Spend More on Libraries

From: Albert Henderson <NobleStation_at_compuserve.com> <harnad_at_cogsci.soton.ac.uk> <harnad_at_cogsci.soton.ac.uk>
Date: Tue, 11 May 1999 09:59:47 -0400

========== Following are my comments, in two parts, ======
========== exactly as submitted to NIH ==================

FROM: Albert Henderson, Editor, PUBLISHING RESEARCH
QUARTERLY <70244.1532_at_compuserve.com>

We share a vision of effective science using the rapid
communication features of information technology. We
differ when it comes to the means of reaching this goal.

My comments focus on

(A) problems inherent in the present rather radical proposal
and (B) solutions that I believe you have not considered

The complaint of NIH director Harold E. Varmus that,
"researchers spend hundreds of dollars of their NIH awards
on subscriptions to scientific journals," (1) reveals
serious defects in policy choices made over the last 30
years or so. Using grant money for subscriptions has always
been an option. My recollection is that universities, not
research grants, once paid for most subscriptions found in
offices and laboratories. Universities canceled most such
"duplicates" in the 1970s. In recent decades, over the
protests of faculty and faculty senates, universities
continued to cut library spending. They have been canceling
the last remaining copies of many journals. (2) Studies
indicate nevertheless that researchers use libraries more
than ever, with library borrowing rates sharply increased.
(3) It also appears that better financed researchers now
order their own copies of journals no longer found in the
library. Although they purchase these publications with
public money, they neither share them generally nor
maintain formal collections. In other words, local
colleagues may still have to order interlibrary photocopies
of articles that they identify via information services and
citations rather than browsing.

Vannevar Bush charged universities with the responsibility
to conserve knowledge as part of the government-academic
research partnership. (4) It was a Faustian bargain.
Instead of maintaining information produced and used by
government research programs, universities cut library
spending (and Federal agencies permitted it!). Universities
seek further relief from library costs, even in the robust
economy of the 1990s, while confessing that the imbalance
between library and research growth is a source of serious
problems. (5) The millions of dollars of subscriptions now
paid by research grants, described by Dr. Varmus to
Congress, represents the unloading of universities'
traditional responsibility onto researchers who have grants.
The cancelations also drove publishers' prices upward,
providing a foundation for denunciation and calls for new
solutions. Taking the prospect of "no library" another step,
provosts at CalTech and elsewhere propose that researchers
divert even more grant money to self-publishing their work.
(6) The present proposal falls directly into this trap. It
would make it easy for universities to justify the further
elimination of their subscriptions to advanced research
journals and information services by shifting the full
responsibility for conserving knowledge to the government.

The present proposal serves such financial goals without
realistically solving problems in dissemination and the
quality of research. I note generous use of the appeal of
"free" information. What is the cost to the taxpayer? How
much equipment, labor, time, and other resources will be
needed? How many articles a year will be served
electronically? What about standards and obsolescence? Who
will pay for equipment required for access? What will be the
impact on scientists and institutions who are not fully
wired? What will be the impact on the use of publications
produced only in paper, including the corpus of previously
published literature? Will NIH take responsibility for
digitizing that? What about copyrights? There are also
major questions of permanance being asked about the use of
fragile storage for journals often called "archival." How
will the E-Biomed proposal deal with the growth of science
-- now generating millions of articles a year and growing
exponentially? Is NIH prepared to face Congressional
challenges to technology that is far from perfect? I am
certain there are other good reasons that faculty who
quickly embraced email, bibliographic databases, mainframe
computing, and laptops are reluctant to join
administrators' undaunted support of support electronic

I believe Federal agencies were in error when they ended
studies of science communication in the mid-1970s. The
present proposal has no recent science or scientists to
provide a context for evaluation. It has only lobbies with
financial, rather than scientific, priorities and
enthusiasts who more often than not have no experience as
publishers. Many recommendations of pre-1977 studies meant
to improve dissemination. They were ignored, even after
such goals were explicitly adopted by the National Science
and Technology Policy, Organization, and Priorities Act of
1976. (Public Law 94-282; 42 USC 6601+) (7) The
incorporation of library fair use into the 1976 copyright
law apparently misled policy makers to the presumption that
photocopies would solve all dissemination problems.
Researchers obviously did not agree. They simply
circumvented the bureaucracy, its decimated library
collections, and the red tape of interlibrary photocopying.
They ordered their own subscriptions. As a promise of better
dissemination, photocopying backfired by justifying
substandard library collections. Twenty years later, it
still takes an average of two weeks to obtain an
interlibrary loan. (Many referees are asked to respond in
that time.) The present proposal echoes the unfulfilled
promise of the plain paper copier.

If NIH truly wishes to improve dissemination, it must have
the courage to ask how the unarticulated policy that bars
its evaluation of its researchers' information resources --
mainly academic libraries -- interferes with its mission and
its performance. The systemic impoverishment of academic
libraries created a bottleneck in communications and a
hostile environment for investment of resources. (8) One of
the worst effects has been publishers' curtailment of
editorial coverage of indexes. (9) It is questionable
whether one can prepare a comprehensive review relying on
Biosis, Index Medicus, Agricola, and other bibliographic
databases. (10) It was the forbidding commercial nature of
the library market that led the American Physical Society
and many others to avoid aggressive investments in
electronic publishing. (11) This opened the door for Paul
Ginsparg's experiment in welfare for rather rich scientists
and universities. It also provides the foundation for
lobbying that spurred the E-Biomed idea.

NIH's earlier adventure with preprints, in the 1960s, ended
in disaster. (12) Plunging into commercial conflict now
with thousands of privately-owned science journals and
indexes will end similarly. It can only aggravate the
present defect in policy.

A proposal with striking similarities to E-Biomed was made
to the Senate committee on government operations shortly
after Sputnik. Modeled on the Soviets' unified information
system, it was quickly rejected. (sorry I don't have a
reference handy) Policy focused then on improving libraries
and encouraging the development of electronic bibliographic
databases by organizations with appropriate experience. The
1960s became a period remembered for its outstanding
accomplishments in research and development as well as for
the development of electronic bibliographic services,
translation journals, author prepared cold-type publication,
and the science citation index -- all developed in the
private sector, many by entrepreneurs.

First, recalling this highly successful solution to the
Sputnik situation, I suggest another path you must consider.
The tenets of Federal policy promise a fair share of the
indirect costs of research. (13) Government science
accounts for sixty per cent of academic research spending.
Yet grant overhead allowances support only about ten per
cent of library spending at research universities -- none
of it tied to purchase of journals. If libraries matched
the growth of research, neither NIH nor its scientists
would complain about dissemination. Backlogs of accepted
articles would be lower and information technology would
be more fully exploited. NIH would do well to investigate
how better-financed and systematically evaluated academic
libraries would further its interest in rapid dissemination.
A more hospitable economic environment -- supporting robust
academic libraries, open to the public -- would encourage
private investments in technology, synthesis, indexing, and

Second, as a government agency, NIH also has a perfect right
and a moral responsibility to qualify research contractors'
equipment and resources. The present blind spot has done
more to promote mediocrity at research universities than
any other policy. I question whether researchers at an
institution with spotty local resources can prepare
effective proposals and provide reliable peer review.

Finally, there are good reasons that there are thousands of
journals rather than a single unified "government" system
envisioned under E-Biomed. Solutions to dissemination
should evolve as they have since Henry Oldenberg launched
the Philosophical Transactions, aiming for a commercial
profit of 40 pounds a year. That would leave the economic
risks and opportunities in the private sector, where they


Albert Henderson, Editor
Publishing Research Quarterly
POB 2423 Noble Station
Bridgeport CT 06608-0423
203-380-0021 fax 203-380-1703 messages 203-367-1555
email: 70244.1532_at_compuserve.com


1. Paulette Walker Campbell. 1999. NIH may use the internet
to distribute findings of research financed by its grants.
Chronicle of Higher Education. 45,35:A33. (May 7)

2. Albert Henderson. 1999. Information science and
information policy. The use of constant dollars and other
indicators to manage research investments. JASIS: Journal
of the American Society for Information Science.

3. Carol Tenopir and Donald W. King. 1997. Trends in
scientific scholarly journal publishing in the United
States. Journal of Scholarly Publishing. 28,3

4. Vannevar Bush. 1945. Science -- The Endless Frontier.
A Report to the President on a Program for Postwar
Scientific Research. Washington DC:National Science
Foundation. Reprint 1990. NSF 90-8.

5. Anonymous. 1998. To Publish and Perish. Policy
Perspectives. 7,4. (March) Co-sponsored by the Association
of Research Libraries, The Association of American
Universities and the Pew Higher Education Roundtable.
Published by Institute for Research on Higher Ecucation.

6. Lisa Guernsey. 1998. A provost challenges his faculty to
keep copyright on journal articles. Chronicle of Higher
Education. 45,4. (Sept. 18).

7. Charles R. McClure and Peter Hernon. 1989. U.S.
Scientific and Technical Information (STI) Policies: Views
and Perspectives. Norwood NJ: Ablex.

8. Albert Henderson. 1994-95. The bottleneck in research
communications, in Publishing Research Quarterly. 10,4:5-21.

9. Richard T. Kaser. 1995. Secondary information services.
Mirrors of scholarly communication. Publishing Research
Quarterly. 11,3:10-24.

10. L. L. Deitz and L. M. Osegueda. 1989. Effectiveness of
bibliographic databases for retrieving entomological
literature: a lesson based on the Membracoidea (Homoptera).
Bulletin of the Entomological Society of America. 35:33-39.

11. American Physical Society. 1991. Report of the APS task
force on electronic information systems, in Bulletin of the
American Physical Society. 36,4:1119-1151.

12. Eugene A. Confrey. 1966. (Letter) The information
exchange groups experiment. Science. 154:843. (18 Nov.)

13. Vannevar Bush. Op cit.; J. Merton England. 1982. A
Patron for Pure Science. The National Science Foundations
Formative Years, 1945-57. Washington DC: National Science


Albert Henderson

PS. Princeton professor Robert Darnton [not Darnham] is
the author of the article in New York Review of Books.

In my earlier remarks, I indicated that I did not have
citations referencing a 1958 proposal similar to E-Biomed
handy. I can supply some now:

(A) The proposal is described in SCIENCE AND TECHNOLOGY ACT
Washington DC: Government Printing Office. 1958.
pages 26-27; 178-195.

(B) It was quickly disposed of by W. O. Baker et al. in
It says,

        "The case for a Government-operated, highly
        centralized type of center can be no better
        defended for scientific information services
        than it could be for automobile agencies,
        delicatessens, or barber shops."

Recalling that the concern in 1958 was competition with
foreign research, it might be appropriate to consider
E-Biomed from the same perspective. The Cold War is over.
Our contemporary concerns were outlined by HELPING AMERICA
INFORMATION (Office of Technology Assessment 1990.

It said, in part:

        A key area of U.S. strength could and should
        be our scientific and technical information.

I think we should consider whether making that information
available "free" to foreign scientists serves to help the
U.S. compete in a world where technological advantage often
makes a critical difference.

The LANL experiment, I am told, has more than a dozen
copies being made systematically abroad. Indeed, one of its
most vocal advocates, emphasizing the "free" nature of
information that would otherwise be sold by publishers and
documentation services, is located in England.

It would serve this particular concern to build stronger
library collections that are "free" to researchers located
geographic regions within the U.S. The network of 125
research universities and major public libraries fits this
description. The LANL server and the proposed E- Biomed
do not.

Albert Henderson
Received on Wed Feb 10 1999 - 19:17:43 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:45:32 GMT