**2nd UPDATE**

My paper has been published in **Forum of Mathematics Sigma** (open the paper as a pdf file).

**UPDATE**

My paper, On the integral Hodge and Tate conjectures over a number field, has now — after minor revision — been accepted by **Forum of Mathematics Sigma**, and should be appearing shortly. On the web, lots of people seem to conflate the open access model with editorial slapdashery. As you’d expect from the editorial board, there was no sign of that in my experience with FOM. I received two serious and helpful referee reports. Among other helpful recommendations, one referee pointed me to a very relevant reference that I didn’t even know existed, and the other referee pointed out that I didn’t understand a formula that I’d thought I understood pretty well.

**Original from 19 December 2012**

The title of this post is, of course, an exaggeration: I already have some version of nearly all my papers on my department webpage and now diligently post new papers on the arXiv. What I mean is that I’ve submitted a paper to one of the new Open Access journals launched by Cambridge University Press, in my case **Forum of Mathematics Sigma**, where the algebraic geometry strand is edited by Sebastien Boucksom, Ravi Vakil, and Claire Voisin. Sigma has other strands — the nearby algebra strand is edited by Dennis Gaitsgory, Raphaël Rouquier, and Catharina Stroppel. (There is also a second journal, **Forum of Mathematics Pi**, for papers of broad interest.)

The journals are meant to investigate whether mainstream Open Access can be a large-scale solution in mathematics to the problem of the increasingly expensive subscription model, and there is meant to be no corner-cutting on editorial integrity or publishing standards. Since, to really put this to the test, there needs to be serious volume I thought I’d do my bit to send some their way.

But this means the author has to pay a big sum per article, on the other of £200, right? Isn’t this a lot even if it is covered by grants?

There’s a three-year period in which CUP subsidizes all the costs of the journals. Then, as you say, costs are covered by article-processing charges, which CUP says will be $750 for Forum. It is looking more likely — still not certain — that grant organizations and university libraries will be willing to fund these charges. Almost no one takes the position that these charges should be the personal responsibility of individual authors.

Then the issue becomes what is reasonable to pay. We don’t want the exorbitant charges the mathematical community is currently paying for publishing simply to move from the “Subscription” column to the “APC” column. Whether or not you think $750 is a big sum judged in isolation, it’s a lot less than commercial publishers are charging for APCs. Springer Open is quoting a range of $1665–$1905. We know that commercial subscription charges are unreasonable because we can point to what good non-profit journals charge. I think of the CUP journals as an effort to put a non-profit yardstick in place for APCs before the commercial publishers establish industry-standard charges (which they are going to set at a level that compensates them for lost subscription income — based on exorbitant subscription prices).

I’m willing to support this effort by putting my papers where my mouth is.

Sorry for my late reply, it is that time of the year, I guess.

My comment was no critique, I am just interested on how the new model would work. From what you say, do I understand that department libraries may pay the costs of open access publishing for the department members? (At least for selected journals, I guess). At the moment, do research grants rules allow to cover these costs? It seems like a more sustainable model for the future, providing that non-for-profit journals such as Journal of Topology become the norm, and that journals provide a non-open access option from those working in less generous departments or having no grants.

I didn’t take your comment as criticism, and it gave me a chance to write a bit more about my motivation in supporting this effort.

How OA is (or will be funded) in mathematics varies a lot by country, and within the US by university. The current landscape is complicated because it’s transitional. I believe that in Germany, when applying for a DFG grant in mathematics, one can build in additional money for APCs. By contrast, in the UK, where the government is pushing pretty vigorously towards OA, funding is being provided by block grants to universities (possibly just the 30 most research intensive). It is not determined how the universities will allocate this money. Cambridge University has a Open Access Transition Project; see http://www.openaccess.cam.ac.uk/. I would guess that other UK universities have something similar. In the US, there is some funding at some schools. One summary of information is available via SPARC at http://www.arl.org/sparc/openaccess/funds/.

I caution that what I say may not be correct or up to date (though it’s not made up). It’s certainly not complete. It takes me well outside my area of expertise.

I cannot reply to your last reply and I am afraid that if I reply to myself you may not read it. You were indeed right about the UK government pushing for open access. Look for instance at EPSRC policy: “all published research articles arising from EPSRC-sponsored research, and which are submitted for publication on or after 1st September 2011, must become available on an Open Access basis through any appropriate route”.

http://www.epsrc.ac.uk/about/infoaccess/Pages/roaccess.aspx

I wonder what this means exactly. Two scenarios:

-EPSRC postdoc (a mythological creature) publishes something. Now he has to do it in open access. For a start this may reduce the number of journals in which he can submit to. Fortunately for him, there are good ones such as CUP where he can do it. However if he was not careful to include this in his justification of resources, he may not have the funds to pay for it!

-EPSRC-funded student publishes something. OK, here there is no way his department is going to pay the £500+ that would cost to publish an article. What happens then?

Moreover, in the justification of resources for EPSRC funding (not that I’m applying, ahem) there is no mention of this item: http://www.epsrc.ac.uk/funding/apprev/preparing/Pages/jor.aspx

Either fellowships and policy department don’t talk to each other or the Government is pushing for open access but it is not willing to pay for it?

Since you seem to know a lot about this, I’d appreciate any comments you have in this issue. A few people I asked in my department seemed to oblivious to this policy.

Nevermind, I answered myself: http://www.epsrc.ac.uk/funding/managing/Pages/publicationfees.aspx

According the revised RCUK open access policy http://www.rcuk.ac.uk/documents/documents/RCUKOpenAccessPolicy.pdf:

“From 1st April 2013 the payment of APCs and other publication charges related to Research Council‐funded research are supported through RCUK OA block grants provided to eligible research organisations. APCs and other publication charges relating to peer‐reviewed research papers can no longer be included within research grant applications.”

The first Forum of Mathematics papers have been published:

Peter Scholze. p-Adic Hodge theory for rigid-analytic varieties.

Forum of Mathematics, Pi,1, e1 http://journals.cambridge.org/article_S2050508613000012Mihnea Popa and Christian Schnell. Generic vanishing theory via mixed Hodge modules.

Forum of Mathematics, Sigma,1, e1http://journals.cambridge.org/abstract_S2050509413000017