The True Cost of the Essentials

Stevan Harnad harnad at ecs.soton.ac.uk
Thu Jul 24 22:55:05 EST 2003


Michael Kurtz is an astrophysicist and author of:
http://cfa-www.harvard.edu/~kurtz/jasis-abstract.html

Astrophysics is unusual in 3 respects: 

    (1) It has a small, closed, self-contained journal literature,
    with cross-citing restricted to those core journals.

    (2) Virtually all astrophysicists are at institutions that can afford
    the toll-access to all of those core journals.

    (3) Hence astrophysicists have already enjoyed the full benefits of
    free online access for many years now.

Michael Kurtz wrote:

> I have been reading the discussion of the "True Cost of the
> Essentials;" it seems to me the discussion is quibbling over pennies
> when there are dollars lying around.

Unfortunately, unlike astrophysics, most other disciplines do *not*
have a closed, self-contained journal literature to all of which their
institutions can afford the toll-access. There are 20,000 peer-reviewed
journals across all disciplines worldwide, and no institution can afford
toll-access to more than a shrinking fraction of them. That is why there
is a library serials crisis, and it also why no discipline other than
astrophysics is yet enjoying the full benefits of free online access.
http://www.arl.org/stats/index.html

> Basicly the entire cost of the journals is tiny compared with the
> efficiencies gained by having full electronic access to the literature
> in a discipline.  In astronomy, where essentially every professional
> astronomer has had total electronic access to the entire journal
> literature for five or six years, the value of that access, in terms
> of increased efficiencies of research, is about twenty times the total
> production cost of the core journals(*).

That is no doubt true. But the fact is that in other disciplines most
articles are not accessible to most of their would-be users because
toll-access is nowhere near being universally affordable.

> Given this huge difference issues concerning methods to reduce the
> production cost of the journals, or to redistribute these costs,
> seem of secondary importance.

It's hard to see why you would think this was so, since the fact is that
currently most of it is simply unaffordable -- to *any* institution,
let alone most or all institutions.

But note that I am not talking about cost-cutting either: Let us call
those researchers who are at institutions that can access some or
all of their relevant journal literature the "Haves" and let's call
those researchers who cannot the "Have-Nots." (Note that, apart from
astrophysics, every other discipline not only has a majority of Have-Nots,
but that even the Haves are Have-Nots for those of the journals in their
field for which their institutions cannot afford the access-tolls.

Well, self-archiving is a solution for the Have-Nots: Let the Haves
continue to access what they can access via their institutional
toll-access but let every author (except astrophysicists!) self-archive
their own journal papers too, in order to supplement toll-access with
open-access for all their would-be users among the Have-Nots.

This does not entail any cost-cutting or restructuring by journals
(though it might eventual lead to it). It merely gives every researcher,
in every field, the benefits that astrophysicists already enjoy today.

> It is likely true that improving the publication process will be of
> far greater benefit to the progress of research than any restructuring
> of the financial arrangements so that those who currently can't or
> won't afford access to the literature can get it.

I don't know what you mean by improving the publication process (if you
don't mean restructuring and cost-cutting). At the moment, universal
access, as in astrophysics, is simply not affordable to other
disciplines. Their literature is too big and diverse, and their
researchers are not all at rich universities.

Hence it is not the "publication process" that needs improvement (what
improvement?) but the *access to its product*: the refereed journal
articles that currently sit unaffordably behind access-denying
toll-barriers for the Have-Nots. The remedy is for the *authors* of
all those inaccessible articles (i.e., all articles other than those in
astrophysics) to maximize their uptake, usage and impact, by making them
openly accessible to the Have-Nots -- by self-archiving them.
http://www.eprints.org/self-faq/

> A good example of the type of improvement possible is the de facto
> collaboration between the physics journals and the ArXiv.  While this
> has the pleasant side benefit that papers can be read without charge,
> the principal benefit of this collaboration is that the rate of
> information diffusion between active researchers is substantially
> increased(**).  The rate of discovery must go up as a result.  The
> value to society of a 1% increase in the rate of discovery (which
> would mean that we would know twice as much new stuff after 70 years
> as otherwise) is so great as to be uncalculable.

All true, but you are speaking here about that subset of physicists
who already *do* self-archive! The objective in this Forum is to get
not only all the rest of physics, mathematics and computer science to
do so too, but all the other disciplines as well!

There is no point speculating about why those who already do it do
it (whether to get their pre-refereeing preprints or their refereed
postprints out faster and further): they are *doing* it, and this is
about getting those who are not yet doing it, doing it!

> (*) The value of increased efficiency for the electronic astronomy
> library is calculated in section 9 of my recent paper for JASIST
> (http://cfa-www.harvard.edu/~kurtz/jasis-abstract.html). The cost of
> the core journals comes from assuming the cost of the 6,000 journal
> articles is twice the cost of the 3,000 published by the Astrophysical
> Journal and the Astronomical Journal (figures from the AAS annual
> budget report).

I certainly don't dispute the great value of open access to any and
every discipline! But it would take some very adroit calculations
indeed to demonstrate how all 2,000,000 annual articles published in
the planet's 20,000 refereed journals could be made affordable to all
of the planet's Have-Nots! The actual amount (in collective tolls,
pooled from those Have-institutions that can afford the tolls) that
the planet's Haves currently pay (in exchange for access only to the
Haves) is $1500-$5000+ per article. Let's take even the lowest end
(without cost-cutting): 2,000,000 articles annually, times $1500 per
article. You are imagining an orchestrated redistribution of the total
$3,000,000,000 toll-fee among all the Haves and Have-Nots so that every
research institution on the planet becomes a Have for them all.

Now there are some toll-access publishers (and no doubt producers of many
other products as well) who do dream about such a global oligopoly --
a collective site-license for the entire planet -- where each year they
name their price and the whole planet forks up, and then everyone has
"free" access.
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/0436.html
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/1525.html
But even my own scanty knowledge of economics and human nature immediately
suggests that if redistributing costs were this simple there would be
no starvation, disease or war on our planet!

> (**) See Tim Brody's plot at
> http://citebase.eprints.org/analysis/citation_latency.png

Yes, data like these keep demonstrating the benefits of open access.
But what's needed, once the scholarly cavalry have thus been led to the
waters of self-archiving, is something to get them all to stoop to drink!

http://www.ecs.soton.ac.uk/~harnad/Temp/self-archiving.ppt

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 & 03):

    http://amsci-forum.amsci.org/archives/september98-forum.html
                            or
    http://www.cogsci.soton.ac.uk/~harnad/Hypermail/Amsci/index.html

Discussion can be posted to: september98-forum at amsci-forum.amsci.org 





More information about the Jrnlnote mailing list