Open Access vs. NIH Back Access and Nature's Back-Sliding

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

        ** Apologies for Cross-Posting **

NIH's proposed "Public Access" policy of requesting that NIH-funded
research should be made freely accessible online 6-12 months after
publication is not Open Access (OA), nor is it a satisfactory substitute
or compromise for OA, nor is it a policy that helps OA happen sooner.

NIH-6/12 is Back Access (BA) (as in back-issue or back-volume), not Open
Access (OA), and if the NIH-6/12 proposal were cloned and copied by other
research funders and other nations in the mistaken belief that it was OA
or would help hasten OA, that cloned NIH-6/12 policy would in fact lock in
a 6-12 month delay/embargo period for years to come, and this would
(unintentionally) set back the prospects of OA very substantially for
years to come.

Some signs of this untoward effect of NIH's ill-conceived BA-6/12 proposal
are already visible: The Wellcome Trust has already adopted
(pre-emptively) BA-6. And Nature Publishing Group, formerly green on
immediate self-archiving of the peer-reviewed postprint, has recently made
a press release -- perhaps timed (unsuccessfully) to coincide with the
expected (but now delayed) announcement of NIH-6/12 -- to the effect that
Nature is Back-Sliding from its postprint green policy and replacing it by
BA-6 (presumably in line with NIH-6/12):

Although one cannot legislate by lexicon, the meaning of the recently
coined term "Open Access" is:

    Immediate, permanent, online access to the full-texts of
    peer-reviewed research journal articles, free for all users,
    webwide

The term was coined to contrast Open Access with Toll Access, in which the
only users who can access and use the articles online are those whose
institutions (or the users themselves) can afford the publisher's access
tolls (subscription, license, or pay-to-view). Note that TA is spatially
restricted access -- only users at the right place can have access --
whereas OA is spatially unrestricted access for all would-be users,
everywhere. What about temporally restricted access?

The purpose of Open Access is to maximize the usage, impact and benefits
of research articles, by making them available to *all* their would-be
users worldwide, not just to those whose institutions can afford Toll
Access. It is through research uptake and usage that research progresses.
Indeed, that is why research is published at all: to be accessed, used,
applied, built upon.

The difference between current issues and Back Issues or Back Volumes is
clear. It's the difference between current research and past research,
between the growth region and the static core, between cutting-edge
immediacy and past history.

How much difference does a 6-12 month access delay make, then?

Although this will no doubt vary somewhat with the discipline involved, it
is *particularly* true in the fast-moving biomedical sciences (NIH's
focus, after all) that research usage and impact and progress begins from
the moment a refereed piece of research is made available to the world
research community (even earlier, at the pre-refereeing stage, sometimes),
and things can potentially move lightning fast thereafter -- *if* the
results are accessible to use and build upon.

Any needless access-delay from that moment onward is exactly that:
needless delay, hence needless loss of research access, usage, impact
progress, and benefits. And it is precisely so as to put an end to that
needless delay and loss that the Open Access initiative came into being:
Temporal access restrictions are every bit as inimical to the progress of
research as spatial ones are.

It must not be forgotten that it is the online medium (the Web) that has
made it possible to put an end to all needless delay and loss in research
usage and impact. Before the advent of the online medium, the costs and
constraints of paper publication and distribution made Open Access an
impossible proposition, regardless of how beneficial it would have been
for research progress. Now it is 100% feasible and fully within reach to
make all refereed research immediately accessible to all its would-be
users worldwide. Hence all further delay and loss of research access and
impact now amounts to needless and unjustifiable loss and delay.

Can the access delay be justified by considering factors other than its
effects on research? If there were any credible evidence that Toll Access
publishing and cost-recovery cannot peacefully co-exist with authors
immediately making supplementary copies of their peer-reviewed drafts OA
by self-archiving them for all would-be users whose institutions cannot
afford the official Toll Access version then there might be grounds for
further reflection on this. But all the evidence is precisely in the
opposite direction:

There are (Toll Access) physics journals whose articles have been made
accessible for free online in author-provided supplements since 1991, and
for some, 100% of their contents have been freely accessible in this way
for years now, yet their subscription revenues have not eroded. The
American Physical Society (APS) was the first publisher to give its green
light to author-provided free-access online supplements. One physics
journal -- Journal of High Energy Physics (JHEP) http://jhep.sissa.it/ --
launched in 1997 as a (subsidised) Open Access journal, even successfully
converted back to Toll Access cost-recovery in 2002, by migrating to a
subscription-based publisher (IOP http://www.iop.org/EJ/journal/JHEP). All
of JHEP's contents remained freely accessible: before, during and since.

So, no, there is no sense whatsoever in research funders and research
employers mandating self-archiving 6-12 months too late instead of
immediately, when it is needed most; and no justification whatsoever for
Publishers to Back-Slide from giving their green light to immediate
self-archiving in favor of merely encouraging Back Access 6-12 months
after the access was needed most!

Stevan Harnad