On the Need to Take Both Roads to Open Access

[Stevan Harnad <harnad@ecs.soton.ac.uk>]

       On the Need to Take Both Roads to Open Access
                    Stevan Harnad

As something of a veteran in the crusade for open access, I feel that I
have to point out to the growing number of open-access advocates that we
have lately been getting a little carried away with open-access publishing
-- as if it were the *only* way to attain open access, rather than just
one of two complementary ways (open-access self-archiving being the other
way).

This one-sided impression (that open-access = open-access publishing) is
all over the public press at the moment, in the US and Europe. This is a
(gentle) irony that historians will eventually have some fun sorting out:
How did it happen that when at long last we finally began to awaken to the
need for open access to research we first went on to risk losing yet
*another* decade waiting passively for open-access publishing to prevail,
when we could in the meanwhile already have had open access too?

Waiting passively for what? For the 24,000 existing toll-access journals
http://www.ulrichsweb.com/ulrichsweb/ to either convert to open-access of
their own accord or to go belly-up in the face of new open-access
competitors (24,000 of them?) that would capture their authorship. This,
at a time when in reality there existed only about 500 open-access
journals http://www.doaj.org/ -- which is less than 5% of the refereed
research literature even if we double the estimate.

The crux of the matter is this: 24,000 journals (or even ISI's hard-core
8,000) are unlikely to be induced to convert to open-access on the
strength of a press flurry, petitions, declarations, threats to boycott,
promises of government subsidy for open-access author-costs, US
congressional bills, and songs of praise for open access by the research
community and the media worldwide. For there is one glaring omission in
all of this: It is all based on passivity on the part of the research
community.

(It is not even clear what percentage of researchers would actually be
willing to switch from publishing in their currently preferred journals to
open-access journals even if 24,000, rather than just 500, open-access
journals already existed for them to switch *to*!)

Why would publishers take the research community's much publicized
yearning for open access seriously as long as that yearning is expressed
only in this passive way, with the expectation that all the effort should
be made on their behalf by journal publishers, for the sake of this open
access that the researchers purport to need and want so much? Who would
not question the depth of the research community's desire for open access
as long as that desire keeps being voiced only vicariously, rather than
through self-help efforts, as if all possibility and responsibility for
action lay exclusively with publishers?

What will make publishers take the research community's expressed wishes
seriously will be *action* on the part of researchers, taking the powerful
self-help step that is actually within their own power to take right
*now*, in the interest of immediate open access: self-archiving their own
published research output. This will be the only credible (and indeed
irresistible) proof of the research community's desire for open access.
Moreover, it is guaranteed to provide immediate open access for the
research of every author who actually does self-archive.

The only reason the research community is not yet taking this simple
self-help step in sufficient numbers -- they *are* taking it in increasing
numbers, but those numbers are as yet far from sufficient
http://www.ecs.soton.ac.uk/~harnad/Temp/self-archiving.ppt -- is that the
research community does not yet *understand* that this more direct means
of gaining immediate open access for their own research output (through
institutional self-archiving) is already within their reach.

The one-sided emphasis that the research community is currently placing on
the 5% solution (open-access publishing), instead of also promoting -- at
least as vigorously -- the complementary 95% solution (open-access
self-archiving of the remaining 95% of their refereed-research
publications) is now becoming part of the problem instead of the solution,
leaving researchers and their institutions and funders both inactive and
unaware about what they could already be doing in order to provide open
access right now, rather than just waiting passively and hoping that the
500 figure will somehow climb to 24,000 just on the strength of polemics
and wishful thinking alone!

It will take a long time and a lot of effort to spawn or convert 24,000
journals, but their current full-text contents could already be made
openly accessible in to time, if researchers would only take the action
that is already open to them: immediate self-archiving.

The most common brake on researchers' taking this immediate action is an
inchoate worry about copyright. But the proof that copyright cannot be the
real obstacle is already available! Even on the most conservative
construal of the Romeo copyright-policy statistics
http://www.lboro.ac.uk/departments/ls/disresearch/romeo/Romeo%20Publisher%20Policies.htm
at least 55% of the research literature could already be self-archived
(and hence openly accessible) with the journal publisher's formal and
official blessing *today* (indeed, yesterday) -- yet researchers are still
not doing it in anywhere near the numbers that even the most conservative
percentage would allow!

(The potential percentage is in reality much higher than 55%: for the rest
of the authors publishing in journals that do not yet officially support
self-archiving can simply *ask* their publishers, on a per-article basis,
to agree to their self-archiving; many more publishers will agree. And
that percentage can be raised to 100% if the remaining authors, in those
cases where their publisher refuses, simply use the
preprint-plus-corrigenda strategy
http://www.eprints.org/self-faq/#copyright1 ).

But even as the 55% solution, self-archiving trumps the 5% solution by an
order of magnitude, and instantaneously! If only it were actually
practised. But it is not, yet. And that is what needs to be remedied. It
is not remedied by focusing all attention and effort on the 5% solution!

In October, Germany will have a national policy meeting (through its
Max-Planck Societies, and in collaboration with the European Cultural
Heritage Organization) on "Open Access to the Data and Results of the
Sciences and Humanities" with a view to formulating and signing the
"Berlin Declaration," which is meant to be a model open-access policy for
Europe as well as the rest of the world. In November there are Norwegian
and UK national meetings on the same theme. The US has the Public Access
to Science Act (Sabo Bill, H.R. 2613) pending. It is so important that all
of these timely efforts give due weight to *both* of the complementary
open-access strategies, rather than just open-access publishing.

Here is a simple, transparent, unified strategy for an institution, or a
research-funder, or a nation wishing to maximize the access to -- and
thereby the impact of -- its research output:

    (1) All research output should be published in open-access journals
    if and when suitable ones exist (5% of research, currently) and

    (2) the other 95% of research output should be published in
    the researcher's journal of choice, but also self-archived in the
    author's institutional open-access archive -- now.

Our research group at Southampton and Loughborough will soon report data
on the current rate of growth of open access via each of these two
complementary strategies, in terms of the annual number of articles that
are openly accessible each way as a percentage of total published articles
per year so far. We will describe how Tim Brody's citebase
http://citebase.eprints.org/ and citation/usage correlator
http://citebase.eprints.org/analysis/correlation.php can be used to
measure the citation and usage impact of open-access articles and authors,
and how Mike Jewell's standardised open-access CV software
http://paracite.eprints.org/cgi-bin/rae_front.cgi can be used to encourage
and assess research output and impact. We will re-present Steve Lawrence's
finding http://www.neci.nec.com/~lawrence/papers/online-nature01/ that in
computer science, open-access articles are cited 4.5 times as often as
toll-access articles. (And if our own data from an ongoing collaboration
with Charles Oppenheim are ready, we will report the open- vs. toll-access
impact-advantage for other disciplines, in controlled pairwise comparisons
of open vs toll access in the same journal and year, for self-archived and
non-self-archived articles, and across time.)

The cumulative message will be that the 95% solution (self-archiving), if
implemented now, would increase research visibility, research impact, and
hence research progress and productivity substantially. We will will even
estimate graphically how much research impact US, UK, and French research
-- and research in general -- are losing daily, monthly and yearly,
because of *lack* of open access, and how long it would take to stanch
that daily/monthly/yearly loss if the research community pursued only the
passive 5% solution, rather than also actively self-archiving immediately!

A proposal for an institutional
http://www.ecs.soton.ac.uk/~harnad/Temp/archpolnew.html 
and national 
http://www.ariadne.ac.uk/issue35/harnad/ 
self-archiving policy will also be described, to help focus and put into
context open-access efforts such as the Public Access to Science Act
http://publish.uwo.ca/~strosow/Sabo_Bill_Paper.pdf 
and the Bethesda Statement 
http://www.earlham.edu/~peters/fos/bethesda.htm

Stevan Harnad

***

NOTE: A complete archive of the ongoing discussion of providing open
access to the peer-reviewed research literature online is available at
the American Scientist September Forum (98 & 99 & 00 & 01 & 02 & 03):

    http://amsci-forum.amsci.org/archives/september98-forum.html
                            or
    http://www.cogsci.soton.ac.uk/~harnad/Hypermail/Amsci/index.html

Discussion can be posted to: september98-forum@amsci-forum.amsci.org