Open Access Publishing and Citation Archives: Background and Controversy
Open Access Publishing and Citation Archives:
Background and Controversy
Updated December 12, 2006
Genevieve J. Knezo
Specialist in Science and Technology
Resources, Science, and Industry Division
Open Access Publishing and Citation Archives:
Background and Controversy
Controversies about open access publishing and archiving confront issues of
copyright and governmental competition with the private sector. Traditional
publishers typically charge readers subscriber fees to fund the costs of publishing and
distributing hard-copy and/or online journals. In contrast, most open access systems
charge authors publication fees and give readers free online access to the full text of
articles. Supporters of the open access “movement” object to the rising costs of
journal subscriptions; share peer reviewers’ reluctance to do free reviews for journals
rapidly escalating in price; and believe that scientific collaboration, advancement, and
utilization will be hastened by free access to information. Traditional subscriber-pays
commercial publishers and some scholarly associations object to most open access
publishing because it may weaken the publishing industry and erode profits. Critics
seek to limit free government-run repositories only to articles and citations from
federally sponsored research; others oppose fees in the thousands of dollars charged
to authors to pay the costs of publishing articles or view as unreliable foundation
donations that sustain some open access activities.
In response to congressional action in 2004 and 2005, the National Institutes of
Health (NIH) implemented a policy that requires authors it funds to voluntarily
submit copies of their manuscripts to NIH’s free access electronic database, PubMed
Central (PMC), as soon as possible after a journal accepts the article for publication,
but within 12 months. The policy allows a publisher-imposed embargo, or delay,
before allowing free public access to the manuscript. Many publishers oppose this
policy and there is only about a 4% compliance rate by grantees. In September 2006,
NIH publicized procedures to permit publishers to post manuscripts or articles
directly to PMC and to give NIH free access to some articles for the embargo period.
In the 109th Congress, report language on H.R. 3010, signed as P.L. 109-149,
endorsed NIH’s policy to post peer-reviewed manuscripts and mandated NIH to
develop its open access repository, PubChem, and to avoid duplication with private
efforts. H.R. 5647 would have mandated NIH-funded researchers to submit final
manuscripts to PMC; S. 2104 would have required submission within six months.
S. 2695, the Federal Research Public Access Act (FRPAA), would have required
federal agencies with research funding exceeding $100 million annually to require
all their federally funded researchers to deposit final manuscripts in a publicly
accessible archive within six months of acceptance by a publisher.
During the 110th Congress, issues likely to generate controversy could include
the FRPAA, which may be reintroduced; modification of NIH’s Public Access policy
to require the government to link to the original journal’s website to read articles;
monitoring the added costs of expanding PubMed Central; determining if other
agencies will use governmental nonexclusive licensing to allow access to
commercially published journal articles, regardless of copyright ownership; assessing
the quality of science published in open access journals; and evaluating the economic
impacts of open access publishing on traditional publishing. This report will be
updated as needed.
In troduction ......................................................1
Definitions of Open Access Publishing and Database Models...............1
Selected Illustrations of Nongovernmental Open Access Activities ..........3
Illustrations of Open Access Systems ..............................3
Public Library of Science (PLoS).............................3
Faculty of 1000.......................................4
Illustrations of Academic-Related Systems..........................5
Illustrations of Dedicated Subject or Disciplinary Archives.............6
Major Issues Relating to Open Access Publishing ........................7
Journal Publishing Costs and Sources of Revenue....................8
Who Pays?: Traditional, Subscriber-Pays Journals................9
Who Pays?: Open Access Journals............................9
Policies For Paying Publication Costs in Relation to the Future of
Open Access Publishing...............................10
Rising Subscription Costs......................................14
The Role of Foundation Support for Open Access Journals........15
Publishing Revenues Support Scientific Societies...................16
Commercial and Open Access Publisher Practices...................16
Timing of Free Access to Journal Articles......................17
Commercial and Open Access Search Engines..................18
Peer Review and Quality of Articles In Open Access Journals..........19
“Enhanced Public Access Policy”: National Institutes of Health (NIH) and Other
Legislative Origins of NIH Policy................................22
NIH’s Public Access Policy and NIH’s PubMed Central (PMC) Database
Legislative Action in the 109th Congress...........................28
Criticisms of “NIH’s Enhanced Public Access Policy”................30
Legislative Proposal to Extend Open Access Policies to Other Agencies: The
Federal Research Public Access Act (FPRAA)..................32
Government Purpose and Copyright Issues.........................34
Issues Relating to Federal Open Access Archives and Publishing...........36
Objections to Government-Operated Databases: Censorship and
Competition in the Free Market.............................37
Allegations of Governmental Censorship......................37
Curbs on Department of Energy Information Systems............37
The Federal Database: PubChem ............................39
Speculation About Differences in Federal Agency Policies........41
Summary of Policy Issues and Questions..............................46
Monitoring of NIH Public Access Activities and Other Federal
Initiatives, Including PubChem..............................47
Appendix 1. Open Access Publishing: Selected Questions in Academia......50
Open Access Publishing and Citation
Archives: Background and Controversy
This report begins with an inventory of basic information: definitions and guides
to histories of the growth of open access publishing and citation archives and
descriptions of selected major open access activities. It moves on to summarize
major points of difference between proponents and opponents of nongovernmental
open access publishing and databases, and then highlights federal, including National
Institutes of Health (NIH), open access activities and contentious issues surrounding
these developments. The report also briefly describes open access developments in
the United Kingdom (where a number of governmental and nongovernmental
initiatives have occurred) and in the international arena. Finally, controversial issues
which could receive attention the 110th Congress are summarized.
Definitions of Open Access Publishing and
The “open access movement” is said to have begun in 1966.1 The term
describes a variety of activities that includes access to archives of indexed citations
of articles, access to separate journal articles that were published in traditional,
subscriber-pays journals, and access to free, online journals.2 According to a May 23,
1 With the inception of Educational Resources Information Center (ERIC), launched by the
U.S. Department of Education’s Office of Educational Research and Improvement and the
National Library of Education. This database contains bibliographic citations for privately
published journal articles and allows retrieval of the text of other nonpublished materials.
Medline, a bibliographic system, was launched by the National Library of Medicine in 1966
(but was not free until 1997). Source: “Timeline of the Open Access Movement,” by Peter
Suber, last revised Apr. 13, 2005, at [http://www.earlham.edu/~peters/fos/timeline.htm].
This is an extensive history since 1966, with hotlinks to different systems and databases.
2 See the following information about open access publishing: Martin Frank, Margaret
Reich, and Alice Ra’anan, “A Not-For-Profit Publisher’s Perspective on Open Access, as
it was planned to be published in Serials Review, vol. 30, no. 4, 2004; “Budapest Open
Access Initiative,” available at [http://www.soros.org/openaccess/read.shtml]. See also
Peter Suber, “What You Can Do to Promote Open Access,” Last revised April 5, 2005, 11
p. [http://www.earlham.edu/~peters/fos/do.htm]; “Budapest Open Access Initiative:
Frequently Asked Questions,” last revised March 27, 2005 [http://www.earlham.edu
/~peters/fos/boaifaq. htm]. Several open access online journals and indexes of collections
of these are available. For instance, see Open Access Bibliography: Liberating Scholarly
2005 Wall Street Journal article,”[c]urrently, the open-access movement makes up
between 1% and 2% of the market, experts say. While that number seems small, the
concept is assuming an important role channeling academic discontent” about the
rising costs of journals.3
In traditional, subscriber-pays publishing, the publisher, who holds the
copyright to an article, pays most printing and distribution costs and, in order to read
an article, the journal subscriber pays fees, whether for hard-copy or online versions.
Sometimes an author is required to pay printing page charges for complex graphics
or color presentations.
“Open access” publishing generally means that the author or publisher, who
holds the copyright to an article, grants all users unlimited, free access to, and license
to copy and distribute, a work published in an open access journal (which may be
published initially electronically or in hard-copy). Users can also make copies for
their personal use, if authorship is properly attributed.4 Open access publishing often
requires an author to pay for publishing or posting of a paper. Estimates of fees
charged vary, but generally range from about $500 to $4,000. These charges may be
paid by individual authors, or by institutions, pursuant to institutional subscription
contracts with open access journals that cover publication charges for all authors
affiliated with that institution. Typically, open access publishers require that a
complete version of the work and related materials be deposited electronically in an
online database that permits open access, distribution, interoperability (allowing users
to extract and use the data in other research), and long-term archiving.5
In “free access” publishing neither an author nor a reader pays for articles to be
published or posted on the Internet,6 but other open access features may not be
A few commercial publishers have adopted some open access features in their
business models. However, the fundamental difference is that traditional publishers
Literature with E-prints and Open Access Journals, information is available at
[http://www.escholarlypub.com/oab/oab.htm]; “List Issues: Open Access (Journal)
Collections: Electronic Resources in Libraries,” [http://www.joanconger.net/ERIL/
list_issues_openaccess. html]; SCIELO, available at [http://scielo.org]; HighWire Press
[http://highwire.stanford.edu/lists/freeart.dtl], and PubMed Central, at
3 Bernard Wysocki, Jr., “Peer Pressure: Scholarly Journals’ Premier Status is Diluted by
Web,” Wall Street Journal, May 23, 2005, p. A1.
4 This is a variation of “Creative Commons” copyright licenses free for public use. See
5 Based on “Definition of Open Access,” which uses a modified version of the “Bethesda
Meeting on Open Access,” [http://www.plos.org/about/openaccess.html]. See “Open-
Access Publication of Medical and Scientific Research,” a Public Library of Science
Background Paper, Dec. 12, 2003.
6 Joanne S. Hawana, “Multiple Publishing Models Critical To Advancing Science, Journal
Publishing Societies Argue,” Washington Fax, Mar. 17, 2004.
generally require readers to pay to read or print an article, or to search indexes of
abstracts or citations. Open access publishers generally do not require readers to pay
for these services. Some traditional publishers say they already provide open access
in that they may make papers freely available online — but this is usually a year or
two after publication. The publishers still hold copyright, and they may or may not
allow the author to post his or her published articles in an open access repository or
database, or on the author’s own website.
The scope of open access repositories or archives varies. Some contain
published journal articles or nonpublished “grey literature” in all fields of science or
in specific scientific disciplines. Some archive a specific researchers’ preprints,
articles, or research reports; or, as in the case of the National Institutes of Health
model, articles, data, or other materials funded by an agency, but prepared for
publication by traditional publishers. Some open access repositories archive only
citations for articles or other materials; some archive both citations and full text
materials; some allow free downloading and some do not.
Selected Illustrations of Nongovernmental Open
A variety of nongovernmental open access publishing activities is illustrated
next with summaries of some current major open access information systems or
publishers. These are categorized by general type, including commercial open access
systems, academic-sponsored systems, and subject or disciplinary systems. NIH’s
PubMed Central (PMC) system is described in detail in the section of this report that
focuses on NIH.
Illustrations of Open Access Systems
Public Library of Science (PLoS). PLoS is a nonprofit group, spearheaded
in large part by Dr. Harold Varmus, former NIH director. It provides readers with
free access to peer reviewed articles published in PLoS’s electronic journals. The
activity is supported by author payments starting at $1,500 per article and multi-
million dollar philanthropic foundation contributions. PLoS’s journals include PLoS
Biology, PLoS Medicine, PLoS Computational Biology, PLoS Genetics, and PLoS
Pathogens.7 PLoS seeks to launch journals in other disciplines. It has the goal of
publishing highly selective, top-quality articles competitive with the quality of
articles in traditional, subscriber-pays journals like Science and Nature. Different
from traditional, subscriber-pays publishing, which requires authors to cede copyright
to the publisher, authors who publish in PLoS retain copyright to an article, but are
required to deposit a copy of the article in an open access, online repository that
allows long-term archiving.8 Reportedly, one of the group’s major goals is to make
7 Janet Coleman, “Public Library of Science to Launch 3-4 New Open-Access Scientific
Journals in 2005,”Washington Fax, Oct. 29, 2004.
8 Information about PLoS and related archives is available at [http://www.plos.org/
research more accessible by eliminating publishers as copyright holders and by
ending the “balkanization” of scientific information in separate databases. Under
PLoS’s editorial policy, “any data can be integrated into new work as long as the
original author is credited appropriately. The model is inspired by GenBank, the
central repository of DNA sequence whose open access policy has driven much of
the progress in genomics and biotechnology of the last decade.”9 PLoS has
announced that it will assist scientists in developing countries by providing Internet
access for readers of limited bandwidth, and will waive or defray author charges for
those who cannot afford to pay.10
BioMedCentral (BMC). This is a British-founded, independent, commercial
publishing system, which provides free access to peer reviewed biomedical research11
published online. It publishes its own approximately 120 biomedical journals and
says articles are rapidly peer reviewed; peer review policies are determined by each
journal’s board. Authors retain copyright of their work. BioMedCentral charges
authors or their institutions for the costs of peer review and publication. “Other
sources of revenue include subscription access to commissioned articles, sales of
paper copies of our journals to libraries, sales of reprints, advertising and
sponsorship, and ... a range of subscription-based value added services such as
literature reviews and evaluation, personalized information services delivered
electronically, provision of editorially enhanced databases, tools that help scientists
collaborate, and other software research aids.”12 It archives materials in PubMed
Central, NIH’s free archive of biomedical literature.
Faculty of 1000. BioMedCentral has created a fee-based subscription service
called Faculty of 1000.13 It originated because the publication of so many articles in
online journals (sometimes free to readers) with varying degrees of peer review has
spawned a new industry: peer reviewers or experts who evaluate articles after
publication and provide a selected list of articles recommended for reading to their
9 Amy Harmon, “New Premise in Science; Get the Word Out Quickly, Online,” Dec. 17,
10 “PLoS, Frequently Asked Questions,” available at [http://www.plos.org/faq.html].
11 Available at [http://www.biomedcentral.com/info/].
12 Available at [http://www.biomedcentral.com/info/about/whatis].
13 Available at [http://www.facultyof1000.com/about/key].
patientINFORM. In spring 2005 patientINFORM14 was launched by the
American Cancer Society, the American Heart Association, and the American
Diabetes Association, in partnership with more than 20 publishing firms, to provide
immediate access to free, selected full-text research articles and materials from the
three organizations’ websites, which provide links to different types of published
materials. “According to the group, the initiative ‘is being driven by recent trends
indicating that public awareness of clinical research, heightened by media coverage
and fueled by the spread of broadband Internet, has led more and more patients to go15
online to find the latest information about treatment options.’” NIH’s decision to
launch its system, reportedly, accelerated the formation of patientINFORM. After
a period of evaluation, “ ... the group will determine whether to expand its focus past
the three initial diseases into rarer conditions.”16 This system may not permit
permanent access to materials on it, since the organizations maintaining the website
may, overtime, replace or remove materials posted.
Illustrations of Academic-Related Systems
Some universities ensure that their scholars’ publications are available online
in a free open access repository by creating their own archives or participating in
networked open access archives. Several examples are outlined next.
EScholarship Program. The EScholarship Program of the University of
California system was launched in the fall of 2003. It is an electronic, searchable
repository that makes freely available an archive of the publications (and other
media) and some research databases of University of California researchers. The
vehicle is also used to disseminate the university’s own open access, peer reviewed
published journals.17 Supporters of systems like this say that indexing materials
improves access to them and, if full text is available, widens reader access, and
improves utilization of federally financed research and development.18
14 Information is available at [http://www.patientinform.org/]. Participating publishers and
associations include the International Association of Science, Technical and Medical
Publishers; the Association of American Publishers/Professional and Scholarly Publishers;
Johns Hopkins’ Welch Medical Library; and the National Library of Medicine’s
MedLinePlus; the American Association for the Advancement of Science; the American
Medical Association; the American Physiological Association; Oxford University Press;
Blackwell Publishing; Elsevier Publishing; BMJ Publishing Group; Nature Publishing
Group; and Springer and Wiley.
15 Andrew Hawkins, “Journal Publishers, Advocacy Groups Spearhead New Open Access
Initiative,” Washington Fax, Dec. 13, 2004.
16 Hawkins, op. cit., Dec. 13, 2004.
17 Available at [http://escholarship.edlib.org].
18 “UC to Launch Open-access Journals” The Scientist, June 16, 2003.
DSpace. A number of research universities19 are participating in DSpace, a
networked multi-member electronic repository that indexes and shares some research20
data, articles, and other media. It was developed by the Massachusetts Institute of
Technology (MIT) in collaboration with Hewlett-Packard. Some universities, such
as Cornell, reportedly, are using it to provide free access to peer reviewed
Highwire Press. This is an archive run by Stanford University that provides
online, full-text articles for biomedical and other scientific journals. It adheres to the
post-publication timing policies of each journal, with most articles archived and
made accessible between 6 and 24 months after publication in the original traditional,
subscriber-pays published journal. Some of these articles, but not all, may be viewed
Illustrations of Dedicated Subject or Disciplinary Archives
Some repositories permit free searching for citations, abstracts, articles, or other
materials in specific disciplinary fields or areas of application, or by researchers
affiliated with specific academic systems, or by other researchers. A few illustrations
are given next.
arXIV.org.22 Initiated in 1991, this is a free, online archive which allows
physical science researchers to make preprints of their papers available before formal
publication. Maintained by the Cornell University Library23 (in cooperation with the
National Science Foundation and the Department of Energy), it includes articles in
the following subjects: physics, mathematics, nonlinear sciences, computer science,
and quantitative biology. According to PLoS, “This server expanded from its initial
19 Including Brigham Young University; Case Western Reserve University; Chapel Hill
School of Information and Library Science Electronic Theses and Dissertations; Cornell
University; Digital Repository at the University of Maryland; DLEARN at the University
of Arizona; Drexel University; DSpace@Cambridge; DSpace at MIT; DSpace at University
of Rochester; Edinburgh Research Archive; Embry-Riddle Aeronautical University; WETD
of Indian Institute of Science, Bangalore (etd@IISc); George Mason University; Hong Kong
University of Science and Technology; IDeA, Indiana University Purdue University Indiana;
Dspace at Indiana University Of Pennsylvania; Kansas State Publications Archival
Collection, Kansas State Historical Society and Kansas State Library; KU ScholarWorks;
Los Alamos National Laboratory, New Mexico; Portfolio@Duke University; RIT Digital
Media Library; SMARTech Scholarly Materials and Research at Georgia Tech; Texas A&M
University Libraries Institutional Repository; T-Space at The University of Toronto
Libraries; University of New Mexico, DSpaceUNM; University of Oregon Scholars’ Bank;
University of Tennessee in Knoxville; University of Texas at Austin, School of Information;
University of Washington, Seattle; University of Wisconsin; Vanderbilt University
e-Archive; Washington University, St. Louis; Woods Hole Open Access Server.
20 Available at [http://www.dspace.org/].
21 Available at [http://highwire.stanford.edu/lists/freeart.dtl].
22 Available at [http://arXIV.org].
23 “Scientific Publishing: Who Will Pay for Open Access?,” Nature, Oct. 9, 2003. See also
role as a vehicle for sharing preprints in theoretical high-energy physics to its current
role as the principal ‘library’ for a large fraction of research literature in physics,
computer sciences, astronomy, and many mathematical specialities. Today, more
than half of all research articles in physics are posted to this server prior to their
publication in conventional journals. In many fields, these ‘eprints’ are the de facto
publications of record.
CogPrints. Some types of foreign open access publishing include access to
U.S.-generated research findings. CogPrints is a free, British-run, self-archive of
full-text, electronically available, published, peer reviewed journal articles as well as
preprints of unrefereed articles in the “cognitive sciences, including any area of
psychology, neuroscience, and linguistics; many areas of computer science (e.g.,
artificial intelligence, robotics, vison, learning, speech, neural networks); philosophy
(e.g., mind, language, knowledge, science, logic); biology (e.g., ethology, behavioral
ecology, sociobiology, behavior genetics, evolutionary theory); medicine (e.g.,
psychiatry, neurology, human genetics, imaging); anthropology (e.g., primatology,
cognitive ethnology, archeology, paleontology), as well as any other portions of the
physical, social and mathematical sciences that are pertinent to the study of24
Major Issues Relating to Open Access Publishing
Controversies arise because developments in open access systems and policies
seem to have outpaced society’s ability to design equitable and efficient mechanisms
and economic reward structures to manage transitions between traditional and open
access publishing and archiving.25 There is evidence that greater acceptance of online
and open access publishing is “forcing traditional journals to address fundamental
financial and philosophical challenges,”26 which has generated heated discussions in
the scientific publishing community.27
24 See [http://cogprints.org/].
25 See, for instance, Julie M. Esanau and Paul F. Uhlir, eds., Open Access and the Public
Domain in Digital Data an Information for Science, Proceedings of an International
Symposium, Published by U.S. National Committee for CODATA, National Academies
Press, Washington, D.C., 2004.
26 “Study Probes ‘Open Access’ and Scholarly Publishing,” Science, Dec. 23, 2005, p. 1918.
Data about these changes and policy implications appear in: Kaufman-Willis Group, The
Facts About Open Access, Association of Learned and Professional Society Publishers,
Worthing, UK, 2005; Mark Ware Consulting, Ltd., Scientific Publishing in Transition: An
Overview of Current Developments, Bristol, UK, September 2006, 30 p. See also John
Willinsky, The Access Principle: The Case for Open access to Research and Scholarship,
Cambridge, MIT Press, 2005, 307p.
27 See for instance, the online blog, “Open Access News,” available at
[ h t t p : / / www.e a r l h a m.e du/ ~ p e t e r s / f o s / f o s b l o g.ht ml ] .
Major arguments28 made by supporters of open access publishing (largely
scientists, librarians, and some non-profit publishers) are that it rides the new wave
of inevitable changes in publishing and electronic dissemination of information due
to development of the Internet,29 hastens scientific progress, gives access to more
readers, promotes economic development, and, in the case of federally funded
research, provides citizens with ready access to the results of research and
development that their taxes funded.
Opponents of open access publishing (primarily traditional publishers and major
scientific associations) cite such issues as the doubtful permanence of electronic
archives, questions of copyright ownership and reductions to traditional publishers’
profits, costs to researchers who have to pay to have their manuscripts published in
open access journals, the possibly dubious quality of articles published, questions
about peer review processing and quality, perceptions of the academic community
and the academic reward system which appear to give more status to articles
published in traditional, subscriber-pays journals, and so forth.30 See Appendix 1 for
a list of additional issues raised about the impact of open access publishing on the
academic community, scholarship, and teaching.
The following sections elaborate on some of these issues.
Journal Publishing Costs and Sources of Revenue
The costs of publishing a journal article include preparing the manuscript for
publication (initial sorting and selection of manuscripts to be refereed, peer review,
selection, editing, layout, table of contents, overhead, letters to the editor, etc.) and
distribution. According to a Wall Street Journal story, costs for publishing an article31
typically range from $3,000 to $4,000. However, these costs can average more than
$10,000 for some journals, such as Science magazine, which publishes only a small
28 For a comprehensive review of major arguments pro and con about open access publishing
and archiving, see Organization for Economic Cooperation and Development (OECD),
Working Party on the Information Economy, Digital Broadband Content: Scientific
Publishing, Sept. 2, 2005, DSTI/ICCP/IE(2004)11/FINAL.
29 David Stern, “Archival Issues Regarding Electronic Scientific Literature,” Presentation
at session on “The Future of Scientific Communication (Formerly Known as Publishing),”
American Association for the Advancement of Science (AAAS) Meeting, Apr. 21, 2005.
30 According to one author, barriers to open access publishing include legal framework
issues; differences in IT-infrastructure and technologies; business models and costs;
indexing services and standards of materials placed in open-access archives; the academic
reward system; and marketing and critical mass issues. The importance of each type of
barrier varies with the type of open access repositories, whether open-access journal,
subject-specific repositories maintained by disciplinary groups; or institutional repositories,
maintained by academic institutions. The author provides a matrix and specific details for
each of the 18 cells in his analysis in: Bo-Christer Bjork, “Open Access to Scientific
Publications - An Analysis of the Barriers to Change?”, Information Research, Jan. 2004.
31 Bernard Wysocki, Jr., “Peer Pressure: Scholarly Journals’ Premier Status is Diluted by
Web,” Wall Street Journal, May 23, 2005, p. A1.
fraction of the articles submitted (about 7%),32 but has high value-added costs, which
include reviewing all articles submitted and selecting those that will be published,
layout, graphics, distribution, and so forth. Another author has estimated costs for
publishing an article in other journals: BioScience, about $7,000 per article; Nature
and New England Journal of Medicine, in excess of $1,500.33
The comparative costs of publishing online only versus traditional journals that
print hard-copy are uncertain. While some observers say that article processing costs
are similar for print and electronic publications, other research shows that electronic
publishing and distribution are cheaper than hard-copy publishing.34 A private
British funding group reported that research it commissioned showed that author
pays, open access publishing models are a viable alternative to subscription journals
and “have the potential to serve the scientific community successfully.”35
Specifically, “Open access publishing should be able to deliver high-quality, peer-
reviewed research at a cost that is significantly less than the traditional model while
bringing with it a number of additional benefits.”36
Who Pays?: Traditional, Subscriber-Pays Journals. Traditional
publishers usually incur most of the costs of publishing an article. Revenue comes
from subscriptions, advertising, reprints, and, in some cases, from authors who are
asked to subsidize the costs of color printing or printing of complex graphics, or page
charges for publishing articles in traditional hard-copy journals. Data for 2004 from
a study by the Kaufman-Willis Group, which surveyed sources of revenue for
traditional and open access publishers, indicates that the three largest sources of
revenue for traditional journal publishers were subscriptions, which provided, on
average, about 70% of total revenue; industry support (advertising and sponsorship)
at about 15% for some journals and membership dues at about 8% for others; and
author fees and charges.37
Who Pays?: Open Access Journals. Reportedly, most, but not all, open
access journals require authors to pay from about $500 to $4,000 for publishing costs.
32 Wysocki, May 23, 2005, op. cit.
33 David Malakoff, “Opening the Books on Open Access,” Science, Oct. 24, 2003, p. 551.
34 Donald W. King, “The Economics of Science Publishing,” Presentation at Session on
“The Future of Scientific Communication (Formerly Known as Publishing),” American
Association for the Advancement of Science (AAAS) Meeting, Apr. 21, 2005.
35 “Costs and Business Models in Scientific Research Publishing,” press summary available
at [http://www.wellcome.ac.uk/print/wtd003185_print.html], describing Costs and Business
Models in Scientific Research Publishing: A Report Commissioned by the Wellcome Trust,
Compiled by SQW Limited, April 2004.
36 Costs and Business Models in Scientific Research Publishing: A Report Commissioned
by the Wellcome Trust, op. cit. On this specific finding, see p. 4.
37 The exact amount of revenue from author fees and charges varies from 0.8 % for some
types of publishers to 9.3 % for others. “The Facts About Open Access: A Study of the
Financial and Non-Financial Effects of Alternative Business Models for Scholarly
Journals,” Researchers: Kaufman-Wills Group, LLC, published by the Association of
Learned and Professional Society Publishers, 2005, pp. 45-46.
Open access journals also receive funds from advertising, corporate sponsorships,
government grants, the use of volunteers, and foundation grants.38 The study by the
Kaufman-Willis Group, cited above, identified the three largest sources of revenue
in 2004 for open access journal publishers as industry support (advertising and
sponsorship) at 37%; author fees and charges at 30%; and grants at 13%.39
This same study showed that, contrary to expectations, author fees were charged
by a larger fraction of traditional, subscriber-pays journals than open-access
journals.40 Author fees include charges for color printing, page layout, page
publication charges, and so forth. This finding, in combination with the data on
percentage sources of revenue, appears to mean that in relation to the total number
of publishers, traditional publishers more than open access publishers charged fees
to authors, but the payments (as a percentage of publishers’ total revenue) were less
to traditional publishers than to open access publishers. The fees traditional,
subscriber-pays publishers charged to authors were primarily for small changes, color
views, and related items, rather than the larger fees open access journals charge
authors to publish in the open access journal.41
Policies For Paying Publication Costs in Relation to the Future of
Open Access Publishing. Among the issues related to “author pays,” and
possibly to the future of open access journals, is whether the federal government will
continue to allow some research grant funding to be used to pay charges levied on
authors or institutions for the costs of publishing articles resulting from federally
38 PLoS’s webpage includes the following information: “PLoS is a tax-exempt, 501(c)3,
nonprofit corporation headquartered in San Francisco, California (Federal Tax ID
68-0492065). PLoS is governed by an eleven-member Board of Directors. PLoS co-founder
Harold Varmus is Chairman of the Board. PLoS has received financial support in the form
of grants from the Gordon and Betty Moore Foundation, the Sandler Family Supporting
Foundation, the Irving A. Hansen Memorial Foundation, the Open Society Institute (OSI),
and the Joint Information Systems Committee (JISC). PLoS also receives support through
donations, sponsorships, and memberships from private citizens, universities, and other
organizations” [http://www.plos.org/about/index.html]. It reported that it received a $9
million grant from the Moore Foundation to start operations for four years
[http://www.plos.org/ about/index.html ].
39 “The Facts About Open Access....,” op. cit., pp. 45-46.
40 “The Facts About Open Access...., op. cit., p. 44. See also, regarding an interim version
of the study, Lila Guterman, “New Study Compares Open-Access and Traditional
Publishing,” Chronicle of Higher Education, Mar. 25, 2005. The interim study was
“Variations on Open Access: A Study of the impact of Alternatives Business Models on
Financial and Non-Financial Aspects of Scholarly Journals,” Preliminary Results Presented
14 March, 2005, London Book Fair. “The survey was conducted by the Kaufman-Wills
Group, publishing consultants based in Baltimore. It was financed by groups that are
affiliated largely with traditional journals: the American Association for the Advancement
of Science, the Association of American Medical Colleges, the Association of Learned and
Professional Publishers, and HighWire Press, which produces online versions of journals
and is operated by Stanford University.” An author of the interim study, “Variations on
Open Access,” op. cit., agreed with this interpretation of the data. (Interview held July
41 See also “Variations on Open Access,” op. cit.
funded research. This may become a more prominent issue if open access publishing
becomes a larger part of the market.
Now, pursuant to OMB’s guidelines, federal agencies that award funds for
scientific research permit investigators at universities, colleges, and nonprofit
institutions to charge the costs of publishing a scientific article as an allowable direct
cost (usually paid in full) if the funding agency agrees that they are an appropriate
part of the project. If the costs of publishing are disallowed as direct costs, the
federal governments likely will pay for these costs as part of “facilities and
administrative” (F&A) indirect costs, if the research was federally sponsored and if
the journal levies similar charges on all research papers published by the journal.42
If the cost is covered as an F&A indirect cost, full reimbursement may not occur due
to limitations on recoveries of some indirect costs.
Some federal agencies have issued policy guidance about allowing as a direct
cost of project support, fees for publication and page charges in order to disseminate
reports of the agency’s federally funded research results. The National Science
Foundation (NSF), for instance, says,
The proposal budget may request funds for the costs of documenting, preparing,
publishing or otherwise making available to others the findings and products of
the work conducted under the grant. This generally includes the following types
of activities: reports, reprints, page charges or other journal costs (except costs
for prior or early publication); necessary illustrations; clean up, documentation,
storage and indexing of data and databases; development, documentation and
debugging of software; and storage, preservation, documentation, indexing, etc.,
of physical specimens, collections or fabricated items.43
According to NIH, the following publication costs are allowed:
Page charges for publication in professional journals are allowable if the
published paper reports work supported by the grant and the charges are levied
impartially on all papers published by the journal, whether or not by government-
sponsored authors. The cost of reprints and publishing in another media, such as
books, monographs and pamphlets, also are allowable. Publications and journal
articles produced under an NIH grant-supported project must bear an
acknowledgment and disclaimer as appropriate, as provided in Administrative
42 Rules for educational institutions are found in Circular A-21, “Cost Principles for
Educational Institutions,” Revised May 1, 2004. See section D, for information about direct
costs. See section J39, for treatment of publication and printing as F&A “facilities and
administrative” indirect costs. Circular A-21 also allows costs of subscriptions as “facilities
and administrative” indirect costs. Rules governing nonprofit institutions are discussed in
OMB Circular A-110, “Cost Principles for Non-profit Organizations.” See the section,
Attachment B, items 33 and 41.
43 National Science Foundation, Grant Proposal Guide, September 2004, NSF 04-23,
Section II. 2.g.vi(b).
Requirements — Availability of Research Results: Publications, Intellectual44
Property Rights, and Sharing Research Resources.”
Publication costs, library fees, and journal subscription costs related to a specific
research project may be allowed as costs of a federally supported research project.
It is not known if the federal government will extend these allowances to include the
costs of institutional subscriptions that open access publishers or journals may sell
to colleges or universities to cover publication fees for all authors affiliated with a
specific institution. At least one report cautions that some federal agencies may not45
allow publication costs to be covered. Harold Varmus, a co-founder of PLoS,
considers “publishing fees as the final, relatively cheap step of a research project”46
and contends that the federal government should pay for these costs.
In 2003, the UK Wellcome Trust, a large research charity that supports
biomedical research in the United Kingdom, announced its support of online open
access journals and said it would allow scientists it funds to use a portion of their
grant to pay author charges required by the journals.47 The U.S.-based Howard
Hughes Medical Institute allows grantees to use up to an additional $3,000 to spend
for publishing in open access journals.
Some professional groups have developed, or widened, policies for “author
pays” publishing allowing free access to readers. For example, in 2006 the American
Chemical Society and Elsevier, both of which publish large numbers of scholarly
scientific journals, announced that they would establish mechanisms permitting
authors to pay a few thousand dollars to allow their articles to be viewed online for48
free after publication of the journal. Similarly, anticipating a the release of many
important papers after the 2007 start-up of the Large Hadron Collider (LHC),
particle physicists are seeking free access to all articles published in their field. In
a report released in June 2006 a task force, led by CERN,49 a physics laboratory in
Switzerland, proposed “that a consortium of labs and funding agencies pay
publication costs for particle physics papers. It would cost $6 million or more a year
to include all the journals willing to offer an open-access option, the group estimated.
44 “Selected Items of Cost,” in Part II, Terms and Conditions of NIH grant Awards, Subpart
a: General — File 3 of 5,” in NIH Grants Policy Statement (12/03).
45 Catherine Zandonella, “Economics of Open Access,” The Scientist, Aug. 22, 2003.
46 Malakoff, op. cit., Oct. 24, 2003, p. 553.
47 Declan Butler, “Wellcome to Fund Publication in Open-access Journals,” Nature, Oct.
48 Scott Jaschik, “Momentum for Open Access Research,” Inside Higher Ed., Sept. 6, 2006
and Susan Brown, “Coalition Works to Secure Open Access to Published Research,”
Chronicle of Higher Education, Sept, 22, 2006. See also “ACS Offers Open-Access Option
to Authors,” Chemical and Engineering News, Sept. 4, 2006, p. 11.
49 CERN stands for the European Organization for Nuclear Research. The name CERN is
That would cover up to half of the 6,000 or so original theory and experimental
papers published each year.”50
Supporters of open access sometimes contend that now most publishing costs
are borne by research sponsors, such as the federal government, and that allowing
these sponsors to shift support to pay for open access publishing will not cost more
and will provide more benefits to society. For instance,
Asking research sponsors to pay for publication of the research they support may
seem to impose new financial burdens on the government agencies, foundations,
universities and companies that sponsor research. But these organizations already
pay most of the costs of scientific publishing — a huge fraction of the US $9
billion annual revenue of scientific, medical, and technology journals comes
from subscriptions, site licenses, and publication fees ultimately billed to grants
or employers. Much of the rest is borne by society in the form of increments to
university tuitions; healthcare costs, including drug prices; and state and federal
taxes that subsidize healthcare, libraries, and education. Surely the cost of open-
access digital publishing cannot, in total, be more than we are already paying
under the subscription and licensing model. By simply changing the way we
support the scientific publishing enterprise, the scientific community and public
would preserve everything we value in scientific publishing and gain all of the51
benefits of open access.
In opposition, some say if the government paid such costs, money would be
diverted inappropriately from research to publishing. Some universities say their
costs will increase if they need to reimburse researchers to pay author fees for open
access journals and if they still have to pay high costs for subscriptions to traditional
journals.52 In addition, some young scientists/investigators say that business models
that force authors to pay for publication in open access journals could hurt them since
they often have smaller grants and “... an author-pays model could amount to a ‘tax53
for productivity.’ “ Another issue is that in some applications-oriented fields, such
as medicine, engineering, computer science, management, and pharmacy, users of
journals, including open access materials, are often private sector parties who read
the journals, but likely would not be authors who would contribute to journal
publication costs. As a result, researchers who produce knowledge would bear
disproportionate costs for journal publication.54
50 Jocelyn Kaiser, “Particle Physicists Want to Expand Open Access,” Science, Sept. 1,
51 Patrick O. Brown, Michael B. Eisen, and Harold E. Varmus, “Why PLoS became a
Publisher,” PLoS Biology, vol. 1, no. 1, p. 1.
52 Lila Guterman, “The Promise and Peril of ‘Open’ Access,” The Chronicle of Higher
Education, Jan. 30, 2004, op. cit.
53 Andrew J. Hawkins, “Scientists at NIH Open access Meeting Fear Author-pays Publishing
Would Hurt Young Investigators,” Washington Fax, Aug. 21, 2004.
54 Organization for Economic Cooperation and Development (OECD), Working Party on the
Information Economy, Digital Broadband Content: Scientific Publishing, Sept. 2, 2005,
DSTI/ICCP/IE(2004)11/FINAL, p. 65.
Rising Subscription Costs
It has been reported that traditional, subscriber-pays academic publishing has
a $5 billion global market,55 and that one of the leading publishers, Reed Elsevier
journals, “bring[s] in about $1.6 billion in annual revenue with an operating-profit
margin of about 30%.” This profit, according to the same source, could be cut to
between 10% to 15% if open access publishing were expanded.56 (The total scientific
and technical journal market has been estimated at $9 billion.)57
Subscription costs vary depending upon the journal and how many journals an
institution subscribes to. Prices also vary for individual versus institutional
subscriptions. According to one article, in October 2003 two scientists at the
University of California at San Francisco were charged $91,000 “from Elsevier’s Cell
Press unit for one-year’s access to six biology journals.”58 The University of
California in 2003 was reportedly charged $7.7 million a year for subscriptions to
negotiated after faculty moves to boycott Elsevier journals if the original bill price
were not reduced.59 Reportedly, sometimes sales are increased by publishers forcing
libraries to subscribe to more than they want because publishers often “... bundl[e]
... journal subscriptions into large contracts often not well matched with institutional
research interests.”60 This includes bundling together journals that are made
available electronically in database systems that access current and archived journals.
Bundling of this sort can force libraries to pay for access to the same journal several
times if it is included in more than one database to which the library subscribes.61
Rising journal subscription costs, it is argued, are too expensive, making it
difficult for libraries, especially university libraries and the public to afford many
journals,62 and forcing them to sacrifice spending on other media. Reportedly, Rick
Johnson, former Director of the Scholarly Publishing and Academic Resources
Coalition (SPARC), said that because of rising costs, library spending on print media
is shifting from monograph and other materials to support largely journal
subscriptions, with price per journal reportedly having doubled within 15 years. He
illustrated this by saying that while the Consumer Price Index increased 64 percent,
55 This is the figure for academic market sales and is less than the $9 billion figure cited by
Brown, Eisen and Varmus, op. cit., possibly because it excludes some sales and consumers.
56 Wysocki, May 23, 2005, op. cit. The profit margin reduction figures, according to
Wysocki are from an estimate by “Sami Kassab, analyst at investment house Exane BNP
Paribas in London....”
57 Brown, Eisen and Varmus, op. cit.
58 Wysocki, May 23, 2005, op. cit.
59 Wysocki, May 23, 2005, op. cit.
60 Jocelyn A. Rankin and Sandra G. Franklin,”Open Access Publishing,” Emerging
Infectious Diseases, July 2004, pp. 1352-1353.
61 Interview with CRS librarian, July 2005.
62 Rankin and Franklin, op. cit.
libraries are paying 227 percent more for journal subscriptions.63 According to a
National Library of Medicine (NLM) report, Access to Biomedical Research
Information, prepared for Congress in June 2004, “prices of commercial biomedical
titles increased 224% from 1988 to 1998, while the prices of nonprofit titles
increased 129%.”64 The NLM report was quoted as saying that “ ‘These trends have
adversely affected the ability (from a cost standpoint) of academic and health science
libraries to continue to support the needs of the research and health care provider
communities for access to biomedical literature ....’”65
The current open access movement has been fueled by actions of academics and
librarians located at the University of California campuses, as well as at other
academic sites, who, in late 2003 and 2004, mounted strenuous objections to
increases in costs for subscriptions to scientific journals. Some demanded a 25%
reduction in subscription fees from major scientific publishers, with Reed Elsevier
often cited as a major target, and said if fees were not reduced, they would relinquish
journal editorial board memberships or stop providing free peer reviews for major
The Role of Foundation Support for Open Access Journals. The
question as to whether open access journals can exist without subsidies may still be
unanswered. Some observers wonder whether open access journals and archives can
be sustained without philanthropic contributions and what will happen if foundation
contributions are ever reduced. It has been reported that several journals which
attempted to provide free access to readers reversed policies due to falling
subscription rates and revenues for print journals. These journals reportedly included
the Journal of High Energy Physics, which published online for free for six years;
it originally did not charge authors a fee, but ultimately decided “to impose a
subscription fee of about $1000 a year” for readers.67 There is also a question of
whether, if publishing patterns and revenue sources change, publishers will obtain
enough revenue to be able to risk starting up niche journals in narrow fields of
science and which have a small readership, which many traditional publishers have
been able to do given their revenue margins.68
63 Damon Brown, “Open Access Journals Offer a New Way of Publishing,” Journal of the
American Dietetic Association, 2004, p. 1060.
64 As cited in Bradie Metheny, “Open Access Publishing Language in House Labor/HHS
Bill Stirs Controversy,” Washington Fax, July 20, 2004.
65 Cited in Metheny, July 20, 2004, op. cit.
66 This last point has been made by Wysocki, op. cit., and others.
67 David Malakoff, “Money Woes Force Some to Change Course,” Science, Oct. 24, 2003,
p. 553. For additional information about financial issues, see Catherine Zandonella,
“Economics of Open Access,” The Scientist, Aug. 22, 2003 and Martin Frank, Margaret
Reich, and Alice Ra’anan, “A Not-For-Profit Publisher’s Perspective on Open Access,”
preprint as forthcoming in Serials Review, vol. 30, no. 4, 2004, p. 6.
68 Wysocki, May 23, 2005, op. cit.
Publishing Revenues Support Scientific Societies
The point is often made that scientific societies, which may publish on their own
or may use commercial publishers to publish their journals, reap considerable profits
from their share of journal revenues. They then use these profits to support societies’
activities, which can include advocacy and assistance to new researchers in the field.
Critics of this practice say that these professional associations need to find different
business models, or alternative ways to raise money, to support their activities instead
of using publishing profits, which are based on payments from subscribers, university
libraries, and, in many cases, indirect costs of federally funded R&D.
On the other hand, revenues to scientific societies may not decrease since, at
least according to one professional association, the rise of online publishing does not
reduce subscriptions to print journals. For instance, according to the American
Physical Society (APS), which receives journal publishing profits, preprints of
articles in physics, computer science, and mathematics are published on arXIV.org,
an open and publicly accessible archive. The editor-in chief of the American
Physical Society, reportedly said that
there has been no decline in the subscriber base of journals in those disciplines.
In fact the ‘contrary is true,’ he said. He explained APS journals have a very
liberal copyright policy that gives back to the author the right to post articles on
e-print servers even before journal publication. They also allow authors to
update articles on the servers, using the corrected journal form, after publication69
Commercial and Open Access Publisher Practices
Proponents of open access have alleged that some traditional publishers’
practices limit equitable access to scientific information. These practices include
“restrictive licensing terms overriding copyright and fair use practices, [controls on]
long-term archival access to electronic content, and ... selective deletions of
published articles from database and e-publications.”70 Traditional, subscriber-pays
publishers often disagree and say that they are beginning to adopt some features of
open access publishing, including, but not limited to, developing multimedia
enhancements, allowing authors to self-archive their articles, and improved content
Journal Enhancements. Some traditional publishers (like many open
access publishers) have taken steps to enhance the content of journal articles they
post online by permitting digital access, permitting access to ancillary databases and
related materials, or allowing posting of preprints in author’s websites or institutional
repositories.71 However, often traditional, subscriber-pays publishers charge a fee to
view the journal article or enhancements, “... with fees ranging from a few dollars to
69 Bradie Metheny, “Public Representatives Call for Egalitarian Access to Published
Research,” Washington Fax, Aug. 10, 2004.
70 Rankin and Franklin, op. cit.
71 Guterman, Jan. 30, 2004, op. cit..
a few tens of dollars.”72 Open access proponents say that fees should not be charged
for access to these kinds of information.
Timing of Free Access to Journal Articles. Subscriber-pays, traditional
publishers have a wide variety of policies regarding free access to the articles they
publish. The British Medical Journal (BMJ), for instance, allows free access to all
readers for all materials in its journal for one week after publication. After that, non-
subscribers have free online access only to original research articles that were
published in the hard-copy journal. Only paying subscribers can access editorials and
news articles published in the journals and articles that are published only in an
online version. After a year access is free to all BMJ materials. Generally, traditional
publishers may permit free access to journal articles anywhere from a few months to
two years after publication. Proponents of open access have argued that the public
or other researchers should not have to wait a year or more to have access to research
findings, especially for biomedical research findings, that could be used to improve
a patient’s health outcome. Another view is that “... limited access to the full text of
research articles is bad for science. Such restrictions make it difficult for researchers
to build on the entirety of what has gone before and for readers to check whether they
have done so. The practice might contribute to citation bias since authors will only
reference journals they can access.”73 Still others may find that traditional publishers
do not allow electronic access to data in a form that other researchers can easily use
to verify findings or to compare in other research projects.
Self-Archiving. Open access publishers require or allow authors to self-
archive their articles immediately and to make them accessible for free. Some
traditional, subscriber-pays publishers now allow authors to self-archive on the
author’s own website an electronic version of the preprint of their article, or, after a
delay, the published journal article. There are a variety of models for this, sometimes
with fees charged. Some traditional publishers allow authors to self-archive the
preprint and then link to the printed version after publication (American
Meteorological Association); some do not allow posting of the article until a year or
more after publication in the journal (American Association for the Advancement of
Science); some allow posting of an author’s article only on an institutional or
educational server, not the author’s personal self-archive, (American Anthropological
Association); and so forth. The policies of hundreds of U.S. and foreign journals,
associations, and publishers are summarized in an inventory, published by SHERPA,
a British open access project.74
Critics say that archiving only on the author’s website makes it hard to find sets
of related articles in particular subjects because articles are more accessible when
placed in freely searchable repositories that archive articles in many fields by many
authors and which can be searched by index or keyword terms.
72 Guterman, Jan. 30, 2004, op. cit.
73 Citing others, this quote is from Pritpal S. Tamber, Fiona Godlee, and Peter Newmark,
“Open Access to Peer-reviewed Research: Making It Happen,” The Lancet, Nov. 8, 2003,
74 “Publisher Copyright Policies and Self-Archiving,” SHERPA, [http://www.sherpa.ac.uk
Commercial and Open Access Search Engines. Some commercial
publishers make available free search engines that allow readers to search for
citations or abstracts in specific fields or types of information. Some of these
repositories link to a text version which can be viewed for free. However, most full
text articles found through these searches are not accessible for free; costs to read or
download an article average $30 per article, which users or libraries are required to75
pay. An example is Scirus, a search engine limited to science literature managed
by Elsevier, which provides access to short abstracts or excerpts. Open access
bibliographic or citation archives have a wide range of policies regarding access to
scientific articles. Open access bibliographic archives generally provide free access
to abstracts or citations in multiple fields, and often to full-text manuscripts or
NIH’s PubMed/MEDLINE is a free bibliographic database that the public can
use to search for journal articles. It gives access to references from millions of
articles published in almost 5,000 biomedical journals dating back to the 1950s. It
gives a citation and abstract and links to full-text articles that are available for free
(via publishers’ websites or on PubMed Central) as well as to full-text articles on the
publishers’ websites that users may have to pay to view. NIH’s PubMed Central
(PMC) includes the full-text version of almost all articles cited in its database
(usually from open access publishers) and in limited cases has links to the full-text
versions on publishers’ websites. Many journals routinely deposit material in PMC
and generally all their published articles are made available for free. PMC also is the
repository for articles resulting from NIH-funded research that are submitted under
the agency’s Public Access Policy. These articles comprise only a small part of the
PMC database and many NIH-funded researchers publish the results of their research
in journals that do not contribute article to PMC. (PMC is discussed in greater detail
below in the section on NIH’s Public Access Policy.)
Open access proponents say that there are multiple benefits to providing free
access to articles in online repositories of collections of articles since a reader could
identify many related papers on one topic and would bypass the need to search
individual authors’ websites or to use commercial indexing databases that typically
charge a fee to read an article.
Copyright Issues. Supporters of traditional, subscriber-pays publishing
argue that publishers, as copyright holders, need copyright protection in order to
market journals and sell reprints which support the costs of publishing and archiving
both hard-copy and electronic materials. Some also say that copyright ownership is
required to guarantee a researcher’s accuracy and the authenticity of authorship of an
article. In open access publishing, the author of the article retains copyright
ownership, but access to the article normally remains free to readers. As will be
discussed below, a mixed model is used in the case of NIH’s Public Access Policy,
which asks authors to voluntarily submit to PubMed Central (PMC) the peer
reviewed version of a manuscript accepted for publication in a journal. This should
be done as soon as possible, but within 12 months of acceptance of the article by the
publisher. Free access to the manuscript is prohibited until after journal publication
75 From [http://www.scirus.com].
or for the embargo period specified by the publisher. Publishers, who hold the
copyright to articles that are not published using the open access model, retain the
exclusive right to disseminate the work for the time before free access is permitted
on PMC, but authors are encouraged to conclude agreements with publishers that
allow them to place the manuscript in the database. According to NIH, regardless of
the publisher’s decision, the agency has the right to utilize the journal article under
the government purpose license doctrine (even though NIH says it is not exercising
this authority). In the future, other agencies may seek to implement public access
policies similar to NIH’s, but may modify it to use government purpose licensing
provisions, which may be controversial. (See the section on NIH, below, for more
Open access publishing, according to many proponents, helps promote
economic, social, and technical development and equitable access to scientific
knowledge by researchers in countries unable to afford the costs of scientific journals
by hard-copy or subscription web access. Many open access systems also say that
they will waive publication charges for authors from developing counties who cannot
afford to pay to have their articles published.
But some traditional publishers say that scientists in developing countries
already have free and ready access to most scientific journals. For example, many
traditional publishers “... participate in projects sponsored by the World Health
Organization and the Food and Agriculture Organization of the United Nations to
provide medical and agricultural journals to readers in developing countries at low
or no cost.”76 In addition, more than 2,000 biomedical journals are accessible online
to researchers and health workers in developing countries via a philanthropic project
called Health InterNework Access to Research Initiative (HINARI) supported in
collaboration with the World Health Organization.77
Peer Review and Quality of Articles In Open Access Journals
There is a diversity of views about whether the articles that appear in open
access journals have been subject to the same kind of rigorous peer review as those
published in traditional, subscriber-pays journals and about whether they are of
comparable quality. The peer review process traditionally involves review of quality
of the article and selection of articles to be published in a journal. Usually journal
editors or editorial boards make an initial selection of articles to be peer reviewed
from among those submitted; use a panel of expert scientists who may volunteer their
time to review submissions; select articles to be published from among the articles
peer reviewers ranked as high quality; and sometimes do some editing.
76 Lila Guterman, Jan. 30, 2004, op. cit.
77 Brian D. Crawford, “Open-access Publishing: Where is the Value?” The Lancet, Nov. 8,
A long-held principle is that the accumulation of high-quality scientific
knowledge rests on a foundation of publication, typically in traditional commercially
distributed scientific journals, with the findings and results vetted and validated
through a process involving peer review and fee-based journal subscriptions. Critics
allege that the open access “author pays” model of paying for publishing costs,
including peer review, prevents quality control mechanisms from working correctly
and that, in the long run, scientific articles published in open access sources may be
less credible than those published in journals which charge subscription fees.78 A
survey published in 2005, funded by traditional, subscriber-pays journal publishers,
is reported to have found that the quality of peer review was lower in open access
than in traditionally published journals:
Open-access journals ... received fewer submissions and were less selective in
choosing among submissions. [It continued] essentially all of the journals
reported using editorial review to select and edit submissions. But nearly all of
the traditional journals used external peer review, while only editorial staff79
members reviewed submissions of about 30 percent of the open-access journals.
According to another study, the most rigorous peer review, “as measured by
their [journals’] reliance on external reviewers,” was largely by traditional publishers,
and that, in contrast, “full open access journals tended to depend heavily on editorial
staff only for peer review,” except for two subsets of open access journals — BioMed
Central (BMC) and Internet Scientific Publications (ISP) journals, which had
practices more like traditional journals.80 On the other hand, a study published in
2005 by a publishing analysis firm showed that the quality of nearly 200 open access
journals was almost as high in specific medical disciplines as the quality of articles
in traditionally published journals.81
Some analysts say that peer review in open access journals suffers from the
difficulty of finding enough scientist peer reviewers for both the growing number of
open access journals and traditional journals. There is also the view that editorial
boards of open access journals, may not filter out unacceptable manuscripts as much
as traditional, subscriber-pays journal boards do. Thus peer reviewers for open
access journals, who interact and report primarily electronically, may be
overwhelmed by the number of articles they are given to review, and, ultimately,
there may be delay in the system. Publication in peer reviewed journals figures
prominently in promotion and tenure processes in academia. Some observers
contend that members of the academic and scientific communities may not view
78 Crawford, op. cit.
79 Lila Guterman, “New Study Compares Open-Access and Traditional Publishing,”
Chronicle of Higher Education, Mar. 25, 2005.
80 OECD, “The Facts About Open Access,” op. cit., p. 25. See also: Janet Coleman,
“Financial Future of Open Access Journals Uncertain, Study Finds: 41% in Red, But
Revenues Growing,” Research Policy Alert, Oct. 13, 2005.
81 Alison McCook, “Open-access Journals Rank Well,”The Scientist, Apr, 27, 2005.
publication on the Internet or in an open access journal to be as prestigious as
publication in a traditional, subscriber-pays peer reviewed journal.82
Others use citation data as a surrogate measure for quality. Some analysts cite
data showing that articles posted in open access journals or freely available on the
Internet are used and cited more frequently than those published in traditional
journals and are, therefore, a better model to ensure the speedy utilization of
scientific research. For instance,
!Experience in physics where researchers publish in traditional
journals and then self-archive their papers in a free database is
conducive to scientific communication and favorable to authors
because “papers listed in free archives often get more citations....”83
!A recent study showed that in four disciplines, philosophy, political
science, electrical and electronic engineering, and mathematics,
articles that are freely available via open access publishing have a
greater research impact than those not available via open access.
Impact is measured by citations made by other researchers to the
literature in the ISI Web of Science database.84
!In computer sciences, “a 2001 study in Nature, showed that, at least
in one set of disciplines, papers that appear free online are more
likely to be cited by other researchers than those that do not. A
scientist at NEC Research Institute analyzed nearly 120,000 papers
in computer science and related titles. Those that were freely
available online had been cited more often in other papers than were
those not online, he found. The average number of citations of
offline papers was 2.74, compared with 7.03 for those freely
!A study published in an open access journal, suggested that articles
published online in open access journal got cited more often than
those cited in subscriber or pay for view journals. The articles
examined were published in the Proceedings of the National
Academy of Sciences in 2004; authors paid $1000 to allow their
papers to be read immediately and without cost.86
82 Points raised in the discussion session of a meeting on “The Future of Scientific
Communication (Formerly Known as Publishing),” American Association for the
Advancement of Science (AAAS) Meeting, Apr. 21, 2005.
83 Alison McCook, “Open Access to U.S. Govt. Work Urged,” The Scientist, July 21, 2004.
84 Kristin Antelman, “Do Open-Access Articles Have a Greater Research Impact?,” College
and Research Libraries, vol. 65, no. 5, pp. 372-282 (Available via E-LIS).
85 Lila Guterman, Jan. 30, 2004, op. cit.
86 Gunther Eysenbach, “Citation Advantage of Open Access Articles,” PLoS Biology, vol.
One implication of these data should be noted. Ease of access to articles readily
available online, as opposed to those that may be accessible only in hard-copy
journals, may increase the propensity to cite them. Thus citation data may not so
much measure quality as accessibility.
“Enhanced Public Access Policy”: National
Institutes of Health (NIH) and Other Agencies
On June 26, 2003, Representative Martin O. Sabo introduced the “Public Access
to Science Act” (H.R. 2613, 108th Congress), which would have denied copyright
protection to publications resulting from federally funded basic scientific research in
order to encourage free dissemination of research results to the public.87 No action
was taken on this bill.
Legislative Origins of NIH Policy
Subsequently, the House Appropriations Committee’s report on the FY2005
Labor/HHS bill, H.R. 5006, July 14, 2004, contained language that led to the NIH’s
“Enhanced Public Access Policy” (H.Rept. 108-636, p. 104). The language, reported
to have been authored by Representative Ernest J. Istook, Jr.,88 “recommended” that
NIH permit open access to NIH-funded research by “requiring” researchers to deposit
peer reviewed articles accepted for publication and associated supplemental materials
in NIH’s PubMed Central, an free access repository information system, within six
months after publication of the article in a scientific journal. If NIH awarded funds
for publishing, the research would be made available immediately upon publication.
It also instructed NIH to draft a report by December 1, 2004 on how it would
implement this policy. Reportedly “librarians and the Scholarly Publishing and
Academic Resources Coalition, or SPARC,” lobbied “the Appropriations Committee
87 It proposed to “Amend ... Federal copyright law to declare copyright protection
unavailable to any work produced pursuant to scientific research substantially funded by the
Federal Government to the extent provided in the funding agreement entered into by the
relevant Federal agency pursuant to this Act; Require ... any Federal department or agency
that enters into a funding agreement with any person for the performance of scientific
research substantially funded by the Federal Government to include in the agreement a
statement that copyright protection is not available for any work produced pursuant to such
research under the agreement; and express the sense of Congress that any Federal
department or agency that enters into such funding agreements should make every effort to
develop and support mechanisms for making the published results of the research conducted
pursuant to the agreements freely and easily available to the scientific community, the
private sector, physicians, and the public.” (CRS Summary).
88 See “Departments of Labor, Health and Human Services, and Education, and Related
Agencies Appropriations Act, 2005,” Remarks of Rep. Ernest Istook on the Floor of the
House, Congressional Record, Sept. 8, 2004, p. H6833; Andrew J. Hawkins, “Istook Will
Clarify NIH Open Access Publishing Language Intent on House Floor,” Washington Fax,
Aug. 31, 2004; and Jocelyn Kaiser, “... Congress Puts Similar Heat on NIH,” Science, July
behind the scenes to include the open-access language in the committee’s report ....”89
The conference report on the FY2005 Consolidated Appropriations Act, P.L.
108-447 (H.Rept. 108-792, p. 1177), which included funds for Labor/HHS, directed
NIH to consider input from publishers as it developed its public access policy,
directed NIH to continue to work with publishers to insure the integrity of the peer
review system, and requested that NIH “... provide the estimated costs of
implementing this policy each year in its annual budget justification ...” in response
to concerns from publishers that NIH’s database cannot easily handle the new articles
it will be required to archive.90
NIH’s Public Access Policy and NIH’s PubMed Central (PMC)
NIH maintains a database, called PubMed Central (PMC). This is an electronic
system that was launched in 2000 and which contains bibliographic citations and the
full text of some several thousand peer-reviewed articles that were published in91
journals in the fields of biomedical, behavioral, and clinical research. The goal of
the database system is to develop a publicly accessible, permanent, and searchable
electronic archive of life science literature, that is separate from publishers’
individual databases Some of the journals participating in PMC (typically open
access publishing journals) make full text articles available to users immediately
upon publication; some require waiting periods of up to three years to obtain free
access to full text of an article. Some journals which participate allow the
publication of bibliographic information about an article, but require the reader to
link to the journal’s website to view the abstract or full article, which may be
accessible only to subscribers or for a fee. Many major biomedical journals do not
participate in submitting materials to PMC, including such journals as the Journal
of the American Medical Association, the American Psychologist, the Journal of
Abnormal Psychology, and the New England Journal of Medicine. Electronic access
to articles in these journals is usually available for a fee — sometimes over $25 per
article — at the journal’s website. Publishers may make the text of articles published
in some of these journals available to readers for free, but usually only on a delayed
basis averaging about 12 months after publication in the journal.
In response to congressional mandate, NIH’s policy to archive published articles
that resulted from its funding was released for public review and comment in
September 2004.92 After holding several meetings with stakeholders and considering
89 Andrea L. Foster, “House Committee Tells NIH to Post Research Results Online and
Make Them Free,” Chronicle of Higher Education, July 19, 2004.
90 Shirley Haley, “Omnibus Report Language on Open Access Called a Win By Scientific
Societies,” Washington Fax, Dec. 7, 2004.
91 Personal communication with NIH official, Oct. 23, 2006.
92 The proposed “NIH Public Access Policy,” which solicited comments, appeared in the
NIH Guide for Grants and Contracts on September 3, 2004 [http://grants.nih.gov/
grants/guide/notice-files/NOT-OD-04-064.html] and in the Federal Register on September
numerous comments from traditional publishers and others submitted during the
public comment period,93 NIH issued the final policy, which was published in the
Federal Register on February 3, 2005.94 Implementation of the policy started on May
It asks authors funded by NIH to voluntarily submit as soon as possible to NIH
for inclusion in the NIH PMC system, manuscripts that have been edited through the
peer review process and accepted for journal publication. Such manuscripts are to
be submitted as soon as possible, but within 12 months after acceptance of the article
by a scientific journal (instead of six months as originally proposed). They are
supposed to be posted and made available for public viewing after the embargo
period, or sooner if the publisher agrees, but within 12 months. According to NIH,
the requirement is not mandatory and no penalties would be imposed if an author did
not submit a manuscript to the free archive.95 Thus, NIH-funded scientists are asked
...submit an electronic version of the author’s final manuscript, upon acceptance
for publication, resulting from research supported in whole or in part by NIH.
The author’s final manuscript is defined as the final version accepted for journal
publication, and includes all modifications from the publishing peer review
process. The policy gives authors the flexibility to designate a specific time
frame for public release — ranging from immediate public access after final
publication to a 12 month delay — when they submit their manuscripts to NIH.
93 These are described on the NIH website at [http://www.nih.gov/about/publicaccess/] and
are summarized in many articles, such as: Jocelyn Kaiser, “Seeking Advice on ‘Open
Access,’ NIH Get an Earful,” Science, August 6. 2004; Bradie Metheny, “Public
Representatives Call for Egalitarian Access to Published Research,” Washington Fax, Aug.
10, 2004; Andrew Hawkins, “Open Access Should Be A ‘Cooperative Venture’ Between
NIH and Journals, NAS Urges,” Washington Fax, Nov. 15, 2004; Meredith Wadman,
“Director Hits back at Critics of Free Archive Plan,” Nature, Nov. 25, 2004; M.T.
Cavanaugh, “Open Doors: All NIH-funded Work Could Be Freely Available,” Nature, Nov.
25, 2004; Shirley Haley, “Publishing Delegation Offers Advice, Alternatives to NIH
Director on Open Access Plan,” Washington Fax, Nov. 5, 2004; Lila Guterman, “NIH
Proceeds With Plan to Provide Open Access to Scientific Papers,” Chronicle of Higher
Education, Sept. 1, 2004; Andrew Hawkins, “Publishers Argue for Public Access
Flexibility, Links to Journals,” Washington Fax, Nov. 19, 2004; Andrew Hawkins, “Public
Access Will Harm Journal/NIH Relationship, AAI Charges; Advocates Dispute Legal
Analysis,” Washington Fax, Nov. 22, 2004; Jocelyn Kaiser, “ NIH Unveils Public Access
Policy,” Science, Feb. 3, 2005; Andrew J. Hawkins, “NIH Says Public Access Policy will
Change How Science Is Understood,” Washington Fax, May 2, 2005.
94 “Policy on Enhancing Public Access to Archives Publications Resulting from NIH-Funded
Research,” Federal Register, Feb. 9, 2005, v. 70, no. 26, pp. 6891-6900.
95 NIH said in section P of the Federal Register rule, that while the House Appropriations
report proposed requiring submission, the NIH policy requesting rather than requiring
submission “is consistent with the final report language found on page 1177 of the Joint
Explanatory Statement in H.Rept. 108-792.” See also: NIH. “Questions and Answers: NIH
Public Access Policy,” Feb. 2005.
Authors are strongly encouraged to exercise their right to specify that their
articles will be publicly available through PubMed Central (PMC) as soon as96
The version required to be submitted voluntarily is not the final version of the
article as copyedited and printed in the journal. Since publishers use different
formats for publishing materials electronically, NIH is using a standardized format97
to archive and make accessible the submitted manuscript in PMC. NIH’s policy says
that it would accommodate any changes made to the manuscript by the publisher if
submitted to PMC and that manuscripts would not be made available from PMC until
after the article was published in a journal. PMC will provide a link to the
publisher’s website (which could possibly charge a fee for viewing) to enable the
public to read the article as published in a journal. Specifically,
. . . under the Policy, the final manuscript will not be made available to the public
through PMC until after the copyedited version is published by the journal.
Corrections and other necessary revisions of author’s final manuscripts will be
accommodated. Furthermore, when publicly available, the published article on
the journal-sponsored website and the author’s final manuscript in PMC will be
appropriately linked through PubMed. Corrections and post-publication
comments referring to a publication are currently identified and linked in
PubMed, and this capability will be linked to the corresponding manuscript in
PMC. If publishers wish to provide PMC with the publisher’s final version, this98
version will supersede the author’s final manuscript in PMC.
NIH allows researcher/authors to use the submission of the manuscript to meet
certain NIH grant reporting requirements.99 According to NIH, its policy is
compatible with existing publishing models. The agency said it,
examined the access policies of the top 20 journals based on citation impact for
medicine and medical research and of the 50 journals published by members of
FASEB [Federation of American Societies for Experimental Biology]. As of
October 2004, 80% of the 20 high impact journals allow public access of some
sort through HighWire press within 12 months of publication; of the 50 FASEB
journals, 78% offer public access within 12 months.”100
NIH Director Zerhouni justified the new policy by explaining that it provides
electronic access to NIH-funded research, permits formation of a central archive of
NIH-funded research publications, advances science by creating an information
96 “NIH Calls on Scientists to Speed Public Release of Research Publications,” NIH News,
Feb. 3, 2005.
97 This format is known as the NLM Journal Article Extensible Markup Language (XML)
Document Type Definition (DTD).
98 Federal Register, Feb. 9, 2005, op. cit., pp. 6893-6894.
99 NIH, “Final NIH Public Access Policy Implementation,”March 15, 2005. The database
is available at [http://www.pubmedcentral.gov/].
100 NIH Director Zerhouni, “NIH: Advancing Science in the 21st Century,” Mar. 24, 2005,
at FLICC Forum on Federal Information Policy, Library of Congress.
resource that scientists can mine, and helps NIH “better manage its entire research
NIH has also created a Public Access Advisory Working Group of the National
Library of Medicine (NLM) Board of Regents, composed of stakeholders to advise
NIH and NLM on policy implementation and evaluation. Modifications are to be
made to the system as it becomes operational and is studied by the group. The NIH
Public Access Advisory Group met on November 15, 2005. Among its
recommendations was that the NIH policy, which is now voluntary, be made
mandatory; that manuscripts be posted within six months, instead of the current 12
months; and that the final copyedited version be posted, instead of the author’s final
In November 2005, in response to several publishers’ concerns, NIH revised the
existing public access policy to allow publishers, in addition to authors, to request
that articles which infringe copyright be removed from PubMed Central, even though
the author has the copyright agreement with a publisher, and the public access policy
agreement is between an author and NIH. Such infringement might occur if a
publisher has not granted permission for an article to be displayed in PMC before 12
months has elapsed or if the author provided NIH with a final copyedited version of
the article, which a publisher might oppose.103
According to minutes of the Public Access Advisory Working Group’s April 10,
the majority of members confirmed the opinions expressed at the previous
(November 15, 2005) meeting of the working group: (1) the policy should be
mandatory; (2) submission should occur within six months (with flexibility to 12
months in the case of journals that publish quarterly or less frequently); and (3)
the final manuscripts as published should be the favored form. A minority favors
a 12 month submission deadline and submission of the author’s final manuscript104
rather than the final published form.
As announced on September 1, 2006, NIH released a press release describing
a modification to the existing implementation process for its open access policy.105
101 Zerhouni, op. cit., Mar. 24, 2005.
102 Janet Coleman, “NIH Public Access Policy Should Be Mandatory, Advisors
Recommend,” Research Policy Alert, Nov. 17, 2005. See also Letter from Thomas Detre,
Chair, Board of Regents, National Library of Medicine, to Dr. Elias A. Zerhouni, Feb. 8,
103 Janet Coleman, “Revised NIH Public Access Policy Allows Publishers to Request
Removal of Articles That Infringe Copyright,” Research Policy Alert, Nov. 18, 2005.
104 “NIH Public Access Working Group of the NLM Board of Regents Meeting Summary
April 10, 2006,” [http://www.nlm.nih.gov/od/bor/PublicAccessWG-April%202006.pdf.
105 NIH, Office of Extramural Research, “NIH Public Access: Journals That Submit
Manuscripts on Behalf of Authors,” Press announcement, Sept. 1, 2006
[http://publicaccess.nih.gov/submit_process_journals.htm]. See also: “NIH Partners With
The policy says essentially that an author does not have to directly submit a
manuscript to PMC if he or she publishes in a journal which automatically deposits
all of its content on PMC and makes its contents available to the public (which is
most of the journals that partner with PMC). “All but a handful” of Public Med
Central journals permit free access usually immediately to the final version of an
article and authors who publish with them do not have to take further steps to satisfy
NIH public access policy. Seven of the almost 300 journals accessible via PMC
require the author to submit manuscripts.
At this time, NIH also initiated a new system called the PubMed Central (NIH
Portfolio) project — only for NIH-funded research — which is apparently designed
to satisfy NIH needs and the demands of some nonprofit publishers. Although a
signed agreement is not necessary,106 some publishers have signed an agreement for
participation with NIH. According to one publisher, it stipulates that published
journal articles resulting from NIH-funded research be made available only for
internal use in an NIH-funded archive during the embargo period, that the embargo
period last no longer than 12 months, and that following the embargo period NIH
could provide links to the journal and could also distribute the article directly through
PMC.107 If a journal does submit articles via the bulk system, authors have to
confirm the version that is posted on PMC. So far only one journal, Blood, has
agreed to participate, but negotiations are underway with other publishers.108 Authors
that publish in any other journal not considered a regular PMC journal or NIH
portfolio journal (identified by NIH, for example, as Elsevier journals) need to
continue to submit manuscripts to comply with NIH’s policy to submit final
manuscripts to PMC.109 These changes occurred after several months of discussions
intended in part to allay criticisms110 (dealing primarily with abridging a publisher’s
embargo periods before submitting a final journal article, increasing compliance by
NIH-funded researchers, and averting additional action to mandate compliance).
Journal Publishers to Facilitate Participation in NIH Public Access,” NIH Extramural Nexus,
September 2006. According to a personal communication with an NIH official, Oct. 23,
2006, although NIH had the infrastructure to allow publishers to submit author manuscripts
in bulk since Dec. 2005, publishers apparently did not begin to use the system until Sept.
106 Personal communication from NIH Office of extramural Research, Oct. 23, 2006.
107 Blood, Journal of the American Society of Hematology, “ASH’s New Alternative to NIH
Policy on Public Access,” Sept. 7, 2006, available at [http://bloodjournal.org
/preview_misc/ASHalternativeNIH.shtml] hot linked from [http://publicaccess
.nih.gov/submit_process_j ournals.htm] .
108 Interview with NIH official, Oct. 10, 2006.
109 “NIH Partners With Journal Publishers to Facilitate Participation in NIH Public Access,”
NIH Extramural News, September 2006.
110 Early discussions are summarized in Janet Coleman, “NIH Public Access Discussions
With Publishers Proceeding, But Obstacles Remain,” Research Policy Alert, Apr. 13, 2006.
Legislative Action in the 109th Congress
On June 21, 2005, the House Appropriations Committee approved H.Rept. 109-
3010). The House bill was passed on June 24, 2005. The report endorsed NIH’s
objectives in establishing the “Public Access Policy” and included language requiring
NIH to develop an “aggressive” outreach program to ensure full participation by
grantees in volunteering to submit their journal manuscripts to the NIH archive. It
also requested the NIH Director to report to Congress by March 1, 2006 on the
number of “articles”111 deposited and the length of the embargo by publishers — that
is, the delay between submission of each peer reviewed “article” to NIH and its
subsequent posting on the PubMed Central website — and to estimate the total
number of articles available for deposit.112 S.Rept. 109-103 on this bill endorsed the
objectives of the policy but also emphasized the need for interaction between NIH
and stakeholders. It urged NIH to work with stakeholders as it implements the new
policy; and asked NIH to report by February 1, 2006 on the number of peer reviewed
“articles” deposited in the database, on “the extent to which the implemented policy
has led to improved public access,” on the impact on the peer review system, and on
the cost of operating the database.113 The bill enacted after conference committee
action was sent to the President for signature on December 28, 2005 (signed as P.L.
109-149). The NIH report to the committees was released in January 2006.114 It
reported that for the first eight months of the system, the rate of submission was low,
less than 4% of the total number of articles estimated to be eligible for submission,
that is 1,636 out of about 43,000 that could be deposited. Lack of awareness, it
reported, does not appear to be the primary reason for the low submission rate. The
report did not describe the reasons for the low rate of participation, but publisher
resistance seems apparent. The NIH said it will continue to work with participants
and stakeholders to improve public access. It also identified three issues the working
group was continuing to examine, which undoubtedly contribute the low
!Should investigators’ participation in the Policy be mandatory or
!Which version of an article should be deposited in PMC: the author’s final
peer reviewed manuscript, or the final, edited article as it is published in
111 The word “articles” is used in the House and Senate reports even though the NIH policy
as it appears in the Federal Register and in NIH Notice Number NOT-OD-05-022, uses the
words “final manuscript”, which is “defined as the final version accepted for journal
publication, and includes all modifications from the publishing peer review process.”
112 H.Rept. 109-143, op. cit., p. 104. See also Jocelyn Kaiser, “House Approves 0.5% Raise
for NIH, Comments on Database,” Science, June 17, 2005.
113 S.Rept. 109-103, op. cit., p. 159.
114 National Institutes of Health, Report on NIH Public Access Policy, January 2006, 8 p.
Available at [http://publicaccess.nih.gov/Final_Report_20060201.pdf].
!What should be the length of the embargo period before public access to115
an article is permitted through PMC?
In 2006, the House Appropriations Subcommittee on Labor, Health and Human
Services, and Education again addressed the issue of NIH’s public access policy.
Testimony on the issue was delivered at an April 6, 2006 subcommittee hearing. The
full Appropriations Committee reported an original bill, H.R. 5647 on June 20, 2006.
Sec. 220 would change NIH’s policy to make it mandatory that all NIH-funded
researchers submit electronic versions of final, peer-reviewed manuscripts to
PubMed Central within one year of acceptance for publication in a journal. No
further action has occurred. The Senate bill, S. 3708, does not contain this language.
Both committee reports on the bills contain language which commends the PMC
repository and observe that NIH is continuing to work with researchers, publishers,
societies and other stakeholder to improve public access.116
The American Center for Cures Act, S. 2104, was introduced on December 14,
2005, by principal sponsors, Senators Joe Lieberman and Thad Cochran; it contains
a provision on translational research, including a section requiring NIH grantees to
provide NIH with a final version of all peer-reviewed manuscripts accepted for
publication within six months from date of publication.117 According to the
American Psychological Association, a Member of Congress had planned to, but did
not finally, introduce an amendment during committee consideration of H.R. 6164,
the NIH Reauthorization Act, “that would have required all journal articles about
federally funded research to be deposited in a free, open archive (NIH’s Pub Med)
no later than six months after they were accepted for publication.”118 The bill was
reported out of the authorizing committee, the House Committee on Energy and
Commerce, and approved in the House on September 26, 2006. According to the
report, “The Committee has listened to stakeholder concerns about NIH’s current
open access policy with respect to making published literature available online. The
Committee will continue to monitor the open access policies adopted by the NIH,
including the management of the program and the participation levels of scientific
journals” (H.Rept. 109-687, pp. 22-23). The Senate passed H.R. 6164, amended,
115 National Institutes of Health, “Report on the NIH Public Access Policy,” January 2006,
116 H.Rept. 109-515, p. 121, and S.Rept. 109-287, p. 157.
117 In summary, “Section 499H-1. Publication Requirement for Research: The Director of
the NIH shall require that for any research funded by the NIH, Centers for Disease Control
and Prevention (CDC), and the Agency for Healthcare Research and Quality (AHRQ), there
will be a standardized report of this research for public viewing. Department of Health and
Human Services (DHHS) grantees shall provide the NLM an electronic copy of the final
version of all peer-reviewed manuscripts accepted for publication for display on their digital
library archive, PubMed Central, within 6 months from the date of its publication.” Source:
Statement upon introduction of S. 2104, American Center for Cures Act, by Mr. Reid (For
Mr. Lieberman (For Himself, Mr. Cochran, Mr. Carper, and Mrs. Hutchison), Congressional
Record, Dec. 14, 2005, p. S. 13577ff.
118 “Open Access Amendment to NIH Reform Act is Withdrawn,” [SPIN] APA’s Science
Policy Insider News; September, 2006.
without written report on December 8, 2006; the House agreed to the amended bill
on December 9; and the bill was sent to the President on December 9.
Reportedly, some Members of Congress have supported the position of major
opponents of the NIH public access policy. Senators Larry Craig, Mike Crapo, and
Kit Bond, according to a news article, sent a letter to NIH Director Zerhouni on
November 18, 2005, which supported the FASEB group position119 of having NIH
post abstracts which are linked to publishers’ websites to read the full text of articles.
The Senators also questioned NIH’s ability to fund the public access system due to
limited resources and requested that NIH meet with representatives of the group to
consider their proposal. Dr. Zerhouni reportedly said he would welcome a
Criticisms of “NIH’s Enhanced Public Access Policy”
Criticisms of the NIH policy have come from traditional, subscriber-pays
publishers as well as proponents of open access.
For instance, PLoS’s supporters have criticized the NIH policy for its voluntary
compliance requirement and said “... the agency’s language should have been to
‘require’ or ‘expect’ rather than ‘request’ the deposition of NIH-funded articles in the
National Library of Medicine’s free-to-use Internet repository, PubMed Central.”121
In addition, according to PLoS “... the maximum allowable delay before articles’
public release should have been at most 6, rather than 12 months — particularly since
no publisher has presented evidence that the free availability of a fraction of its
journals’ articles half a year after publication would adversely affect subscription
revenues.”122 Others say that the 12-month delay for public access falls short of
achieving goals of congressional intent and is too lengthy “in a field as dynamic as
biomedicine,” where patients need immediate access.123
NIH policy has also been criticized by some who say that NIH should utilize
free access policies that exist in the not-for-profit publishing community, which, they
say, are more cost-effective. They suggest that instead of putting articles in PMC,
NIH should create a search engine that has the capability to crawl the full texts of
existing journals, including nonprofit journals, to allow access to articles on the
original journal’s website and to provide access to other articles on the topic.
Publishers often charge a fee to access articles this way. Among the groups who have
119 Described further below.
120 Coleman, Dec. 20, 2005, op. cit.
121 Andy Gass and Helen Doyle, “PLoS Position on NIH Public Access Policy,” Letter to
the Editor, Science, Apr. 15, 2005, p. 352.
122 Gass and Doyle, Apr. 15, 2005, op.cit.
123 Comments made by the Alliance for Taxpayer Access as cited in Andrew J. Hawkins,
“NIH Public Access Policy Unenforceable, Violates Copyrights, Opponents Charge,”
Washington Fax, Feb. 7, 2005.
commented on this position is the Washington DC Principles for Free Access to
Science124 and the American Physiological Society.125
By way of example, Google Scholar,126 which was launched in 2004, is a free
Internet search engine that allows readers to search for peer reviewed articles,
preprints, abstracts, grey unpublished literature and other scholarly analyses. If it
links to a full-text article, the article is likely to have been published at least a year
before the date of the search. There is no assurance that the search engine captures
all current or archived materials available in a field. Full text of publisher-controlled,
copyrighted materials may be indexed with a citation, but a reader may be linked to
the publisher’s website to obtain full text of the published version for a fee. In
addition, there may be a direct link to the full text of a preprint or a version posted
by an author or university archive website.
Some focus on the notion that NIH policy may promote the forfeiture of patent
rights. A legal analysis contends that pre-publication “manuscripts placed on the
PMC database ‘likely’ can be considered ‘printed publications’ for patent purposes,
thus ‘triggering the one-year time period for filing a U.S. patent application covering
research disclosed in the manuscript ....’ “127 “Current practice,” it is charged, “relies
on the date of journal publication to start the clock.”128
A report prepared for the American Physiological Society criticized the NIH
policy as limiting technology development and commercial competition, specifically
that “the open access plan ‘undermines the principle of [Bayh-Dole] that the private
sector is the preferable vehicle to move federally-funded research results to the public
124 Washington DC Principles for Free Access to Science, “Nor-for-Profit Publishers Call
New NIH Rule a Missed Opportunity,” available at [http://www.dcprinciples.org/
125 Haley, op. cit., Nov. 18, 2004, citing a legal analysis by Foley and Lardner, law firm.
126 Available at [http://scholar.google.com/].
127 Shirley Haley, “ Open Access Plan Faces Copyright, Regulatory Compliance Questions,
Legal Analysis Finds,” Washington Fax, Nov. 18, 2004.
128 Haley, op. cit. For other criticisms, see Jocelyn Kaiser, “Seeking Advice on ‘Open
Access,’ NIH Gets an Earful,” Science, Aug. 6, 2004; John T. Softcheck, “PubMed Central’s
Capacity to Host Open Access Articles Concerns ASM [American Society for
Microbiology], Washington Fax, Sept. 1, 2004; Danielle Belopotosky, “Online Federal
Library on Health Research Sparks Outcry,” Government Exec. Com, Sept. 3, 2004; Jeffrey
Young, “Journal Publishers Ask Senate to Intervene Against NIH Open Access Policy,”
Washington Fax, Sept. 10, 2004; Jeffrey Young, “ ‘Unnecessary’ NIH Open Access
Proposal Should Be Discarded, FASEB [Federation of American Societies for Experimental
Biology] Says,” Washington Fax, Nov. 5, 2004; Andrew Hawkins, “Public Access Will
Harm Journal/NIH Relationship, AAI [American Association of Immunologists] Charges;
Advocates Dispute Legal Analysis,” Washington Fax, Nov. 22, 2004.
and the marketplace.’ “129 It should be noted that the Bayh-Dole law applies to
technology transfer, not to publishing of research results.
According to NIH officials, voluntary participation in the public access system
has been very limited: only about 4% to 5% of articles by NIH grantees have been
submitted,130 and a survey (by a publisher group) contends that only about 18% of
NIH grantees understand how to submit a manuscript for posting in the public access
archive.131 In response, the Federation of American Societies for Experimental
Biology (FASEB), which opposes the NIH policy as configured, proposed that the
public access policy be modified so that NIH links readers from abstracts of articles
to publishers’ websites to read an article, rather than to the article itself,132 and that
NIH create an archive of full text of articles for internal NIH use only. NIH officials
are reported to have objected to this proposal, saying it would prevent achieving the
policy’s three core goals: a stable and permanent archive, an archive available to
awardees to help communicate research findings, and an archive accessible to the
Legislative Proposal to Extend Open Access Policies to
Other Agencies: The Federal Research Public Access Act
S. 2695, the Federal Research Public Access Act of 2006, was introduced on
May 2, 2006, co-sponsored by Senator John Cornyn and Senator Lieberman. It
requires all federal departments and agencies that invest $100 million or more
annually in research to develop a public access policy that requires all final
manuscripts or articles that result from federal funding to be posted in a free publicly
accessible archive as soon as possible but no later than six months after
129 Based on a legal analysis of the technology transfer implications of the NIH proposal by
a Foley and Larnder law firm analysis for the American Physiological Society, as reported
in Haley, Nov. 18, 2004. The Bayh-Dole act, (35 USC 200-212) allows the government to
transfer control of a federally funded invention to a university or business to promote
commercialization; the government can license the invention to a third party if it believes
it is not being made publicly available on a reasonable basis. See also CRS Report RL32076,
The Bayh-Dole Act: Selected Issues in Patent Policy and the Commercialization of
Technology, by Wendy H. Schacht.
130 Statistic attributed to NIH Director Elias Zerhouni. See Janet Coleman, “NIH Grantees
Compliance With Public Access Policy Will Take Time, Zerhouni Says,” Research Policy
Alert, Nov. 10, 2005.
131 Andrew J. Hawkins, “Limited Understanding of NIH Public Access Policy Found Among
Researchers,” Research Policy Alert, Mar. 6, 2006. The original survey is NIH Author
Postings, A Study to Assess Understanding of, and Compliance With, NIH Public Access
Policy, Report on Behalf of the Publishing Research Consortium, Feb. 21, 2006, 45 p.
132 Eugene Russo, “FASEB Urges NIH to Adopt New Public Access Policy,” Research
Policy Alert, Oct. 19, 2005.
133 Janet Coleman, “FASEB Public Access Proposal Would Prevent NIH From Meeting Core
Goals, Agency Says,” Research Policy Alert, Nov. 21, 2005.
publication.134 The bill also would require agencies to “make effective use of any law
or guidance relating to the creation and reservation of a Government license that
provides for the reproduction, publication, release, or other uses of a final manuscript
for Federal purposes” (Sec. 4 (c)). The following could be among the agencies
affected, the National Science Foundation, the Department of Energy, the
Department of Transportation, the Department of Defense, and the National
Aeronautics and Space Administration. This proposal, like the NIH public access
policy, has generated considerable reaction. In July, 2006, the provosts of 25
universities, including Harvard, the University of Chicago, and the University of
California, jointly released “An Open Letter to the Higher Education Community,”
supporting the bill as “good for education and good for research.”135 Subsequently
the presidents of 53 liberal arts colleges, organized by the president of Oberlin
College, issued a joint letter supporting the legislation.136 Several library groups have
also supported this proposal.137 Additional support has come from major New
England university provosts.138 Some scholarly associations, academics,139 and
publishers objected on the same grounds as objections to the NIH policy — for
instance, that the costs of a broader policy would detract from research spending, the
government might not maintain databases, some journals would be forced to close
for lack of income, and the government should not interfere in private activities by
creating such publication databases.140
On October 20, 2006 a forum was held on “Improving Access to Publicly
Funded Research,” cosponsored by leaders of higher education and library
organizations, including the Association of American Universities (AAU), the
Association of Research Libraries (ARL), the Coalition for Networked Information
(CNI), the National Association of State Universities and Land-Grant College
(NASULGC), and the Scholarly Publishing and Academic Resources Coalition
134 Rick Weiss, “Bill Seeks Access to Tax-Funded Research; Grant Recipients Would Be
Required to Post Findings on Internet,” Washington Post, May 3, 2006, p. A21. See also,
Janet Coleman, “Mandatory NIH Public Access Policy, With Six-Month Posting, Sparking
Congressional Interest,” Research Policy Alert, Dec. 20, 2005.
135 Scott Jaschik, “Rallying Behind Open Access,” Inside Higher Ed,, July 28, 2006
[ h t t p : / / wwww.i n si dehi gher ed.com/ news/ 2006/ 07/ 28/ pr ovost s ] .
136 Scott Jaschik, “Momentum for Open Access Research,” Inside Higher Ed, Sept. 6, 2006.
137 “University Support for Public Access Act Expands,” August. 3, 2006, at
[ h t t p : / / www.ar l .or g/ spar c/ oa/ Li b r a r yGr oupsComme ndPr o vo st s_06AUG.pdf].
See also, [http://www.taxpayeraccess.org/frpaa]
138 “New England Provosts Call for Broader Access to Publicly Funded Research.” Sept. 19,
139 See, for instance, Letter Sent to Senator Cornyn from Academic Officials, Sept. 22, 2006,
reprinted by the Washington D.C. Principles for Free Access to Science, c/o The American
Physiological Society, Bethesda, MD.
140 Jaschik, op. cit.
(SPARC). Many of the speakers supported S. 2695 and open access publishing of
federally funded research.141
Government Purpose and Copyright Issues
NIH documents indicate that its Public Access policy upholds the principles of
copyright since submission of manuscripts is voluntary and the statutory fair use
privilege still applies to public use of the archived articles. The agency issued
guidelines for authors on how to include, in a copyright agreement with a publisher,
language that acknowledges the author’s obligation to provide a copy of the article
NIH relies on obtaining permission from authors as the basis for its policy even
though “NIH does not need to seek permission from journals who may acquire
copyrights from authors or institutions because any copyright transfer or assignment
is currently subject to the government purpose license pursuant to 45 C.F.R.
74.36.”143 The term “government purpose license” is not used per se in the cited
regulation, but is implied. NIH says it is not relying on use of government purpose
license to implement its policy. The regulation reads,
The recipient may copyright any work that is subject to copyright and was
developed, or for which ownership was purchased, under an award. The HHS
awarding agency reserves a royalty-free, nonexclusive and irrevocable right to
reproduce, publish, or otherwise use the work for Federal purposes, and to
authorize others to do so (45 CFR 74.36(a)).
The concept of nonexclusive right to use the work is similar to the concept of
“government purpose license” that is used in the Federal Acquisition Regulation,
which governs federally funded contracts. Government purpose licensing permits
agencies to disseminate to the public scientific and technical articles based on, or
containing data produced from, research funded by the agency. The government may
subsequently use and distribute the scientific and technical articles as submitted to
a publisher or as published in a journal if the publisher has not added any original
materials, such as publisher-prepared abstracts or peer review comments. However,
generally an agency should obtain a publisher’s written permission to reuse or
republish the article as published in the journal.144 Use of “government purpose
141 Association of Research Libraries, “Improving Access to Publicly Funded Research.
Policy Issues and Practical Strategies,” Association of Research Libraries Press Release. Oct
20, 2006. Includes links to papers and remarks at [http://www.arl.org/forum06].Jennifer
McLennan, “Higher Education and Library Leaders Voice Support for Free Access to
Federal Research,” Oct. 25, 2006, available [http://www.ar.org/arl/pr/forum06.html].
142 Questions and Answers, op. cit.
143 Federal Register, Section P. Legal Issues.
144 According to the source: “FAR Subpart 27.4 — Rights in Data and Copyrights provides
copyright guidance for the civilian agencies and NASA. In addition, agencies may have
their own FAR Supplements that should be followed.” The authority granted to the
government to use the published version of an article resulting from federally funded
authority” per se to disseminate published journal articles to the public may be
limited to contracts funded by those agencies whose originating or authorizing
legislation mandates them to preserve and/or disseminate information to the general
public about the agencies’ activities and research results.145 Agencies may attach
separate and different interpretations to this function and purpose.
Other agencies that support scientific grants are governed by OMB Circular
A110-section 36, which allows copyrighting by the owner of the work produced from
the award of federal funds, but gives the government a nonexclusive right to use it.
The recipient may copyright any work that is subject to copyright and was
developed, or for which ownership was purchased, under an award. The Federal
awarding agency(ies) reserve a royalty-free, nonexclusive and irrevocable right
to reproduce, publish, or otherwise use the work for Federal purposes, and to146
authorize others to do so.
The Circular A-110 language does not appear to require agencies’ enabling
legislation to mandate dissemination of research findings, although agency
regulations generally require grantees to publish or disseminate the findings of their
research and to share data generated by such research. See, for instance, the NSF
Grant Policy Manual which specifies that “Investigators are expected to promptly
prepare and submit for publication with authorship that accurately reflects the
contributions of all those involved, all significant findings from work conducted
under NSF grants.”147 However agencies may have different rules relating to the
dissemination of research findings and definitions of “Federal purpose.”
If other agencies were to develop Public Access policies like NIH’s, they might
use a policy of voluntarily submitted manuscripts like NIH. But research funding
agencies might also chose to invoke government purpose license or nonexclusive
right to use policies to archive articles.
research support is implied to be applicable to grants also. See section 4, “Works Created
Under a Federal Contact or Grant,” of Frequently Asked Questions About Copyright A
Template for the Promotion of Awareness Among CENDI Agency Staff,
CENDI/2004-8.Updated August 2004, HTML last modified May 04, 2005, Edited and
updated by Bonnie Klein, Defense Technical and Information Service and Gail Hodge,
Information International Associates, Inc., Published by CENDI Secretariat, Information
International Associates, Inc., Oak Ridge, TN, August 2004. CENDI is a federal
interagency committee, the Commerce, Energy, NASA, Defense Information Managers
Group. Available at [http://cendi.dtic.mil/publications/04-8copyright.html].
145 Gary G. Borda, NSA Headquarters, “Government Data Rights Under the FAR,” March
146 “Uniform Administrative Requirements for Grants and Agreements With Institutions of
Higher Education, Hospitals, and Other Non-Profit Organizations,” OMB Circular A-110
(Revised 11/19/93, As Further Amended 9/30/99), Section 36(a)).
147 Section 734, Dissemination and Sharing of Research Results.
Issues Relating to Federal Open Access Archives
In addition to NIH’s Public Access policy and PMC, other federal agencies have
engaged in open access activities. Several federal agencies publish free, open access,
peer reviewed, Internet accessible journals. These journals include Emerging
Infectious Diseases, by the Centers for Disease Control and Prevention; and
Agricultural Research and the Journal of Agricultural Research, maintained by the
U.S. Department of Agriculture and the National Agricultural Library. Others have
free, searchable, electronically available repositories that include abstracts, links to
full-text articles, and other research reports, some of which may be read online.
However, some agencies have confronted serious obstacles to maintaining such
systems and have been forced to terminate them. Below is an overview of agency
activities and a review of some of the general issues raised about federal involvement
in open access publishing and databases.
Federal Scientific and Technical Archival Databases
Some agencies maintain databases or repositories containing citations, articles
or reports that resulted from government-funded research or research funded by other
sources, and some include preprints of scientific and technical materials. For
instance, the DOE Information Bridge allows readers to access for free all available
Department of Energy (DOE) preprint report literature (preprint reports prepared for
the government via grant or contract that are usually longer than articles published
in journals). DOE also has a tool called E-print that allows the user to search major
preprint systems and university sites where articles are posted. E-print is a gateway
to over 17,208 websites and databases worldwide that hold “... e-prints in basic and
applied sciences, primarily in physics but also including subject areas such as
chemistry, biology and life sciences, materials science, nuclear sciences and
engineering, energy research, computer and information technologies, and other
disciplines of interest to DOE.”148 The system permits documents to be “... circulated
electronically to facilitate peer exchange and scientific advancement. Included are
pre-publication drafts of journal articles (preprints), scholarly papers, technical
communications, or similar documents relaying research results among peer
Other federal agency open access systems include:
!The GrayLIT Network,150 which includes the searchable full text of
gray literature from the Defense Technical Information Center, the
DOE, the NASA Jet Propulsion Lab, NASA Langley, and the
Environmental Protection Agency.
148 Available at [http://www.osti.gov/eprint].
149 Available at [http://www.osti.gov/eprint].
150 Available at [http://graylit.osti.gov].
!The Federal Research and Development Project Summaries151
system contains information about research projects from the DOE,
the National Institutes of Health and the National Science
!The U.S. Department of Agriculture’s (USDA) AGRICOLA
(AGRICultural OnLine Access) system, an online bibliographic
data base which provides citations, abstracts, and links, when they
are available, to published and non-published agricultural literature
in the National Agricultural Library.153
!The Astrophysics Data System (ADS) is a National Aeronautics and
Space Administration (NASA)-funded project which maintains four
bibliographic databases containing more than 4.2 million records,
including links to external resources dealing with: Astronomy and
Astrophysics, Instrumentation, Physics and Geophysics, and
preprints in Astronomy. The system also contains full-text scans of
much of the astronomical literature (almost 50 astrophysics154
Objections to Government-Operated Databases: Censorship
and Competition in the Free Market
Allegations of censorship and governmental competition with free market
mechanisms are often cited in opposition to government-maintained databases of
scientific and technical information.
Allegations of Governmental Censorship. Some critics focus on
dissemination issues and contend that governmental operation of archives and
databases of abstracts and journal articles resulting from federally funded research
or research funded by other sources implies government “censorship and
encroachment upon scholarly discourse.”155 Federal officials, rather than private
publishers, some allege could end up determining what research gets archived or
disseminated and what does not.
Curbs on Department of Energy Information Systems. Some
publishers have objected to government-run scientific and technical databases
containing abstracts or articles, saying these threaten their publishing activities and
151 Available at [http://www.osti.gov/fedrnd].
152 Marydee Ojala, “PubSCIENCE Joins the Endangered Special List,” Information Today,
Oct. 1, 2002.
153 Available at [http://agricola.nal.usda.gov/].
154 Available at [http://adswww.harvard.edu/].
155 See, for instance, statement of the Association of American Publishers’ Patricia
Schroeder in Danielle Belopotosky, “Online Federal Library on Health Research Sparks
Outcry,” GovExec.com, Sept. 3, 2004.
employees’ jobs. This controversy is illustrated by the experiences of at least two
The DOE E-print system, described above, has been controversial, and,
according to a DOE official, a few years ago several publishers threatened to prohibit
publication of articles that authors posted on it. But eventually the publishers
relented and now each publisher has different rules regarding the posting of
PubScience, was a U.S. Department of Energy effort to provide a free
multidisciplinary database for physical sciences literature. It contained indexed
abstracts or citations for federally funded and other literature published in
commercial journals. Readers could access indexed abstracts for free, but were
directed to the commercial website link to obtain the full text article, usually for a
fee.157 The system was initiated on October 1, 1999 and closed on November 4,
... the effort quickly became the target of intense lobbying, spearheaded by the
Washington-based Software & Information Industry Association (SIIA), a
coalition of for-profit and nonprofit members including Reed Elsevier, ISI,
Chemical Abstracts Services, and Cambridge Scientific Abstracts. The SIIA
claimed that such a service competed with its members’ services and argued that
government initiative should confine themselves to government information158
DOE’s Office of Scientific and Technical Information (OSTI) operated
PubScience. According to one DOE official, intense lobbying by publishers and their159
associations threatened OSTI’s budget. The House Appropriations committee
report on the DOE FY2002 appropriation bill, H.R. 2311 (H.Rept. 107-112, pp. 108-
156 Interview, DOE official April 2005.
157 According to an article written shortly before the termination of PubScience:
“PubSCIENCE launched in October 1999 with the mission of providing free Web search
capabilities for journal article abstracts and citations in the physical sciences. Reading the
abstract is free, but hyperlinking to the full text generally involves paying for the article. The
collection contains over 1,200 journal titles from 35 publishers, including both professional
associations (American Association for the Advancement of Science, American
Meteorological Society, American Physical Society, American Society for Microbiology,
Royal Society of Chemistry, and the Society for Industrial and Applied Mathematics) and
private publishers (Blackwell Science, Kluwer Academic Publishers, Nature Publishing
Group, Springer-Verlag, and Taylor & Francis Publishers, Ltd.). A few university presses
also contribute to the database. Clearly modeled after PubMed, PubSCIENCE wanted to
attract scientists and the general public to its information. Noting that the U.S. federal
government funds 80 to 90 percent of scientific research and development, DOE touts
PubSCIENCE as a significant taxpayer benefit.” (Source: Ojala, op.cit.).
158 Andrew Albanese, “PubScience Dies Despite Comments,” Library Journal, Dec. 15,
2002. See also: Ojala, op. cit., and “SIIA Releases Comments on DOE’s PubScience
Decision,” Nov. 15, 2002.
159 Interview with OSTI official, April 2005.
asked DOE to keep its efforts focused appropriately. The existence of the
commercial database Scirus160 and another called Infotrieve161 were cited as
competing commercial vendors.162
The Federal Database: PubChem . Efforts were made in 2005 to curtail
or close an NIH database initiated to advance science by assisting basic researchers
to identify chemicals related to genetics and cellular research. According several
articles, the American Chemical Society (ACS) initially sought closure,163 and then164165
modified its position to seek limitations, on PubChem, which, it says, duplicates
ACS’s commercial, fee-based Chemical Abstract Service (CAS).
Reportedly, NIH launched PubChem in fall 2004 to provide data and to index
hyperlinks to articles on the chemical structures of small organic molecules and
information on their biological activities to support the “molecular libraries and166
imaging component of the NIH Roadmap Initiative,”which is a strategic planning
process initiated by the NIH Director.167 PubChem contains data organized into three
databases: PubChem Substance, PubChem Compound, and PubChem BioAssay.
According to NIH,
Links from PubChem’s chemical structure records to other Entrez databases
provide information on biological properties. These include links to PubMed
scientific literature and NCBI’s protein 3D structure resource. Links to
PubChem’s bioassay database present the results of biological screening. Links168
to depositor web sites provide further information.
The system, reportedly, will expand as it includes more data from the Molecular
Libraries centers and data from other online open access chemical database
PubChem, operated by the National Center for Biotechnology Information
(NCBI), also provides readers with free access to links to other NCBI databases. It
is operated by 13 staff members with a budget of about $3 million.
160 Available at [http://www.scirus.com].
161 Available at [http://www4.infotrieve.com/default.asp].
162 Andrea L. Foster, “Energy Department Seeks to Close Web Site That Searches Scientific
Journals,” The Chronicle of Higher Education, Sept. 6, 2002.
163 Jocelyn Kaiser, “Science Resources: Chemists Want NIH to Curtail Database,” Science,
May 6, 2005.
164 Andrew J. Hawkins, “Chemical Society Entreats Congress to Pull Funding For NIH’s
PubChem,” Washington Fax, May 26, 2005.
165 Available at [http://pubchem.ncbi.nlm.nih.gov/].
166 Source: [http://pubchem.ncbi.nlm.nih.gov/].
167 Available at [http://nihroadmap.nih.gov/overview.asp].
168 From [http://pubchem.ncbi.nlm.nih.gov/].
According to the ACS, PubChem jeopardizes its own CAS service, which is
reported to “... employ ... more than 1,200 people in Columbus, Ohio, and makes a
significant contribution to the society’s $317 million in annual revenue from
publications.”169 CAS subscribers receive summary data on chemicals and links to
about 24 million abstracts from about 9,000 journals, as well as patent abstracts on
more than 25 million chemical substances.170 NIH is reported to have said that its
database provides indexes and links only to biological journals that overlap only
slightly with the journals linked by CAS and focuses on “biological information such
as protein structures and toxicology,” which CAS does not deal with, not broader
chemical reactions which CAS covers.171 An NIH official, Christopher Austin, senior
advisor at the NIH Chemical Genomics Center at the National Human Genome
Research Institute, was reported to have said that limitation of PubChem would have
profoundly negative effects on medical discoveries.172 One report said “The overlap
between the two databases occurs in the indexes of chemical names. NIH maintains
the overlap is ‘quite modest’ and for the most part is ‘complementary’ to CAS. ACS
disagrees, saying PubChem duplicates CAS’ platform and replicates its search
features and information.”173 Several articles noted that the ACS lobbied Members
of Congress, especially Appropriations Committee members, to have PubChem
terminated174 or limited to include only compounds derived from federally funded
R&D and to avoid overlap with a commercial enterprise.
Both the House Appropriations Committee and the Senate Appropriations
Committee addressed this issue in their reports on the FY2006 appropriations bill
that includes appropriations for NIH (H.R. 3010). They did not reduce funding for
the database. Both reports said essentially the same thing — that they understood
that the database will include chemical compound information from the NIH-funded
molecular libraries screening center network and from other sources. But they both
expressed concern about duplication of effort with the private sector and urged NIH
to work with private sector publishers to avoid unnecessary duplication.175 After
conference committee action, the bill was cleared on December 21, 2005 for the
President’s signature176 and signed as P.L. 109-149.
169 Kaiser, May 6, 2005, op. cit.
170 Hawkins, May 26, 2005, op.cit..
171 Kaiser, May 6, 2005, op. cit.
172 Aliya Sternstein, “Chemical Publisher Goes After NIH,” FCWCom, May 27, 2005.
173 Hawkins, May 26, 2005, op. cit.
174 “AmChem Soc Calling for Shutting Down Govt. Chem. Database,” email from Patrice
McDermott, American Library Association, May 17, 2005; Hawkins, May 26, 2005, op. cit.
175 U.S. Congress, House, Committee on Appropriations, Departments of Labor, Health and
Human Services, and Education, and Related Agencies Appropriation Bill, 2006, H.Rept.th
109-143, 109 Congress 1st session, p. 112 ,and U.S. Congress, Senate, Committee on
Appropriations, Departments of Labor, Health and Human Services, and Education, andth
Related Agencies Appropriation Bill, 2006, S.Rept. 109-103, 109 Congress, 1st session,
176 See also Jocelyn Kaiser, “House Approves 0.5% Raise for NIH, Comments on Database,”
Reportedly, “Supporters of PubChem see the House language as a victory for
NIH.”177 An ACS official is reported to have said that the language is a “
‘tremendous step in the right direction.’ “178 In late August 2005 NIH rejected an
offer from ACS to create and manage for free “a database for NIH to deposit bioassay
data from its molecular screening project.”179 Instead, on September 1, 2005, NIH
announced in the Federal Register that it was inviting participation from private
sector providers and users of chemical information to participate in a new working
group “to advise on interactions with private sector information providers in the
development of PubChem.”180 Subsequently, it was reported in October 2005 that
the American Chemical Society objected to what it characterized as the retrospective
process that the group was to use to assess biomedical relevance of compounds in the
data base,181 and sought that prospective analysis be used instead. Reportedly, NIH
database managers said that NIH cannot know “...’a priori which compounds should
and shouldn’t go into the collection.’ “ The private-sector panel and NIH officials
met on December 19, 2005, and, reportedly, “No definitive conclusions were reached
at the end of the meeting, although industry representatives said they left with a better
understanding of PubChem and of NIH’s intentions. Agency officials said it was
unclear whether the working group would meet again.” 182
According to NIH, many new private sector depositors have contributed to
NIH’s PubChem system. A “live” list of depositors is available at
[http://pubchem.ncbi.nlm.nih.gov/sources/sources.cgi]. This site report that
MDL/Elsevier deposited over 2 million structures under the “DiscoveryGate”
and “xPharm”, names, for example, and Prous Science deposited several
thousand structures from their “Drugs of the Future” review journal. This shows
that some of the major providers in the chemical information industry have found
PubChem useful and complementary to their commercial services.183
Speculation About Differences in Federal Agency Policies. There are
no unequivocal answers as to why some agencies can maintain open access systems
more easily than others. It may be that publishers, despite their misgivings,
Science, June 17, 2005.
177 Kaiser, op. cit., June 17, 2005.
178 Quoted in Andrew J. Hawkins, “NIH Should Reign In PubChem’s Duplicative Services,
House Appropriators Warn,” Research Policy Alert, June 21, 2005.
179 Andrew J. Hawkins, “NIH’s PubChem Compromise: To Solicit Advice From Industry,”
Research Policy Alert, Sept. 2, 2005. See also: Shirley Haley, “NIH Rejects ACS Offer to
Create a PubChem-Like Database For the Agency,” Research Policy Alert, Aug. 24, 2005
180 “National Library of Medicine; Request for Nominations,” Federal Register, Sept. 1,
181 Andrew J. Hawkins, “NIH’s PubChem Odyssey Continues With American Chemical
Society Counteroffer,” Research Policy Alert, Oct. 18, 2005.
182 Andrew J. Hawkins, “NIH Meets With Chemical Information Vendors To Settle Pub
Chem Dispute,” Research Policy Alert, Dec. 21, 2005.
183 Personal communication with NIH official, Oct. 23, 2006.
moderated their opposition to congressional action to put manuscripts on NIH’s
PubMed Central since the posted items are limited to those that resulted from NIH
funding. However, NIH may be in a different position from other federal agencies
since it has a mandate to preserve and provide health information to the public; other
agencies may not have such clear mandates to distribute information and the results
of their research funding to the public. Furthermore, support for NIH’s open access
activities seems based not only on the need to allow taxpayers access to results of
research their taxes funded,163 but also on the emotional argument about need for
rapid access to information to improve health and save lives, a compelling rationale
to many Members of Congress.
Reportedly, DOE’s Scientific and Technical Information Advisory Board is
discussing, at the highest levels, the question of whether it should establish an open
access policy like NIH’s to make DOE-funded articles available in its own database
and is preparing a report on this subject. According to several federal agency staff,
it seems that in the absence of guidance from the congressional appropriations
committees, agencies, other than NIH, would likely find it difficult to mount a system
like NIH’s because of publisher opposition.164
Scientific publishing and communications methods are slowly changing as
Internet publishing becomes more prevalent. Some observers say that government-
supported researchers and sponsoring agency staff should participate in shaping these
new methods of delivering scientific information. CENDI (the Commerce, Energy,
NASA, Defense Information Managers Group), an interagency committee composed
of senior Scientific and Technical Information (STI) managers from 12 U.S. federal
agencies, has working groups that are studying open access publishing, indexing, and
archiving and has issued reports on it to help develop uniform standards and methods
of international cooperation.165
163 Istook, op. cit.
164 Interview with CENDI official, May 2005.
165 CENDI’s members are: Defense Technical Information Center (Department of Defense);
Office of Research and Development & Office of Environmental Information
(Environmental Protection Agency); Government Printing Office; NASA Scientific and
Technical Information Program; National Agricultural Library (Department of Agriculture);
National Archives and Records Administration; National Library of Education (Department
of Education); National Library of Medicine (Department of Health and Human Services);
National Science Foundation; National Technical Information Service (Department of
Commerce); Office of Scientific and Technical Information (Department of Energy);
USGS/Biological Resources Discipline (Department of the Interior). These programs
represent over 96% of the FY2004 federal research and development budget. Among
CENDI’s open access-related working groups are those that deal with “Archiving,
Preservation, and Permanent Access” and “Content Management and Access.”According
to CENDI, “In 1999, CENDI and the International Council for Scientific and Technical
Information (ICSTI) jointly sponsored a review of the state of the practice of digital
archiving. Over 30 organizations were surveyed and 18 were interviewed to collect
information Regarding technology, policy, procedures, and metadata in operational or
Several international organizations and other countries are examining wider
implementation of open access publishing. Following the release in 2003 of the
“Berlin Declaration” which called for open access to knowledge and its signing by
representatives of selected European universities, research groups, and government
sectors,166 the European Union began a study on changes in markets for scientific and
technical publishing in Europe. Among its topics of inquiry is the subject of “open
access to research findings for all and the need to reconcile authors’ rights and the167
economic interests of publishers.” The report, Study on the Economic and
Technical Evolution of the Scientific Publication Markets in Europe, January 2006,168
endorsed but did not require open access to publicly funded research.
In 2004, the Organization for Economic Cooperation and Development’s
(OECD) science ministers declared their commitment to a “... principle that research169
data from public funding should be openly available” on the rationale that
providing such access promotes long-term economic benefits, more informed
governmental decisionmaking, and hastens the advancement of scientific research.
The ministers asked OECD to develop guidelines to “facilitate optimal cost-effective170
access to digital research data from public funding ...” that would be balanced in
terms of opening access while recognizing “the need for restriction of access in some171
instances to protect social, scientific, and economic interests.”The guidelines will
be released after approval by the OECD Council. The 2004 OECD work was based,172
in part, on a report that was funded by the U.S. National Science Foundation.
prototype projects. The results of this project were reported in Digital Electronic Archiving:
The State of the Art and the State of the Practice, a report to ICSTI and CENDI. An update
of the report was completed in 2004. The updated version, CENDI 2004-3, is available in
PDF.” (Source: [http://www.cendi.gov]).
166 Available at [http://www.zim.mpg.de/openaccess-berlin/signatories.html].
167 “EU Investigates Open Access Scientific Publication,”News - Medical. Net ..., June 15,
168 The 108 page report is available at [http://europa.eu.int/comm/research/science-society/
169 Peter Arzberger, et al., “An International Framework to Promote Access to Data,”
Science, Mar. 19, 2004.
170 “Science, Technology, and Innovation for the 21st Century. Meeting of the OECD
Committee for Scientific and Technological Policy at Ministerial Level, 29-30 January 2004
- Final Communique.” Annex I., Available at [http://www.oecd.org].
171 Annex 1, Available at [http://www.oecd.org/document/15/0,2340,en_2649_201185_
172 “International Access to Research Data Critical to Advancing Science for the Public
Good, Report Says,” NSF Press Release, NSF PR 04-031, Mar. 18, 2004. The report was
not named in the press release. The author was reported to be Peter Arzberger, director of
life sciences initiatives at the University of California, San Diego.
According to the report’s authors, “The ultimate goal ... is to make data sharing and
the principle of open access the rule rather than the exception.”173
Another OECD report published on September 2, 2005, as Digital Broadband
Content: Scientific Publishing.174 It reiterated the view that governments should
increase access to findings from publicly funded research to maximize social returns
on public investments and presented examples and comprehensive pro and con
analyses of currently used business models of open access publishing and open
access archives.175 It also summarized the pros and cons of “hybrid” business models
which distribute publishing costs among authors and users. One example is a
... two-part tariff for author fees...with fees levied for submission and publication
serving to reduce the tendency for multiple and speculative submission of papers
for publication, and enabling journals to cover the costs of quality through
support for higher rejection rates. Such as model might also serve to increase
revenue certainty for publishers of open access author pays variant journals and,
by reducing the cost of publication in them, enable them better to compete for
authors with subscription-based journals. However, user resistance would be a176
strong possibility compared with simpler author pays models. “
The report also proposed variations of another hybrid model involving
“...segmentation of a journal into subscription and open access on an article-by-article
basis, according to the author’s preference and willingness/ability to pay.”177
Apparently a number of publishers have already adopted such practices, and the
OECD report concluded “Such a model may be a useful way for a journal title to
migrate from a subscription model to an open access model over time, with the pace
and direction of change dictated by author preferences.”178
As noted above, there has been considerable governmental and
nongovernmental activity to promote open access publishing in the United Kingdom.
Some scientific and medical researchers in Britain took steps to make research results
freely available via the British open access publisher, BioMedCentral.179
Subsequently, in 2004, the Science and Technology Committee of Britain’s House
of Commons issued a report endorsing open access to research results by proposing
to require authors to deposit their published papers in online archives and journals
using an author pays model and eliminating subscription fees. It also recommended
that government agencies mandate that government-funded researchers put their
173 NSF PR 04-031, op. cit.
174 Organization for Economic Cooperation and Development (OECD), Working Party on
the Information Economy, Digital Broadband Content: Scientific Publishing, Sept. 2, 2005,
175 Digital Broadband Content: Scientific Publishing, op. cit., pp. 57-75.
176 Digital Broadband Content: Scientific Publishing, p. 74.
177 Digital Broadband Content: Scientific Publishing, p. 74.
178 Digital Broadband Content: Scientific Publishing, p. 75
179 John T. Softcheck, “U.K. Publishing Deal Makes Public Research Results Available to
All,” Washington Fax, June 30, 2003.
articles into the archives180 and that the government pay some publishing fees.181 In
November 2004 the U.K. government (the Department of Trade and Industry)
rejected the proposal, maintaining there is no indication that access to scientific
journals is impeded under current publishing methods, and that according to the
government, “the true costs of open-access publishing are still not clear ...”182 and “it
is ‘not obvious ... that the ‘author pays’ business model would give better value for
money than the current one’ ....”183 In June 2005, the United Kingdom Research
Councils (RCUK),184 the main British supporter of publicly funded research, “which
distribute[s] most government science funding,”185 issued for comment a draft policy
which mandates researchers it funds to archive their journal articles and conference
papers “in a free public archive ‘at the earliest opportunity, wherever possible at or
around the time of publication.’ “186 But the rules may allow publishers to continue
to embargo archiving articles until many months after publication, since the council
says “its mandate is ‘subject to copyright and licensing arrangements’ that can restrict
what authors do.”187 Costs of publishing in “author pays” journals would be covered
by the Research Councils’ funding grant “subject to justification of cost-
effectiveness.”188 The British government said it would review its policy options on
this issue taking into consideration the draft RCUK policy and any changes to it, as
well as other information.189 The executive board of the RCUK issued a policy
statement in the summer of 2006, saying that “...all peer-reviewed journal papers
produced by publicly funded research must be made available for free soon after
180 Lila Guterman, “British Parliamentary Panel Endorses Open Access to Scientific
Literature,” Chronicle of Higher Education, July 20, 2004.
181 Daniel Clery, “Scientific Publishing: Mixed Week for Open Access in the U.K.,” Science,
Nov. 12, 2004.
182 Clery, op.cit.
183 Clery, op. cit.
184 The Research Councils UK have been identified as “... a partnership of the U.K.’s eight
research councils. Funded by the government’s Office of Science and Technology, the
councils are independent public bodies that account for the vast majority of publicly funded
research in the U.K., including medical research as well as research in the humanities, social
sciences, and physical sciences” (Janet Coleman, “Costs, Benefits of U.K. Open Access
Policy should be Studies Before Funders Require, Royal Society Urges,” Research Policy
Alert, Nov. 28, 2005).
185 Jim Giles, “UK Research Councils Claim Success for Open-access Publishing Plan,”
Nature, June 2, 2005.
186 Eliot Marshall, “Scientific Publishing: Britain’s Research Agencies Endorse Public
Access,” Science, July 8, 2005. For earlier history see Giles, op. cit.
187 Marshall, July 8, 2005, op. cit.
188 “RCUK Announces Proposed Position on Access to Research Outputs,” News release
189 Email communication from a staff member of the Office of Science and Technology, a
British Government official, July 21, 2005, who said “The government position will be
reviewed in the light of advice from RCUK, results of studies by JISC and the report from
the EU study.”
they’re completed.”190 But “exactly what that means was not specified, and RCUK
left each research council to set its own rules.”191 RCUK also said it would assess the
results of a two-year analysis of the impact of mandating open access and review the
policies in 2008.192 The Wellcome Trust, a large British medical foundation, recently
announced that it requires all papers produced with its support “... to be submitted to
the NIH archive or to the British equivalent that is being developed.”193
The British Royal Society, an advisory body to the government, which also
publishes seven peer-reviewed journals, whose papers can be accessed without
charge a year after publication, issued a position paper opposing the RCUK policy.
It cited, in particular, the lack of assessment about cost effectiveness of institutional
archives, subject-based repositories, and self-archiving; the potential for the proposed
policy to threaten survival of some existing journals; and the problems observed with
quality control of articles appearing in some open access publications.194 Apparently
some learned societies fear that libraries will cancel subscriptions to their
professional societies’ publications.195
Other foreign and international organization open access activities are
summarized in Julie M. Esanau and Paul F. Uhlir, eds., Open Access and the Public
Domain in Digital Data an Information for Science, Proceedings of an International
Symposium, Published by U.S. National Committee for CODATA, National
Academies Press, Washington, D.C., 2004.
Summary of Policy Issues and Questions
Policies for open access journals and citation repositories are evolving andth
contentious issues may be raised during the 109 Congress. Those that have
implications for academic institutions are discussed in Appendix l. Other policy
issues and questions are emerging, including the following.
!Assessment of which federal agencies, in addition to NIH, would
seek to archive and provide free public access to manuscripts or
articles reporting the results of research that they supported.
190 Eliot Marshall, “A Mixed Bag of U.K. Open-Access Plans,” Science, July 7, 2006, pp.
191 Marshall, July 7, 2006, op. cit.
192 Marshall, July 7, 2006, op. cit.
193 Giles, June 2, 2005, op. cit.
194 Janet Coleman, Nov. 22, 2005, op. cit. The cited report is The Royal Society, “Royal
Society Response to Research Councils UK’s Consultation on Access to Research Outputs,”
Policy document 15/05.
195 Marshall, July 8, 2005 and Giles, June 2, 2005, op. cit.
! Analysis of which agencies might seek to provide access to
manuscripts or articles, using government purpose license or
nonexclusive right to use published articles, regardless of copyright
!Comparison of the quality of peer review processes and of peer
reviewed articles that are published in traditional, subscriber-pays
and open access journals.
!Monitoring of whether academic reward systems react differently to
articles published by traditional publishers or open access publishers
and assessing the implications for professional advancement of
researchers and teachers in academic promotion and tenure systems.
!Assessing the positive and negative impacts on the speed and quality
of scientific research, knowledge synthesis, and knowledge
accumulation flowing from open access publishing and open access
citation/abstract archives in comparison with traditional publishing
and archival methods.
!Analysis of publisher actions to identify whether or not authors who
seek copyright agreement terms allowing them to post manuscripts
in PMC are penalized.
Monitoring of NIH Public Access Activities and Other Federal
Initiatives, Including PubChem
!Assessment of rates of voluntary participation by NIH-funded
authors in the Public Access policy and determination of whether
there are any negative impacts — from research sponsors or the
scientific community — on NIH-funded authors who may not
submit articles for dissemination in PMC.
!Determination if federal open access databases and archival
repositories should be limited to providing access only to
publications that result from federally funded R&D.
!Assessment of proposals for governmental citation archives to link
to publisher’s websites to read published articles, as opposed to
posting articles on a free access government system.
!Follow-up to congressional mandates that NIH monitor the
implementation of its Public Access policy, that it work with
traditional, subscriber-pays publishers to monitor the impacts and
costs of open access archiving of text on PMC, as it posts what is
estimated to be thousands (possibly 60,000) of additional articles on
the system, and that it work with publishers to monitor impacts on
the integrity of peer review processes. (The NIH Director estimated
that the added costs for posting all NIH-funded research studies on
PubMed Central’s digital library at around $2 to $4 million
annually.196 According to NIH, agency-supported research resulted
in 60,000 to 65,000 published papers in 2003.)197
!With respect to PubChem, assessing cooperation between NIH and
private groups on clarifying the possible overlap between NIH’s
archive and that of private activities, including the American
Chemical Society’s Chemical Abstracts Service. Analysis of the
impacts on biomedical research in general and on NIH’s research
and its strategically planned genomic research initiatives if the scope
of PubChem were to be limited.
!Determining whether federal regulations for support of contracts and
grants will continue to allow agencies to pay individual authors or
academic institutions for the costs of publishing articles in open
access journals as part of the research process, especially if open
access publishing becomes more widespread and a substantial
portion of the scientific and technical publishing market. A related
issue is determining the possible effects on research support funding.
!Given that federal research sponsors allow some journal publishing
and subscription costs to be counted as part of the costs to conduct
federally sponsored research, comparing the actual total costs to the
government for publishing and reading of scientific articles
published traditionally as opposed to those published using open
!Analysis of the role that the federal government should play in
funding the start-up of nongovernmental citation archives and
repositories for scientific and technical articles, if the government
also initiates governmental activities with similar purposes.
196 Elias Zerhouni, “NIH Public Access Policy,” Science, Dec. 10, 2004. See also: Janet
Coleman, “Open Access Would Cost NIH Roughly $2.5 Million, Agency’s Lipman
Estimates,” WashingtonFax, Sept. 24, 2004. These costs have been criticized as excessive.
See, for instance Michael Stebbins, et. al., “Public Access Failure at Pub Med,” Letter,
Science, July 7, 2006, p. 43. Others say this amount is a small portion, that is about 0.011%
of NIH’s current annual appropriation, which is almost $28 billion annually and a small
amount compared to the approximately $30 million that NIH says it awards annually to its
funded researchers for publication page charges and so forth (Michael A. Rogawski and
Peter Suber, “Support for the NIH Public Access Policy,” Science, Sept. 15, 2006, p. 1572).
197 Questions and Answers, op. cit.
!Economic analysis of the impacts on the commercial publishing
industry (revenues, employment, sustainability, etc.) if open access
publishing and archiving activities continue to expand.
!Examination of the extent to which professional scientific societies
utilize the profits from publishing to support their activities and of
alternative sources of funding for these activities.
Appendix 1. Open Access Publishing: Selected
Questions in Academia
Continuing questions relating to controversial issues about open access
publishing were raised by Andy Gass and Helen Doyle, “The Reality of Open-Access
Journal Articles,” Chronicle of Higher Education, February 18, 2005. They conclude
that although there are problems, support is growing in academia for open access
journals. Remaining questions include:
!What will become of the market for secondary filters of primary
research articles, services like BioMed Central’s Faculty of 1000,
which highlight important papers published in a wide swath of
journals? Will fee-for-access ventures that collect open-access
articles become a new cash cow for publishers? At present, faculty
members offer their recommendations to the filtering services free,
and publishers sell their aggregated opinions to institutions — will
established professors go on contributing their free labor to such
!How will the role of the research library change, as open-access
scholarly communication becomes more widely practiced? To what
extent will librarians be freed from the burdens of subscription
!Many university libraries now encourage open access by subsidizing
a portion of the publication charges in open-access venues for
authors affiliated with the university, through channels like our
employer’s institutional membership program. Will those subsidies
continue? If so, will they continue to be paid from libraries’ budgets,
or will they come from research budgets — a source that would be
more consistent with the view of open-access proponents that costs
of publication should be part of the costs of conducting research? Or
will external granting agencies, many of which already pay
scientists’ page charges and color-illustration fees, assume the full
costs of their investigators’ open-access publications?
!Will libraries continue to serve as intermediaries through which
researchers find open-access information, as well as that available
only through subscription, and how?
!Those questions relate not just to academic libraries, but to the
mission of colleges and universities. The time has come for a
comprehensive review of how best to pay for the dissemination of
!How will reduced legal barriers to reusing articles — a stipulation
of most formal definitions of open access — affect teaching,
research, and other scholarly activities? There are, of course, good
precedents for having few or no legal restrictions on the reuse of
scholarly work: Every article published by an employee of the NIH
is in the public domain. Some more-restrictive open-access licenses
now available, like the Creative Commons attribution license in use
for articles from our employer and from BioMed Central, permit
users to reproduce scholarly work in any medium, for any purpose,
as long as the author receives proper credit.
!What kinds of educational tools will such licenses make possible?
For example, will we see a proliferation of online articles enhanced
with explanatory links and informational sidebars, which make
scientific discoveries more comprehensible to a wide audience? Will
such resources be produced by commercial enterprises? By nonprofit
organizations? Or by networks of volunteers, as is the case with
open-source computer software?
!Will open-access articles enable more researchers from
less-developed countries to work on the frontiers of science? Given
that all credible open-access journals waive publication fees for
authors who can’t afford to pay them, increased availability — and
therefore knowledge — of the literature might well allow scientists
in the developing world to increase their output of cutting-edge
work. Would that change, in turn, help resolve the “10/90 gap” —
the unfortunate reality that less than 10 percent of the global
expenditure on medical research goes to study the predominant
health needs of 90 percent of the world’s population?
!Most important, what kinds of discoveries might result from
searchable, open archives of peer-reviewed, full-text scientific
literature? The aggregation of gene sequences in a single, freely
accessible information space (GenBank) has spawned entire fields
of research; will open access to journal articles have a similar effect
on areas of work that could benefit from “mining” full texts and
figures? Clearly, comprehensive collections of open-access literature
would make it much easier to systematically review published
!Will open-access literature lead to frequent discoveries of
correlations between phenomena previously thought to be unrelated?
Will it spark more open access to data sets and databases of
laboriously compiled and annotated information? The potential for
open access to lead to new discoveries is its single most compelling
asset, though one that is frequently overlooked.198
198 Andy Gass and Helen Doyle, “The Reality of Open-Access Journal Articles,” Chronicle
of Higher Education, Feb. 18, 2005, p. B13.