Need for systematic scientometric analyses of open-access data
harnad at ecs.soton.ac.uk
Sat Dec 21 16:29:01 EST 2002
On Sat, 21 Dec 2002, Thomas Krichel wrote:
>sh> Now the immediate occasion for this discussion thread was the recent $9
>sh> million grant to the Public Library of Science for the founding of new
>sh> open-access journals (i.e., BOAI-2):
>sh> This is excellent news for open access
> But is it good news for scholarly communication? Probably not.
> They want $1500 per submission. We discussed that with the RePEc
> community. A library would have to cancel one of the expensive
> journals in our discipline for a year to fund one submission.
Thomas, you definitely have a point. But consider this:
(1) The Public Library of Science has a very specific strategy here --
a top-down rather than a bottom-up strategy: They are going into direct
competition with the highest quality/impact journals in the biomedical
hierarchy, rather than simply trying to convert weaker toll-access
journals into open-access ones (or start new low-level journals on
(2) The hope is that -- if the PLoS strategy is successful, and these new
high-level open-access journals successfully compete for the authorship
of their high-level toll-access competitors -- the this will start
a domino effect, from top-down (which is much easier than doing it
bottom-up), with the result that all (biomedical) journals will convert
(3) While most journals are still toll-access, this does indeed mean a
higher cost burden on authors and their institutions (and that is partly
why subsidies are available for those who cannot afford it). But once the
dominos begin to fall, institutions will begin to make windfall savings
from their diminishing toll-expenditures, and there will be more than
enough to pay the publication costs.
Again, though, this is all hypothetical. Any of these expectations may
fail to meet with success. We will have to try and see. And meanwhile,
let us not forget that a second strategy is still in place for those who
are not yet ready to submit their work to open-access journals (or who
do not yet have suitable open-access journals to submit it to), namely,
self-archiving their toll-access papers.
I would also like to make a prediction: With the help of self-archiving
(which will distribute the archiving burden across all the interoperable
institutional Eprint Archives) it will soon become apparent that the
only essential function of an open-access journal is implementing peer
review, which costs at most $500 per paper, which will be much more
affordable, especially once the dominos fall and institutions have at
least four times as much as that in annual windfall savings. Open-access
journals will duly cost-cut downsize to fit that sustainable niche.
> Using data from Ted Bergstrom, Bob Parks made a rough calculation
> that if a library took all the journals in Ted's list, which
> has many journals in economics and certainly the most expensive
> ones, it could fund 42 submissions with the money that it
> would save from cancelling all the subscriptions (assuming that
> it would buy all of them: no library does that). Now note
> that these are submissions, not accepted papers. If they
> have a high rejection rate, you burn all your money for
> your serial budget in trying to get into one of the
> two journals. None except the very well-funded will be able
> to publish there.
All good points, but not the right way to do the estimates, I think.
First, although I am ready to be corrected, I believe the $500
peer-review cost will prove to be per accepted paper, not per submitted
paper (although levying a much lower submission charge as well --
creditable toward acceptance if accepted -- might not be such a bad
idea, to discourage nuisance submissions wasting wasting many referees'
[freely given] time while a paper works its way down the quality hierarchy
until it finds the level that it should have submitted to in the first
place!). (I make no defense of the $1500 publication cost, except that
it may be necessary to test the BOAI-2 top-down strategy.)
So, with the (conservative) estimate of $500 per (accepted) paper
peer-review costs, this is the way that institutions need to do the
(1) What is the current annual number of peer-reviewed papers
published by researchers at your institution? Multiply by $500 and
call that P.
(2) What is the total annual expenditure of your institution in
toll-costs for peer-reviewed journals (subscription, site-license,
pay per view). Call that T.
Prediction: T >> P (probably about 3 or 4:1).
> Can anyone tell me how an organization can cash in $9 Million,
> over 5 years, and not be able to operate two, presumably
> online, journals with this money without charging a submission
> fee, for at least the time that the subsidy runs for?
Not quite fair. The PLoS plans to start further open-access journals out
of that grant too; and some of the funds are also going to subsidize
authors who cannot pay; and, as I said, competing for the very top
niche in the hierarchy is likely to be more costly initially.
But I do agree that if the PLoS strategy had been paired with an explicit
policy to off-load all the archiving onto the authors' institutional
Eprint Archives, that would not only have cut costs (from $1500 to $500),
but helped give more momentum to the self-archiving undercurrent that
(I believe) will hasten us all to the optimal/inevitable endstate,
which is universal open acess.
NOTE: A complete archive of the ongoing discussion of providing open
access to the peer-reviewed research literature online is available at
the American Scientist September Forum (98 & 99 & 00 & 01 & 02):
Discussion can be posted to: september98-forum at amsci-forum.amsci.org
See also the Budapest Open Access Initiative:
the Free Online Scholarship Movement:
the SPARC position paper on institutional repositories:
the OAI site:
and the free OAI institutional archiving software site:
More information about the Jrnlnote