As things change, some interesting new business models are emerging. See http://peerJ.com, for example. It turns on its head the idea of subscribing to a learned society and getting the society's journal. Quite fascinating.
See also the new OA projects by CUP, (£500 publishing fee) and especially Tim Gowers' blog., that describes several key points clearly.
Some of my own thoughts
In August 2008 I did some brain-storming on this subject with my friends C. V. Radhakrishnan and Kaveh Bazargan. Here are some of my notes from those exchanges.
I was thinking about this problem that if the funding model changes from "reader pays" to "author pays", then not everyone could afford to publish. Independent scholars, or those at 3rd world universities, might be priced out of publication.
- An OA journal *must* consider itself free to publish good peer-reviewed research, whether or not there's funding. So there has to be a "let-out" or discretionary waiver clause in any statement about pricing.
- Would it be feasible for the journal to have a sliding scale of
charges that is directly keyed to the budget of the institution to which
the author belongs? University budgets should be publicly available
somewhere, shouldn't they? It might take a bit of work to track them
down, but it could be done. Or the authors could simply be asked to
provide that information. In any case, if a university has a big
endowment or annual budget, then their staff would be charged more to
publish in the journal, and v.v. Independent scholars would be free (?)
The general idea is that a scholar from Cambridge Univ. or the TIFR, Bombay, could be charged $500 to publish an article in our hypothetical OA journal, on the assumption that his department has a budget for this (an "article processing fee"). Whereas Prof. Shivaramakrishna from a Jnanamatha in Trichy, or a Dr Salvador from Havana Univ., could be permitted to publish at $30.
Or if direct keying is not easy to implement, there could at least be general funding bands: we could find somebody else's ranking, perhaps UNESCO, for national education budgets, or educational funding, and use those as bands for submission charges.
- Here's an important tweak. I think this kind of banded charging can be
thought of a bit like Google's AdSense advertising system. Basically,
the university is paying to have its name associated with the research
that is published. So at the top of the article it says,
University College London
and the University pays the journal $100 (or whatever) for document processing.
However, if I - as an author - choose, I can say instead,
and then there would be no charge for document processing.
It wouldn't matter to the journal whether it was actually true or not that DW was an "independent scholar". The point is, if there's no payment, then the university or research sponsor doesn't get its name mentioned, and is therefore not formally associated with the research. This treats the association of the university's name with the research rather like advertising or product placement.
Most research contracts require the academic to include acknowledgement in his publications of the source of the funding. "This research was carried out under grant 123456789 of the National Science Foundation". So if the NSF is making such a requirement, they have to pay for it to be done. It's quite like advertising.