14 September, 2009

Peer-review reviewed

Most of you have probably seen the recent peer-review survey, and the post in Nature blog about it. Well, some scientists are satisfied by referees of their papers, some are not.

What looks a bit weird to me is the "how to improve peer-review" part. Usually there are two ideas:
i) to make the referee's name open, and
ii) to make the review process double-blind, with both names of authors and reviewers hidden from each other.

Well, if we take "a spherical society in a vacuum", say, an ideal one with no politics involved in research, then the first point might probably work. But I don't understand how can one hide the authors' names: people are used to cite their own work, such as an experimental machine they have built or a code they have written. So, the authors will not be obvious only when submitting their first contribution to the field.

That is surprising that 76% of researchers are favoring the double blind system.


Daniel Lemire said...

Somehow, I can never get interested in "fixing" the current system. It is like trying to fix the newspaper industry.

It is much more interesting to consider drastically different models.

Lemeshko said...

Oh, absolutely, being a tenured professor this is definitely true. :)

Daniel Lemire said...

Ah. Actually, here is how I think you change things, no matter who you are:


Lemeshko said...

Thanks for a good post, Daniel.