On Thu, 12 Dec 2002, Arthur P. Smith wrote:
>sh> The following is a wild guess on my part (but, considering how
>sh> conservative was my estimate of the number of would-be users, I think
>sh> it's a well-buffered guess): I doubt that even 10% the planet's would-be
>sh> users have access to even 10% of that annual corpus today.
>
> In physics, at least, that's a wild underestimate - 10% of would-be
> users for the research literature
> in physics is about 20,000 people - our journals are accessible by over
> 100,000 researchers who use them actively (just from online usage
> statistics); I'm not sure what fraction of the literature we publish,
> but I believe it's in the neighborhood of 5-10%. So, assuming we're not
> a major exception in accessibility among physics publishers (and I know
> at least AIP and IOP's subscription numbers aren't that far behind
> ours), you're way off on this one. Maybe it's true in medical or other
> fields, but that level of access loss seems hard to believe.
APS journals are the top of the quality hierarchy in physics, so they are
predictably subscribed to, hence accessible to, the largest proportion
of the total population of would-be users in physics (which you
estimate as 200,000 worldwide). By your estimates, then, the top
10% of physics journals are now accessible to 50% of their would-be
users. Presumably it's downhill from there (for the rest of physics
journals, and the rest of the planet's 20,000 journals and their 2 million
annual articles, across all disciplines). I suspect the journal-access
percentages in physics are the best among all disciplines, not least
because physics is also the furthest along the road to open access through
self-archiving. (That can in principle take care of the other 50%, once
self-archiving spreads to all of physics -- though it is growing too
slowly even in physics and may take another decade at its present linear
growth rate:
http://arxiv.org/show_monthly_submissions ).
>sh> All that is still necessary from peer-reviewed journal publishers is
>sh> a *service,* namely, peer review, and we already know how much that
>sh> service alone would cost the planet: At the conservative estimate of
>sh> $500 per article, it would cost $10,000,000 annually.
>
> Uh, your math is way off there. The total would be $1 billion ($1000
> million for clarity). And your $500 is after a factor-of-three
> improvement in costs that isn't exactly available as yet.
Sorry about the 2-decimal error (jet-lag) but the corrected $1 billion
figure will do. I'm not sure what you meant about factor-of-three
improvement in costs, but if the estimates are correct this time, the
amount the planet is currently paying for the 20,000 peer-reviewed
journals (as a "value-added product," the text, rather than just the
peer-review service) is then (again, conservatively) around $4 billion
(and the punch-line is the same).
>sh> Now what do you
>sh> think the planet is paying now, annually, for those 20,000 journals
>sh> (collectively, in subscriptions, license, and pay-per-view, by those
>sh> institutions that can currently afford it)?
>
> I believe total current sales of scientific research journals is $5 - 10
> billion/year. And I believe that number should be sufficient for close
> to 100% access to 100%, if we could work out the business model
> properly. Which is what the large site license approach and differential
> pricing models are directed towards. Of course the total number will
> have to grow as research volume increases, even so.
Ok, let us raise the ante: Let the mark-up from the cost of peer-review
alone ($1 billion) to the cost of all the other value-added features
of the text-as-product be 400% - 900% instead of just the 300% I had
estimated. Either way, (1) I very much doubt that the planet can afford
much more in access-tolls than it already pays and (2) that still would
not provide anywhere near 100% access to 100% of this literature (20K
journals) for all of its would-be users. Nor is it clear why the planet
would want to keep paying such mark-ups for value-added features, when
it could have open access to the essentials (a peer-reviewed literature)
for a fraction of the current cost -- if researchers simply self-archived
all their peer-reviewed research output.
But the only way to test this is for researchers to provide open access
to the peer-reviewed essentials by self-archiving them now, and then we
can see how much of a market still remains for the add-ons as options.
And we would have the added comfort of knowing that if ever the market for
the value-added options was no longer able to make ends meet, there would
still be several times more than enough money to pay for the essentials
out of the savings.
>sh> If researchers don't have open-access now, because they have not done what
>sh> the authors of 200,000 papers in physics, 500,000 papers in computer
>sh> science, and who knows how many other authors have already done -- namely,
>sh> to provide immediate open access to their own peer-reviewed research
>sh> output by self-archiving it -- then they have only themselves, and
>sh> definitely not their publishers, to blame.
>
> Well... readers have authors to blame (as well as publishers)... they're
> not always the same people, you know.
Ah, but when it comes to the coin of the publish-or-perish realm, namely,
research impact, they *are* the same people! That's why I kept referring
to "would-be users" (rather than, say, "causal readers"). Whichever hat
they happen to be wearing, researchers cannot remain for long blind to
the basic causal connection between access and impact (especially once we
have explicitly quantified it for them scientometrically, and illustrated
their actual and potential cause/effect relations across of time).
http://citebase.eprints.org/
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/2373.html
The temptation to call this "reciprocal altruism" should be resisted.
"Self-archive unto others as ye would have them self-archive unto you"
is strictly a matter of mutual self-interest, because maximizing research
access is a necessary condition for maximizing research impact.
Stevan Harnad
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):
http://amsci-forum.amsci.org/archives/American-Scientist-Open-Access-Forum.html
or
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/index.html
Discussion can be posted to: american-scientist-open-access-forum_at_amsci.org
See also the Budapest Open Access Initiative:
http://www.soros.org/openaccess
the Free Online Scholarship Movement:
http://www.earlham.edu/~peters/fos/timeline.htm
the OAI site:
http://www.openarchives.org
and the free OAI institutional archiving software site:
http://www.eprints.org/
Received on Thu Dec 12 2002 - 19:52:03 GMT