate reviews (maybe!). It would also allow for portability of reviews, so starting at a new journal does not require starting from scratch.
I am sure there are many variations and some better ideas out there.