A minor celebration today…I’ve finally had my first paper accepted for publication. As a measure of how long and convoluted this process has been, consider the fact that this is the very same* paper that I discussed in this post. That was written in October 2010, and was titled “A year’s work, lessons learnt”. Which means, according to my calculations, that the time between beginning work on this project, and actually getting something published has been over 2 years!
Of course, most of this time was not spent actually doing anything related to that particular paper. In fact, the majority of the time was spent waiting for referees to get round to reading the thing. Actually “waiting” is the wrong word, as I have come to realise that the best strategy when submitting papers to journals is not to wait, but to completely put it out of your mind (unfortunately this doesn’t help when you then have to revise it months later), and perhaps set some kind of reminder to get in touch with the editor one year in the future and ask exactly what is going on. I currently have two other papers “under review”, one of which has been “with editor” (I assume this to mean that the editor hasn’t got around to actually looking at it, let alone passing it to a referee) since May, and the other which, perhaps thankfully, I have no way of knowing what is happening with.
There is intermittent hand-wringing about the peer-review system in mathematical circles, and in academia in general. Like exams, and job interviews, it seems to be grudgingly accepted to be the least bad form of evaluation. Recently Timothy Gowers raised the possibility of an alternative system on his blog, which led to much fevered debate (I have just noted that I am at least the the seventh blog to have linked to that particular post, so it is safe to assume the debate sparked by it stretches much further than that particular lengthy list of comments!).
I am always aware of the non-mathematicians – and non-academics in general – that might be reading this blog; so briefly, this is how the current system works: you have some original idea, and write a paper about it. You submit your paper to an appropriate journal. Having ascertained to their satisfaction that you are not a crank/crackpot/charlatan, the editors of that journal identify those among their list of potential referees who are most likely to know your subject well enough to understand what you are talking about, and pass it on to them. These referees read your paper as sceptically as they can, and write a review. This review includes, crucially, their opinion as to whether or not the paper should be published, and if so, what should first be changed. The editor makes a decision based on these reviews, which is passed back to you. You then either make the recommended changes/argue your case, try another journal, or give up entirely, depending on what exactly the judgement has been.
In theory, this sounds like a very sensible and rigorous way of doing things: objective evaluation by impartial experts. However, consider the following statistics (lifted from this site about publishing in economics journals, but which I think apply fairly broadly):
1. The average wait for an acceptance decision is 3 years (on the other hand, the average wait for a rejection is 6 to 8 months. At least you are relatively swiftly put out of your misery…)
2. Assuming the average acceptance rate of “good” journals to be 15%, you need to have 7 papers under review in such journals at all times in order to have one paper accepted per year in a “good” journal.
3. If you want to have 10 papers published in the first 5 years of your career, then you need to have about 12 papers under review at all times
Pretty daunting! Especially for someone who, like me, is taking the first few tentative steps onto the academic career ladder. For established mathematicians, it doesn’t really matter too much how long a paper takes to get published (within reason of course), as they are relatively secure in their positions. However, I am currently applying for post-doctoral fellowships, and the length of a publications list is one of the most obvious ways to judge the quality of such applications. But how is one possibly expected to have such a list at all if a paper takes longer to get published than it takes to do a PhD?
Incidentally, I was once told that part of the current problem with the peer-review system is the fact that there are too many people submitting papers these days, and not enough established referees. It is amazing how many of the world’s problems come down, basically, to over-population…
Whatever the cause of the problems, time-scales like these seem quite anachronistic in the internet age; clearly an alternative system would be desirable. And indeed, there is at least the first murmurings of one with arXiv.org. This is basically a preprint repository, where people post their papers while they wait for the wheels of peer-review to grind their slow and rusty way to a conclusion. It enables people to share their results with the wider academic community immediately and effectively, and within a few years, it seems to have become an integral part of scientific life. However, a site like this is obviously no replacement for peer-reviewed journals (seeing a “proof” of the Riemann Hypothesis posted every few weeks or so serves as a handy reminder to be very sceptical of any new research that hasn’t yet been reviewed).
Basically what Gowers was proposing was a kind of extension of arXiv…you would upload your paper to some website, and it would then be reviewed by other people in your field. The lengthy comments and debates which arose from his post were mainly about the details: what people’s motivation would be to review papers, how easy it would be to manipulate the system, what level of anonymity would be required, and various other seemingly small, but very important details.
Interestingly, it seems as though the publishers of journals are themselves expecting some kind of seismic shift in the way these things are done. Upon confirming the proofs of our paper, I was offered the choice of various extras and options. The price list was as follows:
- To make the paper “Open Access” (that is, to basically buy the copyright from the journal): €2000
- To have figures printed in colour: €950
- To order extra offprints (aside from the measly one free copy you are allocated as author): €200 for 25 copies (same price for electronic copies!)
- To order a poster of your article: €50
I could go on, but you get the point. I have never seen such a blatant attempt to rip a person off outside of my spam folder. When I complained about this to my supervisor, he opined that, because the future is so uncertain for these publishers, they are trying to make as much money as possible now while they can.
Ironically, if they hadn’t set the price quite so high (how do you possibly justify charging €8 for a .pdf? The mind boggles) I would have actually ordered some offprints, as various members of my family would love to have some hard evidence that my years of university education have produced something tangible! They’ll just have to take my word for it instead.
* I say “same”, but actually the paper that was finally accepted bears little resemblance to that one! (The final version is here if you are interested). Same in the sense that a river is always the same river I suppose.
