Wednesday 15 October 2008

Peer review

Referee report dropped into my inbox this morning. Suffice to say our paper concerned a dimensionless object, let's call it O for object, which in the literature has a somewhat dubious, gauge dependent definition. So we gave a nice, sensible, gauge invariant definition for it.

The referee said "this is nonsense, they never explain what is wrong with the old units, and those units are certainly gauge invariant!".

Well. How to respond to that. Perhaps something like:

1. We explained exactly what was wrong with the old definitions of O.
2. O is dimensionless.
3. Sorry, in case that went over your head, O *doesn't have any units*. It is a pure number.
4. Units... gauge invariance? Sorry, what? No, really, what? Please, do explain. I'm waiting. No? No, I thought not. Go back to playing with your table top cold fusion. How's your website, by the way, www.my-own-theory- that-neutrinos-are-alive-and- cause-gravity-to-go-backwards.com ?

On a less facetious note, this example of, frankly, stupidity, raises an important point concerning peer review. This guy or gal now believes that I write rubbish papers. So I can kiss goodbye to any job for which this luddite holds any sway in the decision making process. Also, he is going to sit around with his chums, who presumably are at least partially connected to the field the paper is written for, and slag me and my physics off. My name is now dirt in some department, somewhere in the world (replace "some" with "another" if you like).

On the other hand, the peer anonymous review system means that I don't know who this crackpot is. I won't know now to avoid his papers, not apply for any jobs he might have, and I won't be able to laugh about him with my chums in conferences. Or get smashed during the conference dinner and tell him precisely what I think of his abilities.

Seriously -- this guy's incompetence is damaging my career. It isn't fair. And it's just not possible to get this point accross to the average journal editor, unless he happens to know you and that you aren't a cretin.

SIGH.

But what can we do? Our problem isn't the one they have in certain areas of pure maths, where there are fields so specialised that only six people in the world work on it, so everyone knows exactly who everyone else is and what they're doing. Our problem is that to make reviewing completely anonymous, we're going to have to either give up the arXiv (SHUDDER), make arXiv submission anonymous, or go back to the old ways of submitting to a journal and waiting until it's accepted before sending it around to people (meaning put it on the arXiv). Making arXiv submission anonymous seems the most sensible (until your paper is accepted), now that there's an endorsement system it should cut down on the crackpots. But is it really what we want?

There is another option. How important is journal publication these days? For the postdocs I know, it isn't -- what matters is the citation count on spires. It seems that the older chaps, who still hold the position of power, they are more likely to want to see that your papers are in good journals before they hire you. That's just the way things were done when they were coming up through the system.

Well, they'll be retiring soon... and if there are any jobs left in physics (STFC, I'm looking at you again), the young will be filling them up. People who communicate via e-mail and skype, not by waiting for the latest version of acta. physica. polonica. to be delivered by mountain pony. People who know if a paper is decent or not because they can download it and read it for themselves (or they can just count the cites and leap on the bloody bandwagon). Do they care what some random shadowy fool thinks of their paper when, if they have any sense, they'll give up any pretence of honesty and send their paper to a journal for which their chum is an editor and it will be accepted straight away.

Well. There we are.

Oh, should say, there was a second referee report which was overwhelmingly positive. Good show there.