Recently I have been debating in my head the following question: Does the four page limit for papers in Physical Review Letters squash physics?
Benefits of the four page limit: (1) brevity enforces a focused article, (2) experiemental results can often be described in four pages, (3) you can tell when a paper is submitted to PRL on the preprint server by counting the number of pages.
Problems with the four page limit: (1) brevity means much is left out or compressed to near unreadability, (2) experimental techniques are rarely described in enough detail, (3) the compression to unreadibility means that general readership across the different sections of PRL, one of the supposed goals of the journal, is difficult if not impossible, (4) the papers are often light on citations since it is easier to cut citations to get to the page limits than to cut the content, (5) theories of any complexity are impossible to present in four pages without obmitting or skimming major portions of the work.
O brave new quantum world!
“General interest” is definitely not what I think about when I read PRLs. But part of this, I do think, is that it is very hard to write an article which is accessible to those outside of a specialty in four pages. I mean in four pages I can usually bring someone up to speed on the state of the art, but then to actually explain my results requires at least another few pages.
Heh, I’ve heard of this thing called work. I’ve been pulling 14-15 hour days the last week. And I’m not even an experiementalist, just a useless theorist.
I think a lot of the readability problem comes not from the format, but from the referees. If the people refereeing papers actually put some weight on the “general interest” criterion, PRL could be dramatically improved, and don’t get me started on the writing.
It’s a big hassle, though. I got a paper to referee once in which the authors changed variables between equations 3 and 4 without bothering to inform the reader of the change, or define the new variable, and when I called them on it, they got pissy with me. This is an error that would earn a failing grade in an undergraduate lab report, and yet somebody sent it to the top journal in the field, and got annoyed when it was pointed out. And that was just the most greivous of the many errors of explanation in the paper.
As a referee, I’m not paid to deal with this shit. I managed to get them to define their variables, kicking and screaming all the way, but the end result was only marginally more readable. I just didn’t see any reward in continuing to fight over it, though.
I may blog more about this over at my site, but I probably ought to do some actual work first.
Well there’s always 7-zip.
Jokes aside, perhaps authors can simply put up a more verbose or ‘uncut’ version of their article on the arxiv for people who are interested in the details? Or would that be too much work?
It would be easier if they started from a lengthy article from which they began to pare down to size. Then they could simply put up the original.
From an IM conversation with a fellow physicist last nite – “4 pages… for PRL??? Congratulations!!”. Was really amused cos I really didn’t know about this until I read your post as I am new to this.