On Mon, 15 May 2000, Albert Henderson wrote:

>ah> on Fri, 12 May 2000 Stevan Harnad <harnad@COGLIT.ECS.SOTON.AC.UK> wrote:

> sh> It is important for sophisticates of this Forum to note that there is no
> sh> irony whatsoever in the fact that The Economist does not give away its
> sh> contents for free on the Web.
> sh>
> sh> Why should it? Its journalists write the articles for a fee; their
> sh> entirely valid objective is to sell, not to give away, their work.
> sh>
> sh> The WHOLE POINT of the initiative of freeing the refereed journal
> sh> literature is that this (trade) model does not fit that anomalous
> sh> literature, so fundamentally unlike everything else.
> sh>
> sh> Researchers are not journalists selling their words, they are scientists
> sh> and scholars reporting their findings. Their rewards do not come from
> sh> tolls charged for access to their texts; they come from accessing
> sh> and making an impact on the minds and the research of other researchers.
>
>ah> Not so. Researchers make an economic exchange valued
>ah> more than cash, for recognition and dissemination
>ah> services that will reach their intended audiences,
>ah> present and future.

This often-repeated positive correlation has NOTHING to do with
causation, indeed, if anything, the real causal relationship is
NEGATIVE:

Researchers report their research findings in refereed journals in
order to make an "impact" (let us call it) on research and researchers,
not (like all other authors) to sell their texts. It is that impact (if
any) that then brings them promotion, grants, prizes, renown.

It follows that anything that increases that impact is positive for
researchers, and anything that decreases that impact hinders is
negative.

Access-barriers (Subscription/Site-License/Pay-Per-View, S/L/P] decrease
impact; let us not debate that. It is incontestable that you can't have
an impact on anyone who can't get to your work.

That is the real causal picture. The reason there is nevertheless a
positive correlation between (1) appearing in a refereed journal and
(2) impact is also quite obvious:

In the Gutenberg (paper) Era, the only way (and hence, a fortiori, the
best way) to make an impact on research and researchers apart from
what a researcher could convey by word-of-mouth and writ-of-hand (1-on-1
letters) was through print-on-paper. And producing and disseminating
print-on-paper was expensive.

Hence if those inescapable expenses were to be met (so that the
research could be disseminated at all), researchers had to reluctantly
acquiesce to S/L/P access-barriers to meet them -- because (in the
Gutenberg Era) the negative effects of S/L/P access-barriers on impact
were out-weighed by the positive effects of paper dissemination itself
(compared to word-of-mouth or writ-of-hand).

Albert Henderson writes as if nothing had changed since those days. But
we are now in the PostGutenberg Era of Scholarly Skywriting. Research
reports can now be publicly disseminated much more widely than by
print-on-paper, and at virtually no cost at all, via online
open-archiving: http://www.openarchives.org/

In this new Era, anything that attempts to constrain this new form of
open access is in headlong conflict with impact, and hence with the
interests of research and researchers.

There is still an essential function that journal publishers perform for
researchers, and that is quality-control/certification QC/C (implementing
peer review). For it is not the dissemination of their raw findings that
researchers seek, and need for impact, but the dissemination of findings
that have been certified by the pertinent experts.

The dissemination can now be handled by the researchers, but the QC/C
cannot; individual researchers cannot police themselves. Hence the real
costs of implementing QC/C (implementing only, because referees referee
for free) still need to be covered. But the good news is that these
QC/C costs are only a fraction of S/L/P costs, so they can easily be
covered out of a portion of each institution's annual S/L/P savings.

The crucial difference, then, is that S/L/P costs are
reader-institution-end costs for a reader-institution-end PRODUCT (the
text), hence recovering them depends on erecting reader-access-barriers
(= impact-barriers), whereas QC/C costs are author-institution-end
costs for an author-institution-end SERVICE (QC/C), hence recovering
them does not depend on reader-access-barriers, but rather on
dismantling them.

The funds will be there to cover QC/C costs many times over once S/L/P
barriers are gone. Institutions will be happy to redirect this small
portion of their annual windfall savings from S/L/P cancellation to
cover all QC/C service charges for their publishing researchers,
because their researchers' impact is also their institutions' impact
(as reflected in citations, grant-income, prizes, renown: that's why
institutions reward them through salaries and promotion).

>ah> Publishers bring order out of
>ah> chaos, setting standards for quality and objectivity.

That is QC/C. No longer any need to hold the paper product hostage to
this service via S/L/P access barriers.

>ah> They channel information to the readers who may use it.

That is the second "C" in QC/C. Again, no necessary connection between
it an access barriers in the PostGutenberg Era.

>ah> Research papers are not ads. Nothing is "given away"
>ah> by either researcher or publisher.

They are most definitely given away by their refereed-researcher/authors (no
fee, no royalty), unlike all other authors.

They are indeed not given away by their publishers, but that is the
precise point under discussion here! There is a vast conflict of
interest in the PostGutenberg Era, for this (author)-give-away
literature. And the conflict all concerns access-barriers and potential
impact.

(And I did not say research papers ARE ads, but that they are more
LIKE ads than they are like the non-giveway literature.)

>ah> Thanks to libraries
>ah> and librarians, scientific discoveries and theories are
>ah> preserved and disseminated for the future, often long
>ah> after the authors and publishers have disappeared.

In the PostGutenberg Era we are now in, Networked Open Archives will do
all of that, just as long, and much better -- and without the
access-barriers, thank you very much.

> sh> The access-blocking tolls are hence working AGAINST these rewards, not
> sh> for them. (Charging for access to their research makes about as much
> sh> sense for researchers as charging for access to their ads would make
> sh> sense to the advertisers of commercial products.)
>ah>
>ah> Not so. Financial statistics indicate that access was
>ah> blocked by university managers. They manufactured the
>ah> "serials crisis" by cutting library spending and an
>ah> open season on publishers propaganda campaign to shift
>ah> the blame. Universities have been hoarding money at the
>ah> expense of knowledge assets for 30 years. The average net
>ah> profit of private research universities last year climbed
>ah> to about 25% of revenues.

I will not reply (again) to this oft-repeated conspiracy theory of
Albert's. (It's the old refrain "Spend More On Libraries" and all will
be well.)

I will just say that these PostGutenberg possibilities have nothing to
do with the reality or unreality of the "serials crisis."
Access-barriers are access-barriers, whether they are high or low. And
when there is no longer any need for them at all, there is no longer
any justification for them.

I have described self-archiving as "subversive" precisely because it
is likely to force journal publishers to scale down to the bare
essentials (i.e., QC/C service-provision), because readers prefer
the free-access, self-archived version of refereed final drafts to the
S/L/P alternatives. But as long there still exists a market for the
S/L/P version, let it continue to be sold; researchers' needs are
served by freeing the refereed literature online. How long the two
incarnations of the same literature (for-free and for-fee) co-exist
is anyone's guess, and certainly no concern of mind. (But the redirected
funds for covering QC/C service costs are always latent in the S/L/P
savings, if and when that market collapses.)

> sh> The access-blocking tolls are hence working AGAINST these rewards, not
> sh> for them. (Charging for access to their research makes about as much
> sh> sense for researchers as charging for access to their ads would make
> sh> sense to the advertisers of commercial products.)
>ah>
>ah> Not so. Starting with PHILOSOPHICAL TRANSACTIONS, which
>ah> was founded as a for-profit venture by Henry Oldenburg,
>ah> journal publishing has been a win-win arrangement for over
>ah> 300 years. It is widely supported by researchers and
>ah> considered an important source of financial support for
>ah> other activities that might include policy positions and
>ah> accreditation.

Ah me, back to Gutenberg! Those days are over, Albert. (And once the
inherent trade-off, and its needlessness in the PostGutenberg Era, is
made explicit to them, I am ready to bet that researchers are NOT
willing to have the potential impact of their curtailed in the service
of subsidizing other "good works" of their Learned Societies.)

> sh> In the papyrocentric era, such give-away authors had no choice but
> sh> to make the Faustian bargain (with Gutenberg), that in order to defray
> sh> the substantial expense of typesetting, printing and distribution, they
> sh> would reluctantly acquiesce to the levying of access tolls to recover
> sh> those costs -- knowing that if they did not acquiesce then there would be
> sh> nothing at all for researchers to access (beyond what they reported
> sh> orally or by writing one-on-one learned letters).
>ah>
>ah> Not so. The Faustian bargain was made when academic
>ah> senates gave up control of policy to administrators so
>ah> that faculty could be free to pursue intellectual goals.
>ah> Unfortunately, the quest for knowledge has been undermined
>ah> by the financial priorities and petty ambitions of the
>ah> new bureaucracy...

Ah me. Nolo contendere. I have deleted the rest of this irrelevant
conspiratorial speculation.

Forget about bureaucrats' petty ambitions for a moment and focus on an
objective that has face-validity: Researchers do research and they want
to share their results with other researchers, present and future,
freely. There is a way for them to do this now. So let's just do it.

--------------------------------------------------------------------
Stevan Harnad                     harnad@cogsci.soton.ac.uk
Professor of Cognitive Science    harnad@princeton.edu
Department of Electronics and     phone: +44 23-80 592-582
             Computer Science     fax:   +44 23-80 592-865
University of Southampton         http://www.cogsci.soton.ac.uk/~harnad/
Highfield, Southampton            http://www.princeton.edu/~harnad/
SO17 1BJ UNITED KINGDOM

NOTE: A complete archive of this ongoing discussion of providing free
access to the refereed journal literature is available at the American
Scientist September Forum (98 & 99 & 00):

    http://amsci-forum.amsci.org/archives/september98-forum.html

You may join the list at the site above.

Discussion can be posted to:

    september98-forum@amsci-forum.amsci.org