For the first part of your comment I will just say that research does not work like this. However, that is interesting: > Maybe it's best to improve the proce

For the first part of your comment I will just say that research does not work like this. However, that is interesting: > Maybe it's best to improve the processes. In software we can ship something with a beta label and expect problems. Then we can iterate on the software to improve it. That's more or less an analogy of what people talking about open peer review are saying, and I think it is quite right.