Since the start of The Cost of Knowledge website requesting support in the form of signatures from the academic community to boycott Elsevier in various degrees, I have taken more interest in academic publishing. Elsevier happens to be currently placed in the limelight by dissatisfied academics as a prime example of unfair practices such as journal bundling and exorbitant prices and The Cost of Knowledge site already has 12532 signatures at the time of this post.
The urge to vanquish the likes of Elsevier and others has lead to the open-access publishing model, which has been around for some time. I decided to take a closer look at open-access, being motivated by the recent announcement of a new set of author-pays open-access math journals by Cambridge University Press: "Forum of Mathematics", which consists of two tiers: Pi and Sigma, the former for the high-impact papers that potentially interest many mathematicians, the latter for everything else that doesn't get rejected. These journals are expected start their official acceptance of submissions on October 1st, 2012.
These journals will have their papers available online for free under the Creative Commons CC-BY license, and they will be free to publish in for three years. After these three years, it is expected that publishing will cost the author to pay 750 from grants or departmental funds, unless the author can demonstrate that he or she does not have access to these funds. There have been extensive discussions on the blogs of Terence Tao and Timothy Gowers, and these discussions have surprised me with the amount of negativity towards the new journals.
In this post I shall explain these open-access journals in a bit more detail, and explain why I think such ventures can only be good for mathematics.
What is The Forum of Mathematics, Exactly?
(Of course, one should visit the official site of the journal for definitive information.)
These journals are open-access in the sense that anyone with an internet connection can obtain these papers without additional cost, and the papers will be licensed under the Creative Commons CC-BY license. Among many things, as long as the attribution to the original author is made, this allows any individual or organisation to keep a copy of these papers and offer them for redistribution, so that I myself could host all the papers from these journals on my own website with no possibility of legal action against me. This license also allows for commercial use and modification of the licensed work.
Being open-access of this type, naturally any individual or organization wouldn't need a subscription to access the journal. This also means that, except for people inclined to buy a compendium or t-shirt with a "Forum of Mathematics" logo (perhaps?), the traditional revenue stream is not available. In any case, if anyone is to be paid doing this, then the money must come from somewhere. The journal already has a sizeable monetary backing, so that for three years having a paper published in it will be free.
But, unless further funding is procured, after this 3-year period, Cambridge University Press will ask about 750 dollars for publishing in the journal (although submission will still be free).
This 750 dollar fee for publishing one article has met with a fair amount of opposition. Before I go on and incite further vitriol, I should point where the money should come from: mathematicians would pay this fee from departmental funds or grant funds set aside or requested for such purposes. Should the popularity of open-access increase, a section for publishing money would not be uncommon on future grant proposals, and funding agencies should get used to this. After all, considering how much more expensive every other aspect of research is, adding a bit extra for the publication should be tolerable.
Certainly the money is not intended to come out our own pockets, and indeed the Forum of Mathematics has a list of countries such that the mathematicians in which would be exempt from the fee automatically. Furthermore, mathematicians outside these countries who could demonstrate a lack of ability to pay the fee would also be exempted. Hopefully getting an exemption will be easy and painless for authors without the access to funds, such as many researchers in continental Europe for instance.
I definitely believe that a fair question to ask is: is this amount fair and only a nominal amount to keep paid personnel at a reasonable salary for the function of this journal?. No organisation ought to be above scrutiny, including Cambridge University Press, and as a community we definitely should keep tabs as close as possible to prevent any kind of corruption. This does mean making decisions wisely, and doing some research before submitting a paper to this journal. However, this also applies to any publishing house, or any organisation with which one enters into a contract. Although some already exist, hopefully more open-access journals in mathematics are created soon so that we have an environment of healthy competition to ameliorate the effects of any possible monopoly on open-source mathematical publishing.
To me, these conditions seem fair, especially since the Forum of Mathematics has a monopoly on publishing papers, and there are plenty of traditional publishing houses to choose from. Furthermore, even if the Forum of Mathematics is widely successful and helps the already existing open-access movement to gain significant market share, this surely will happen over a period of time long enough to allow most mathematicians to adjust. Some comments in the discussions on the blogs of Terence Tao and Tim Gowers have called this asking feel unethical, and surely I would agree if the Forum of Mathematics were the only journal and the money had to come out of our own pockets, but it seems every effort is being made to make this not the case.
What About Completely Free?
Yes, free is good. In fact, there are several journals in mathematics already that are free to publish in and free to read. Here "free" refers to free-as-in-cost, to borrow a term from software idealists, which means not having to pay any assets except the nominal price of finding an internet connection. An example that I know of is Homology, Homotopy, and Applications
To some extent, the editing and formatting have been offloaded to volunteers with these journals, so that neither the authors nor the readers have to pay anything to keep these journals running, and these will run as long as there is always a group of volunteers willing to do the work. In some sense, this is the most efficient setup per unit of work done, but it runs only as long as there are volunteers.
It's hard to say if there would be enough volunteers should the appropriate organisational steps take place, and whether such a diversion of effort would be worth it to flood the market with free open-access journals of high quality. If such an effort were to take place, as much of the administrative overhead would have to be covered by computers as possible. An example of such an attempt is the free and open source Open Journal Systems software. Although I have not used this particular program, it is free to run and since most departments already have websites and webservers, this type of solution would be free to implement modulo some volunteer who knows about installing software on a web server (which is not difficult).
Ultimately, if there are enough people willing to spare the time, then there is no reason we cannot eliminate costs altogether. Furthermore, as open-access becomes more common, accepted, and perhaps even viewed as the only proper thing to do, then universities may start to fork out funds specifically to hire administrative staff. Personally, I'm hoping for a world in which universities rabidly compete with each other for having the best open-access journals, much as universities compete now for having the nicest-looking campuses: pristinely formatted research perhaps being showed as proudly in the future as that 3.2 million dollar garden is today. Here's hoping?
Leaving the Publishing House
Perhaps I am an idealistic graduate student, but I feel that the existence of author-paid (from grants, with waivers for those without the funds) in mathematics is a good thing along side free journals. This type of journal offers the ability for authors to find a peer-reviewed, publishing platform that come with paid administrative and editorial staff that needs only a one-time fee, so that the article can be made freely available as long as the internet is around. Even if the journal perishes, the papers will still be available, and I for one intend to make a mirror, which is fine because of the CC-BY license.
The monetary burden placed upon the authors takes care of the copy-editing, and keeping staff so that the journal can function. If it is true that this author-funded part is not really necessary (i.e. that the paid staff is not necessary) then I expect a wealth of free-to-publish open-access journals to sprout up in protest; if no such publishing outlet appears, then in some sense this fee is necessary, at least if one subscribes to the belief that the best thing to do in terms of publishing our research is to make it available to anyone using the most efficient and least costly method available to others, and that we must reach this goal as a community.