|
 |
|
EMBO reports 5, 3, 222–225 (2004)
doi:10.1038/sj.embor.7400107
Access for all?
While initiatives for self-archiving and creating new open access
journals gain momentum, new questions about the legal and economic basis of
scientific publishing arise
Les Grivell
|
|
|
|
Les Grivell is manager of
E-BioSci, the European electronic publishing initiative
e-mail: les.grivell@embo.org
|
|
|
|
Last December, a UN-sponsored world summit on the information society
(http://www.itu.int/wsis) and the announcement that the UK House of Commons
Select Committee on Science & Technology intends to conduct an inquiry into
access to scientific publications were two more developments in a movement that
many regard as a crusade to free up the scientific literature. The basic
aim—to provide toll-free access to the full text of published scientific
articles for anyone wishing to read it—is clearly a noble one to which
few would object. Nevertheless, in recent years the complexity of the issues
and the difficulties involved in bringing about change in scientific publishing
have become more apparent. For those who have not managed to keep up with the
current complicated state of affairs, the following overview may provide food
for thought.
The scientific world can be roughly divided into two groups: those with
access to the scientific literature and those without. Members of the former
group are often blissfully unaware of their privileged status. Their
institutions have paid the required subscription fees, and for them the full
text of journals is available at the click of a mouse. Those without access
face the fact that virtually all recently published research is locked away
behind toll barriers.
The movement to change this state of affairs arose out of what
librarians call the 'serials crisis': the inability of institutional budgets to
keep pace with the spiralling costs of scientific journals. The proponents of
different solutions to the problem of access jostle shoulder to shoulder, each
advocate aspiring to evangelical heights in an effort to promote their point of
view. Broadly speaking, three mutually non-exclusive models specifically
address the issues of access. These are to start up new open access journals,
to encourage authors to provide free access to self-maintained archives of
previously published papers, and to open up access to conventional,
subscription-based journals.
Note that in the previous paragraph I have used two terms, 'free' and
'open' access. This is not sloppiness on my part. These terms are in fact
different and it is the nature of this distinction that is the subject of
heated debate, not least because the terms have in the past often been used
interchangeably, although they do have different connotations.
According to the 2002 Budapest Open Access Declaration
(http://www.soros.org/openaccess), "open access" is defined as
"free availability on the public internet, permitting any users to read,
download, copy, distribute, print, search, or link to the full texts of these
articles, crawl them for indexing, pass them as data to software, or use them
for any other lawful purpose, without financial, legal, or technical barriers
other than those inseparable from gaining access to the internet itself."
The declaration goes on to define the rights of authors: "The only
constraint on reproduction and distribution, and the only role for copyright in
this domain, should be to give authors control over the integrity of their work
and the right to be properly acknowledged and cited."
In contrast, the 2003 Bethesda Declaration on Open Access defines open
access more widely. Works have to meet two conditions: "The author(s) and
copyright holder(s) grant(s) to all users a free, irrevocable, worldwide,
perpetual right of access to, and a license to copy, use, distribute, transmit
and display the work publicly and to make and distribute derivative works, in
any digital medium for any responsible purpose [...] A complete version of the
work and all supplemental materials, including a copy of the permission as
stated above, in a suitable standard electronic format is deposited immediately
upon initial publication in at least one online repository that is supported by
an academic institution, scholarly society, government agency, or other
well-established organization [...]."
|
|
|
|
The proponents of different solutions to the problem
of access jostle shoulder to shoulder, each advocate aspiring to evangelical
heights in an effort to promote their point of view
|
|
|
|
In other words, under the Budapest concept of open access, an author can
choose to make his work freely available, but retains the right to restrict the
ability of others to copy or redistribute it as a whole, or in part. Under the
Bethesda concept, the author waives his or her rights under copyright law.
Publication in a Bethesda declaration-compliant journal is thus the only option
for achieving full open access. This difference has fuelled recent vigorous
mailing-list exchanges between Stevan Harnad from the University of
Southampton, UK, and Michael Eisen from the University of California at
Berkeley (Berkeley, CA, USA), as supporters of the Budapest and Bethesda
definitions, respectively. Eisen redefines the Budapest concept as one that
allows toll-free, but not open access. According to Eisen, in addition to
publisher-implemented toll barriers, the lack of the right to reformat and/ or
redistribute the text is an unnecessary compromise of the interests of the
scientific community in situations in which full texts are used for
computational or data-mining analysis. According to Harnad, indexing and other
forms of text analysis can be performed under the Budapest declaration without
asking authors to sign away their rights. In his opinion, this emphasis on new
open access journals and copyright issues places unnecessary limits on the
speed with which information in the literature can be made freely
accessible.
Copyright restrictions, or rather their abolition, are also the central
theme of a bill introduced in the US House of Representatives in June 2003 by
Democratic congressman Martin Sabo from Minnesota (Sabo,
2003). The Public Access to Science Act aims to amend existing US
copyright law to ensure that research "substantially funded" by the
US government cannot be copyrighted and thus remains freely available to the
public. Whether it will achieve this aim is hotly disputed.
Critics of the proposal have pointed out that the bill in fact does
little to facilitate open access and many have also questioned the wisdom of
removing copyright protection. Notable opponents include the Federation of
American Societies for Experimental Biology (FASEB), the Association of
American Universities (AAU) and many scientific society and commercial
publishers. FASEB argues that the bill "threatens to destroy the current
field of scientific publishing and will harm scientific societies that rely on
publishing revenues to support other professional activities" (Wells, 2003). The AAU is concerned about long-term access
to scholarly research, yet in an open letter to congressman Sabo, AAU president
Nils Hasselmo stressed that copyright protection is an important means of
ensuring the accuracy and authenticity of publications and that its removal is
unnecessary for promoting public access (Hasselmo,
2003). In addition, removal of protection under the bill would
adversely affect other copyrighted works including computer software, and thus
reduce incentives for universities and industry to collaborate in technology
transfer. In a vigorously worded editorial, Michael Held, Director of
Rockefeller University Press, warned that lack of copyright protection would
lead to pirating and uncontrolled reprinting of published materials, thereby
strongly reducing the incentive of the original publisher to invest resources
in their production (Held, 2003).
In an analysis of the Sabo Bill and the initial negative reactions to
it, Samuel E. Trosow, Assistant Professor of Law at the University of Western
Ontario in London, Canada, argued that some of these reactions are based on a
misunderstanding of the purpose and scope of US copyright law and patentability
under US Patent law (Trosow, 2003). Singling out
the AAU for criticism, he remarked that "one is left with the distinct
impression that the large research universities appear to view the Sabo Bill as
a threat to their overall programme of commercialization, and that this
consideration drives their opposition more so than any supposed deleterious
affect the bill would have on the production and dissemination of works derived
from STM [scientific, technical and medical] research."
Amid the hubbub of protests and actions from organizations likely to be
affected by the Sabo Bill, one major player seems to have largely been ignored:
the research scientist. In the role of reader, a scientist may vociferously
demand free or open access. As authors, however, most scientists are concerned
that their papers should be published in good journals and that they should be
subsequently read and cited frequently. Although it has been claimed that
publication in open access journals increases the chances of citation (Lawrence S, 2003), few open access journals in the life
sciences have yet made it into the elite category. Younger scientists in
particular are cautious about putting their careers at risk by submitting work
to what they regard as unknown or sub-optimal journals. Even fewer may be
willing to risk their careers if they either lose control or fear loss of
control over their works through legislation that removes copyright
protection.
These fears are not without foundation. Under Canadian and European
copyright, an author can transfer distribution rights to a publisher as part of
the process of publication, but still retain his or her legally enforceable
moral rights to attribution, integrity and association. Attribution allows an
author to be associated by name with the published work. Integrity protects
against "distortion, mutilation, or other modification".
Association allows an author to control his or her work in association with a
product, service, cause or institution. All three rights provide valuable
safeguards in a situation where, for example, a company might seek to promote a
product by selectively quoting from a published paper. Curiously, the moral
rights of authors do not exist under US copyright legislation (Trosow, 2003). Thus the Sabo Bill cannot erode them
further. I leave it to readers to draw their own conclusions as to whether they
consider this good or bad news.
Academic as the above considerations may appear to someone who, as yet,
has no access at all to the scientific literature, they do have important
effects for both the speed and the mechanism by which access is achieved.
Ignoring for a moment the fact that many profit and not-for-profit publishers
make publications older than 12 months available free of charge, one notable
feature of 2003 was a series of declarations affirming the intention of their
signatories to support the establishment of new open access journals that meet
the two conditions laid down by the Bethesda declaration. The USA-based Public
Library of Science (http://www.plos.org) launched its high profile open
access Biology journal and considerable publicity was devoted to
promoting the view that this form of open access publication should be the way
forward and would give scientists access to the literature, and freedom to
create and distribute derivative works and to carry out sophisticated
computational analysis or data-mining on reprocessed texts.
Others, such as Harnad, are more cautious. While encouraging publishers
of both new open access and conventional scientific journals to make their
content freely available, Harnad is a long-standing proponent of
self-archiving. In this scheme, individual scientists or their institutions set
up web servers on which copies of previously published works are stored and
made accessible by the use of protocols that allow search engines to harvest
relevant bibliographic information or 'metadata'. He points out that open
access journals currently represent only about 5% of the estimated 24,000
journals responsible for publishing a total of 2.5 million scientific articles
per year. In his opinion, the fact that many publishers of conventional
journals already permit self-archiving by authors means that free access to a
very significant proportion of the world's scientific literature could in
theory be achieved virtually overnight. Open access journals, in contrast,
depend for their existence on largely untested models of economic operation,
and, as for any other journal, take time to establish themselves and build up a
scientific reputation.
|
|
|
|
In the role of reader, a scientist may vociferously
demand free or open access. As authors, however, most scientists are concerned
that their papers should be published in good journals and that they should be
subsequently read and cited frequently
|
|
|
|
So, how does self-archiving work? Is it an effective mechanism to
provide free access? If so, why has it not yet become established? Answering
these questions in turn, the first step in self-archiving is to set up a web
server with software that allows data to be stored, searched and retrieved.
EPrints (http://www.eprints.org) and DSpace (http://www.dspace.org) are
perhaps the best-known self-archiving software packages. They are open source,
freely available and comply with the Open Archive Initiative (OAI). This means
that the bibliographic metadata of the stored articles can in principle be
harvested and shared among different self-archive servers, thereby allowing
these to form interoperable nodes in a Napster-like network. Setting up and
maintaining a server is a relatively simple task. Self-archiving is thus an
effective and low-cost mechanism to provide free access. OAI-compliant search
engines, such as the University of Michigan's OAIster, which covers
approximately 2.4 million records from about 250 institutions, provide the
means for finding and retrieving documents.
So why has a large part of the scientific literature not become freely
accessible overnight? As discussed below, there may be difficulties with some
technical issues but the main problems seem to be social and psychological
factors. A scan through the comments of, for example, the American Scientist
Open Access Forum, suggests widespread ignorance of the potential of the
approach and of the right of authors to self-archive; confusion between
preprint and publication servers; lack of motivation on the part of individuals
and/or sluggishness on the part of their institutions to adopt and finance a
self-archiving policy.
The technical issues relate to searchability. More than in other
literature retrieval systems, the quality of the metadata that characterize
individual articles in a distributed self-archiving network is crucial to the
success of cross-searching efforts. Unfortunately, institutional archives often
need to cover a broad spectrum of scientific disciplines, so that the
classification of articles is often limited to major subject classifiers only.
Furthermore, the simple OAI-compliant meta-data models
(http://www.openarchives.org/OAI/openarchivesprotocol.html) used by most
self-archiving software may introduce additional limitations on search
flexibility. Metadata is often incomplete, and archive curators have to add to
limited metadata manually. Clearly, only when a researcher is able to search
rapidly and accurately through full-text articles will self-archiving become a
more attractive proposition. Information platforms, such as E-BioSci
(http://www.e-biosci.org), are likely to have an important role in
realizing the full potential of self-archive repositories.
Given the primary aim of providing toll-free access to as much of the
scientific literature as possible and three non-exclusive ways of achieving
this, what are the likely consequences of each for the longer-term future of
publishing? Let us examine each of the routes separately.
At first sight, self-archiving seems to be the least disruptive of the
three routes and the most friendly to the world of conventional publishing.
However, it takes little imagination to see that in the longer term a
significant degree of self-archiving, coupled with efficient systems for
search, retrieval and caching could have a serious impact on demand for
subscription-based content, with a consequent decline in incentives for
publishers to invest in these activities.
Although costs vary widely from journal to journal, there is no doubt
that publishing is an expensive business in general, and open access journals
are no different from other journals. A new journal can take between three and
five years to establish itself and requires considerable capital investment
during this period. The staffing costs associated with peer review and editing
accepted manuscripts will depend on the volume of manuscripts and acceptance
rates, but are significant. Online publications, although clearly able to reach
more readers at a lower cost than paper-based journals, still need continued
investment in technology and hardware. BioMed Central, a pioneering publisher
in this field, with more than 100 journals under its wing, charges authors
US$525 for processing each manuscript. The Public Library of Science's
elite Biology journal charges US$1,500 per manuscript for
processing. However, other publishers estimate that as about 90% of manuscripts
are rejected to maintain quality, the real costs for this and similar journals
may be anything up to an order of magnitude higher (Butler,
2003).
|
|
|
|
Publication is part of the research process, costs
are associated with any publication system, open access or not...
|
|
|
|
Nevertheless several major funding institutions, including the Howard
Hughes Medical Institute and the National Institutes of Health in the USA, the
Deutsche Forschungsgemeinschaft and the Wellcome Trust in Europe, have so far
been supportive of the principles of free or open access. They have pledged
funds to meet the costs of institutional subscriptions to the journals or
processing charges for individual manuscripts. How this will work out in the
long term remains to be seen. Open access journals have not been in existence
long enough for proper testing of this economic model. However, there could be
painful surprises for laboratories with high publication levels, which could
easily run up bills of many tens of thousands of Euros in this way.
Not surprisingly, few conventional publishers have so far rushed to
convert existing subscription-based, access-limited journals to full open
access. Some publishers— such as the Company of Biologists, the American
Physiological Society and Oxford University Press—have established
curious, hybrid models in which subscriptions are maintained, but authors pay
for open access to their own articles. Nevertheless, the British Medical
Journal's early bold experiment in offering free access to all content has
not been widely taken up. Worse, falling revenues from subscriptions have
forced them to return to paid access. After 10 years of free access, a large
part of BMJ's content will, from 2005 on, disappear behind access controls for
"a year or more (after publication)" (Delamothe
& Smith, 2003). Even though only 12% of the journal's income is
generated from subscriptions with the rest coming from advertising, a 9%
reduction in subscriptions in the past year coupled with other losses has led
the BMJ Publishing Group Board to take this step, albeit reluctantly, to
"defray costs of the journal's web-site and to allow funding of further
developments."
The BMJ's difficulties should serve as a serious warning to those who
have been calling for the abolition of the subscription-based system at any
cost and with it much of the value added to scientific publications by
publishers, whether for profit or not. Some proponents of the free or open
access revolution seem to accept calmly the 'downsizing' or even termination of
current journal publishing activities as inevitable but this would have
dramatic consequences with the permanence, integrity, accuracy and independence
of the scientific record at stake.
Although commercial publishers have been the subjects of the most
virulent attacks so far, it is the learned societies perhaps who stand to lose
most—the ability to support the very communities that founded them. This
paradox is simplistically brushed aside in the FAQ of the self-archiving
community: "Learned Societies are potential allies in and beneficiaries
of the self-archiving initiative. First, they are us. Whatever is good
for research, and for research impact, is therefore good for Learned Societies.
[...] But many of them are also journal publishers, and hence may be facing
downsizing pains. Unlike commercial publishers, however, their first and last
allegiance will of course be to research and researchers, that is, us. We will
hear rationalizations about needing the toll revenues to fund 'good works' such
as meetings, scholarships and lobbying. But it will quickly become evident that
some of these good works are not actually essential, and certainly nothing for
which we would want to sacrifice research impact; and the subset of them that
really is essential (such as meetings) will prove to be able to fund itself in
other ways, rather than needing to be subsidized at the expense of research
impact" (http://www.eprints.org/self-faq).
In summary, the movement to achieve free and/or open access to the
literature has put the world of scientific publishing into a state of flux that
is leading to a re-examination of all aspects of the publishing process and the
roles of publishers and scientists as producers, reviewers and consumers of the
literature. As discussed above, the debate has gone beyond the simple question
of access to the basic economics of publishing. Simple proposals for
redistributing costs have raised complex questions about the ownership,
authenticity and integrity of the scientific record itself.
At the same time, the increasing pressure from research institutions and
funding bodies on authors to publish in new low- or no-cost journals has
triggered heated discussion on the quality of both current and future
publications, the value of peer review and different publication strategies
(Lawrence PA, 2003), and about how scientists in
their role as reviewers and advisors evaluate publications as a major component
of the reward system in science (Colquhoun,
2003).
To the lay public, the complexity of the debate must be confusing in the
extreme. The Public Library of Science's publicity campaigns in the USA have
targeted the taxpayer. They point out that "the current closed system of
publication places the narrow interests of publishers before the public
interest and greatly diminishes the value of the more than US$50 billion
invested by US taxpayers each year in scientific and medical research."
The implication that taxpayers pay twice for something that in the end they
have no access to is clearly calculated to make anyone hot under the collar.
But the full facts are much more complicated and difficult to grasp and are
unlikely to lead to a change in public opinion. Publication is part of the
research process, costs are associated with any publication system, open access
or not, and the 'author-pays' open access models will simply lead to the
reallocation of library funds to research budgets. This state of affairs is
unlikely to change, as Margaret Reich, Director of Publications for the
American Physiological Society, commented humorously: "[...] why should
any of us, scientist and patient alike, have to pay again to read the results
of that [federally funded] research? That sounds good, but some of my tax
dollars also go to wheat and other farm subsidies, and I don't see anyone
handing me free loaves of Wonder Bread™" (Reich,
2003).
In the long term, no doubt access will be available for all, but at a
price. Good (business) sense will hopefully prevail and hopefully too, the
long-term benefits will outweigh the costs for all involved. Although likely to
be just as ruthless in its progress, it probably will be evolution, rather than
revolution (Owens, 2003) that will determine the
future course of events. Changes in attitudes and economic models will lead to
the creation of new niches from which both society and commercial publishers
will be able to continue to serve the scientific community.
|
|
References
Butler D ( 2003) Scientific publishing: who will pay for open access? Nature 425: 554–555 | Article | PubMed | ChemPort |
Colquhoun D( 2003) Challenging the tyranny of impact factors. Nature 423: 479 | Article | PubMed | ChemPort |
Delamothe T, Smith R ( 2003) Paying for bmj.com. BMJ 327: 241–242 | Article |
Hasselmo N ( 2003) Letter from the AAU to M Sabo. http://www.aau.edu/intellect/sabo7.18.03.pdf
Held MJ ( 2003) Proposed legislation supports an untested publishing model. J Cell Biol 162: 171–172 | Article | PubMed | ChemPort |
Lawrence PA ( 2003) The politics of publication. Nature 422: 259–261 | Article | PubMed | ChemPort |
Lawrence S ( 2003) Free online availability substantially increases a paper's impact. http://www.nature.com/nature/debates/e-access/articles/lawrence.html
Owens SR ( 2003) Revolution or evolution? EMBO Rep 4: 741–743 | Article | PubMed | ChemPort |
Reich M ( 2003) Peace, love and PLoS. Physiologist 46: 137
Sabo M ( 2003) Public access to science act. H.R. 2613, US House of Representatives, 108th Congress, 1st session.
Trosow SE ( 2003) Copyright protection for federally funded research: necessary incentive or double subsidy? http://publish.uwo.ca/~strosow/sabo_bill_paper.pdf
Wells RD ( 2003) Letter from the Federation of American Societies for Experimental Biology to M Sabo. http://www.faseb.org/opa/news/docs/sabo.pdf
|
|
 |
|
Top of page MORE ARTICLES LIKE THIS These links to content published by NPG are automatically generated | |
top  |
|
|
|
