I received an email from BMC Medical Informatics asking me to review a resubmission of an article I'd reviewed earlier. So far, no problem. I took a look at my original comments, the revised article, and the authors' responses. Still no problem, as I looked for their responses to my comments. Oh, there it is: "Re the comments made by Andrew Gelman . . ."
Hey! I thought referee reports were supposed to be anonymous! It is a good thing that I liked the paper, otherwise I would've made some quick enemies without even trying!!
Maybe, like Leibniz when he saw an anonymous monograph by Newton, they "recognised the lion by his claw" :-)
Maybe if you didn't advertise your use of boilerplate text in your reviews, you would be harder to spot.
You probably told them that all their tables should be graphs. :-)
I review quite a lot for the BMC journals, and I thought they made it pretty clear that the reviews aren't anonymous, unless you specifically request it to be.
http://www.biomedcentral.com/info/about/peerreview/
I think it says this when you fill in the web form too. (Although I'm not 100% sure, and if you don't fill in the form, because you email it, obviously that doesn't happen.) I have once explicitly asked for anonymity, and it was given.
The reviews are also published on the website for all to see, afterwards, for example: http://www.biomedcentral.com/1471-2288/6/48/prepub
I do *quite* like the idea of this - I think it makes reviewers a bit more likely to consider what they are saying, and how they are saying it.
Jeremy seems to have hit upon the answer. As usual, I didn't read the fine print. I don't mind open reviews; I'd just like to know they're open when I write them!