Stevan Harnad
School of Electronics and Computer Science
University of Southampton
Highfield, Southampton
SO17 1BJ
Abstract:
What the research community needs, urgently, is free online access
(Open Access, OA) to its own peer-reviewed research output. Researchers
can provide that in two ways: by publishing their articles in OA
journals (Gold OA) or by continuing to publish in non-OA journals and
self-archiving their final peer-reviewed drafts in their own OA
Institutional Repositories (Green OA). OA self-archiving, once it is
mandated by research institutions and funders, can reliably generate
100% Green OA. Gold OA requires journals to convert to OA publishing
(which is not in the hands of the research community) and it also
requires the funds to cover the Gold OA publication costs. With 100%
Green OA, the research community's access and impact problems are
already solved. If and when 100% Green OA should cause significant
cancellation pressure (no one knows whether or when that will happen,
because OA Green grows anarchically, article by article, not journal by
journal) then the cancellation pressure will cause cost-cutting,
downsizing and eventually a leveraged transition to OA (Gold)
publishing on the part of journals. As subscription revenues shrink,
institutional windfall savings from cancellations grow. If and when
journal subscriptions become unsustainable, per-article publishing
costs will be low enough, and institutional savings will be high enough
to cover them, because publishing will have downsized to just
peer-review service provision alone, offloading text-generation onto
authors and access-provision and archiving onto the global network of
OA Institutional Repositories. Green OA will have leveraged a
transition to Gold OA.
Open Access
(OA) is: immediate,
permanent, toll-free online access to the full-texts of peer-reviewed
research journal articles. A great deal of the research
community's attention to OA today is focussed exclusively on the
prospect of a direct transition from Toll Access (TA, "Gray") journal
publishing to OA journal publishing
(the "Gold" road to OA). This golden
road is the more radical of the two
roads to OA, and hence the slower
and more uncertain one. I would like to suggest that the research
community has already been waiting passively for far too long.
Cumulative
research impact keeps being lost daily, weekly, yearly,
because of access-denial to would-be users whose universities cannot
afford the access-tolls. OA is already within every author's
immediate
reach, via the "Green" road to OA.
This is why we should be focussing far more of our attention and
effort on the far less radical transition
from
a "Gray" to a "Green" TA journal policy, whereby the journal stops
short of
converting to gold, but gives its "green light" to authors who
wish to provide OA for their own articles by self-archiving them in
their own university's
online "eprint" archives. For a publisher who is currently making
ends meet, this is a far less risky step than a
direct conversion to the OA (author-end)
cost-recovery model. Hence it is
a step that publishers are far less reluctant to take in order to
demonstrate their support for the research community's mounting desire
for OA. Such a step is far less likely to be opposed -- or be held at
arm's
length by delays and embargoes of the kind Shulenburger
(1998) has proposed. ("Shulenburger on open access: so NEAR and yet
so
far"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/3277.html
).
The Green option allows the number of OA articles (not journals) to
grow
anarchically,
article by article, rather than systematically,
journal by journal. This allows TA journals
to adjust gradually to any changes that might arise as the number of
self-archived OA articles grows.
There is, for example, far less risk of library cancellation for any
particular TA journal when OA is not growing systematically, journal by
journal, but anarchically, article
by article: The libraries too will
only
learn gradually whether it ever becomes safe to cancel any particular
journal, for it will not be clear what proportion of any particular
journal's articles is OA as yet.
But this gradual
Green option is at the same time serving the
immediate
best interests of research, right now, for it allows the individual
author to have immediate OA for his own findings, today (thereby
immediately maximizing
their visibility and impact).
In contrast, Shulenburger's "shrinking embargo" proposal -- which
envisioned an eventual direct transition from Gray to Gold, but with
the embargo interval between the time when a journal issue was
published and
the time when it was made freely accessible online being gradually
reduced -- would not provide OA at all
for a long, long time to come. Moreover, much of the
benefit of OA occurs at research's "growth tip" (starting from the
pre-refereeing preprint stage, even
before publication, to about a year after the publication of
the postprint).
That is why OA is defined as immediate
toll-free access. ("Needless Pruning of Research's Growth-Tip" http://bmj.bmjjournals.com/cgi/eletters/328/7430/1#45401
).
Nor does a shrinking embargo provide a buffer against catastrophic
cancellation (as the embargo period approaches zero); nor does it
provide a gradual
transition scenario for converting to the OA (Gold)
cost-recovery model (author-institution publication charge per outgoing
article)
from the TA cost-recovery model (user-institution access-tolls per
incoming journal) -- if and when that should ever evolve.
In contrast, the Green road offers the research community the
option
of immediate OA (but via author/institution self-help, rather than
via publisher-conversion) and it allows both journals and institutions
the time to prepare for a
possible eventual transition (though not a necessary or certain one, as
TA and OA might go on peacefully co-existing indefinitely) to Gold:
a leveraged transition, as
follows:
If growing competition to the journals' TA versions of articles from their authors' self-archived plain-vanilla OA versions ever does start to produce some cancellation pressure -- note that this would not be wholesale cancellation of a particular journal by all libraries, because of the anarchic nature of OA growth (article by article instead of journal by journal) but anarchic individual cancellations, by some libraries, of some individual journals -- then TA journals can gradually adapt to it, first by cutting costs (by cutting out the features that are no longer essential) and then, perhaps (if it should ever become necessary) by converting to the OA journal cost-recovery model (with the institutions' annual windfall institutional TA subscription savings now available to cover their annual OA publication costs)
The biggest uncertainty about the direct transition to Gold today is
whether the Gold cost-recovery model is viable. Will
author/institutions
be willing to pay, and where will they find the funds? (Subsidies to
pay for the small number of gold journals that exist today are not a
realistic predictor: would a subsidy model scale up to all 24,00
journals,
should they all go Gold?) Nor is it yet clear how much
author/institutions will have to pay, because no one knows what the
essentials for Gold journal publishing will be, and what inessential
features and their costs could be cut:
"Online Self-Archiving: Distinguishing the
Optimal from the Optional"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/0248.html
"Separating Quality-Control
Service-Providing from Document-Providing"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/0466.html
"Distinguishing the Essentials from the
Optional Add-Ons"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/1437.html
"The True Cost of the Essentials
(Implementing Peer Review)"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/0303.html
"The True Cost of the Essentials
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/1973.html
"Re: The True Cost of the Essentials
(Implementing Peer Review - NOT!)"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/1966.html
"Journal expenses and publication costs"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/2589
"Re: Scientific publishing is not just about
administering peer-review"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/3069.html
"Author Publication Charge Debate"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/1387.html
It might be, for example, that many of the added-values of TA
journals
will no longer be necessary: Will author-institutions want to keep
paying the publisher for the cost of generating and distributing a
print
edition, of doing XML mark-up or creating PDF, of online distribution
and archiving, even of copy-editing and proof-reading?
This cannot be decided a priori. It is only the "competition"
between
the publisher's enhanced TA version and the author's plain-vanilla OA
version that can settle what is still essential and worth paying for,
and what can be dropped in the era of universal OA. It could even turn
out that a continuing market for the "inessential" added values will be
sufficient to sustain TA publishing for a long time into the OA era,
perhaps forever, with no need for the transition to Gold. (I don't
believe
this will be the case, but it cannot be prejudged with certainty
either.) http://www.ecs.soton.ac.uk/~harnad/Tp/resolution.htm#4.2
It is more likely, though, that the eventual effect of the
cancellation
pressure during the transitional Green period in which TA journals
co-exist with growing OA-provision through author self-archiving, will
be to cause journals to downsize and cut costs by phasing out most or
all
of the inessentials listed above, leaving only the costs of
implementing
peer review to be recovered (offloading all text-generation onto the
author [and soon-to-be-designed XML authoring tools] and all
access-provision
onto the network of
interoperable institutional OAI-compliant
open-access
archives of self-archived articles).
Moreover, this Green leveraged-transition period will not only
have
guided -- via cancellation pressure -- the journal-publishing
community's
downsizing and cost-cutting while at the same time providing the
researcher community's all-important OA, thereby determining what the
essentials really are, and how much they really cost, but it will also
have generated the revenues out of which to pay for them (in place of
the indeterminate subsidies envisioned currently):
For the flip side of the TA "cancellation pressure" that guided the
publisher cost-cutting is of course windfall TA savings for the
cancelling
institution! Those annual windfall savings were the ones that used to
pay the costs of the inessentials in the TA era. Whenever an entire
journal
is cancelled, the institution saves the costs of both the essentials
and
the inessentials. It follows that the fraction of the total amount that
institutions are currently paying for all articles
subscribed/licensed via TA will be available to pay for the essentials
only, per incoming outgoing
article published -- if and when the conversion to the Gold-based
cost-recovery model ever has to be made.
In other words, the funds for covering Gold journals' costs are
there
already ("in
the system," as Peter Suber puts it), probably several
times over (depending on what does and does not turn out to be part of
the essentials). So with a gradual leveraged (Green) transition to
the Gold cost-recovery model, there is no need to rely on the uncertain
factor of finding extra subsidies to cover indeterminate costs.
In contrast, the Shulenburger "NEAR" proposal has been around as a
proposal for years, has brought us no nearer to OA, and contains no
mechanism for a transition to Gold with the shortening of the embargo
interval. Embargoed access is not Open Access. Journal publishers know
very well
that most of their revenue comes from the first year after publication,
and that they give up almost nothing in making their contents available
after that. If a timetable for gradually shortening the embargo
interval
were ever actually implemented (which it has never been!), whether
journal
by journal or collectively, it would only be a recipe for approaching
a catastrophe point, not for gradual adaptation and a smooth, leveraged
transition
to Gold, as the anarchic Green option is. (Harnad, Stevan (2001) AAAS's
Response: Too Little, Too Late.
Science dEbates http://www.sciencemag.org/cgi/eletters/291/5512/2318b
).
It is not that embargoed access for only 6-12 months is not
preferable
to permanently embargoed access (just as lower-toll access is
preferable
to higher-toll access); but open access is the antithesis of embargoed
access! So finite-embargo publishing should not be represented as
either OA publishing or a step toward OA publishing. (Finite-embargo
publishing would probably have happened anyway, in the online age,
irrespective of the possibility or the demand for OA.) And shrinking
finite-embargo publishing is either incoherent or catastrophic. (More
likely just a concept that would put OA on indefinite hold --
embargoing
it indefinitely.)
No: OA can and should be provided right now. It's already long
overdue! But
it need not be had at the cost of putting TA journals at needless risk,
or asking them to make needless sacrifices. The transition to Green is
low-risk. After that, nature can take care of itself.
Nor is subsidized access open access! HINARI provides
subsidised
access (sometimes low-toll,
sometimes no-toll) for the no-market sectors of the world. It is
provided
at the expense of the toll-paying sector. Hence it is by definition not
something that can generalise to open access for everyone. It is just
as incoherent (from the standpoint of a smooth and gradual transition
to
OA) as the shrinking-embargo strategy! Both are Escher-style
"impossible
figures."
Embargoed access is too little,
too late, for the very purpose
of open access, which is to accelerate
and augment research progress and productivity. Publishers do not have
to be pushed or shamed -- or waited for! The Green strategy depends
only on the research community. The only thing
the publishing community needs to be "shamed" into doing is giving
self-archiving its blessing, ex officio! And 93% of journals have
already done that. In exchange, they have a
long
grace period to see what will be the effects of growing OA-provision
via self-archiving -- with plenty of time to adapt to it (and without
being demonized as the enemies of research access and impact in the
meanwhile!).
The Darwinian evolution under the Green solution will not be journal
vs. journal but feature vs. feature. Which of journals' current
features
(and their costs) will turn out to be necessary for journal survival
and which not? Competition between the publisher's enhanced TA version
and the author's self-archived plain-vanilla OA version will settle
this. (My bet is that the only essential feature will prove to be
administering peer-review and certifying its outcome.)
How can the many University Eprint Archives be linked
into a seamless whole? That is what the OAI (Open Access
Initiative) metadata harvesting
protocol was designed for: http://www.openarchives.org/
The medatada
of all OAI-compliant archives are interoperable. That means they share
the same tagging, and therefore it is as if they were all in one global
virtual archive, seamlessly searchable. (All have the AUTHOR, TITLE,
DATE, PUBLICATION-NAME, etc. metadata tags.)
This is why the free GNU Eprints software http://www.eprints.org/
was designed: So that universities and research institutions
could immediately create their own OAI-compliant OA Archives,
all interoperable, hence integrable, with one another.
http://www.arl.org/sparc/core/index.asp?page=g20#6
The "semantic web" -- which is in reality the "syntactic web"!
-- http://lists.w3.org/Archives/Public/www-archive/2002Sep/0114.html
--
is certainly a help too, as are all text-analytic resources, including
citation-based search, navigation and ranking http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/2237.html
http://citebase.eprints.org/cgi-bin/search
and google-style inverted full-text boolean search:
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/3169.html
which the next release of the Eprints software provides (and google-scholar
can already provide, if it is restricted to the OAI subset of
google-space).
So once we reach 100% OA, the full-texts of all the 24,000 journals,
across all disciplines
and
languages and years (2.5 million articles per year) will be as
efficiently
searchable and navigable as just their abstracts and metadata are
today,
in databases such as PubMed, Inspec, Chem Abstracts, Scirus, Scopus and
ISI's
Web of Science -- but augmented also by google-style full-text search.
http://www.openarchives.org/service/listproviders.html
http://oaister.umdl.umich.edu/o/oaister/
The only thing that's missing is those 2.5 million annual articles,
most of which still remain to be self-archived: The ball is in the
research community's court! http://www.ecs.soton.ac.uk/~harnad/Temp/self-archiving_files/Slide0052.gif
and it is already clear who has to pick it up and run with it: Researchers' institutions and funders.
They need to extend their existing publish-or-perish mandates -- which
are already weighted by the research impact of the publications -- to further mandate that
the impact of all published journal articles must be maximized by
self-archiving them in the author's own institutional OA Archives, so
as to make them accessible to all would-be users worldwide, even ift
their institutions cannot afford the TA version.
Researchers themselves have already stated that they will not
self-archive
till they are required by their institutions and research funders to
do
so --
but then the vast majority say they will
do so, and do so willingly. In
other words, it is the same as with requiring them to publish (or
perish):
Incentives are needed; the prospect of one's findings being read, used
and
cited is not enough, unless accompanied by carrots and sticks!
References
Harnad, S., Brody, T., Vallieres, F.,
Carr, L., Hitchcock, S., Gingras,
Y,
Oppenheim, C., Stamerjohanns, H., & Hilf, E. (2004) The
Access/Impact Problem and the Green and Gold Roads to Open Access.
Serials Review 30.
http://www.ecs.soton.ac.uk/~harnad/Temp/impact.html
Cox, J. & Cox, L. (2003) Scholarly Publishing Practice: The ALPSP report on academic publishers' policies and practices in online publishing. Association of Learned and Professional Society Publishers. http://www.alpsp.org/2004pdfs/SFpub210104.pdf
Harnad, S. (1995) Electronic Scholarly Publication: Quo Vadis? Serials Review 21(1) 70-72 (Reprinted in Managing Information 2(3) 1995) http://cogprints.ecs.soton.ac.uk/archive/00001691/00/harnad95.quo.vadis.html
Harnad, S., Carr, L., Brody, T. & Oppenheim, C. (2003) Mandated online RAE CVs Linked to University Eprint Archives: Improving the UK Research Assessment Exercise whilst making it cheaper and easier. Ariadne 35 (April 2003). http://www.ariadne.ac.uk/issue35/harnad/
Hitchcock, S., Woukeu, A., Brody, T., Carr, L., Hall, W., and Harnad, S. (2003) Evaluating Citebase, an open access Web-based citation-ranked search and impact discovery service http://opcit.eprints.org/evaluation/Citebase-evaluation/evaluation-report.html
Kurtz, Michael J.; Eichhorn, Guenther; Accomazzi, Alberto; Grant, Carolyn S.; Demleitner, Markus; Murray, Stephen S.; Martimbeau, Nathalie; Elwell, Barbara. (2003) The NASA Astrophysics Data System: Sociology, Bibliometrics, and Impact. Journal of the American Society for Information Science and Technology http://cfa-www.harvard.edu/~kurtz/jasis-abstract.html
Kurtz, M.J. (2004) Restrictive access policies cut readership of electronic research journal articles by a factor of two. Harvard-Smithsonian Centre for Astrophysics, Cambridge, MA http://opcit.eprints.org/feb19oa/kurtz.pdf
Lawrence, S. (2001) Online or Invisible? Nature 411 (6837): 521. http://www.neci.nec.com/~lawrence/papers/online-nature01/
Shulenberger, D.E. (1998) Moving With Dispatch to Resolve the
Scholarly Communication Crisis: From Here to NEAR. Association of
Research Libraries. http://www.arl.org/arl/proceedings/133/shulenburger.html
Swan, A. & Brown, S.N. (2004) JISC/OSI Journal Authors Survey Report. http://www.jisc.ac.uk/uploaded_documents/JISCOAreport1.pdf http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/3628.html
