Though I was not at the meeting, I have read (not studied!) Michael's
recent two papers at
http://www.prism.gatech.edu/~mm284/ and I think,
with Stevan, that he is looking at the wrong parameter.
OA is needed because it is the only way to achieve the social function
of scientific publication: the increase in scientific knowledge gained
by making the material available to all potential users without barriers.
Even if (to exaggerate a little) we could make all journals available to
90% of the users at half the cost of OA, we should not do so. It does not
matter whether OA is the cheapest way to publish. It is only necessary
that it can be published with the available resources. Subsequently,
having achieved OA, we may want to optimize for cost because that
increases the money available for other things (in the present context,
perhaps for buying books.)
It is obvious that OA journals can be published with the available
resources, because the needed resources are the same as the current
system. We need not -- we should not -- consider whether we could publish
non-OA journals at even less cost, as this would not achieve the purpose
for which journals are published. It is equally obvious that Stevan's
preferred "green" OA can be published within the available resources,
because they are only trivially greater than now. (That is, less than
the annual increased cost of journals under the present model. If we
can afford next year's journals, we can afford this year's OA.)
It is possible that OA journals will be less expensive, because of the
lack of need for access controls; so much the better. It is possible that
green OA will save money, because perhaps the necessary quality control
function can be achieved without quite as many low quality journals;
again, so much the better. (Personally, I think both results very likely,
but we will soon find out for sure.)
I am not an economist, so I may not be properly understanding all the
implications of Michael's analysis. I would find it important to know
whether it demonstrates that OA journals and green OA can or cannot
coexist. The practical implications are obvious, and this seems
beyond intuition and in need of true analysis. If we must now choose,
we will decide not on the basis of which is cheaper, but which would
appear to give better OA. I would be happier if they could coexist,
so we could find out by experiment, but perhaps that is not the case.
Otherwise, the only place I find Michael's discussion relevant to OA
is that it seems to indicate that OA and non-OA journals cannot coexist
(if I understand it rightly). If true, this would be a very significant
result, and one that immediately affects policy, because it would
indicate that we must plan for a very rapid change.
Dr. David Goodman
Associate Professor
Palmer School of Library and Information Science
Long Island University
dgoodman_at_liu.edu
Received on Sat Dec 11 2004 - 17:13:23 GMT