Re: Berlin-3 Open Access Conference, Southampton, Feb 28 - Mar 1 2005

From: Stevan Harnad <>
Date: Thu, 10 Feb 2005 16:10:16 +0000

The avowed purpose of the international meeting that will be hosted
by Southampton University February 28 - March 1

    "Berlin 3 Open Access: Progress in Implementing the Berlin Declaration
    on Open Access to Knowledge in the Sciences and Humanities"

is to *implement* the Berlin Declaration, so as to turn it into a concrete
institutional policy which institutions that have signed (and will
sign) the Berlin Declaration can then commit themselves to adopting.

The Berlin Declaration itself was only an abstract expression of principle:
Scholarly/Scientific research should be freely accessible online to all
potential users worldwide.

Many institutions worldwide signed that they endorsed that Principle.

But not that they would put the Principle into Practice, or How!

Berlin 2 (at CERN in May 2004)

began drafting a "Roadmap" for implementing the Berlin Declaration:

but the Roadmap was still far too vague to provide a basis for a specific,
concrete, practical institutional policy.

That concrete policy is what the Berlin 3 Meeting in Southampton in
February will try to formulate, and there is a candidate proposal (from
Southampton) on the table, as to what this practical implementation policy
should be:

        Unified Institutional Open-Access Provision Policy:

        I. The institution's researchers EITHER publish their research
        in an Open Access Journal (if/when a suitable one exists)


        II. The institution's researchers publish their research in
        a suitable non-Open Access journal AND also self-archive a copy of
        it in their own institutional Open Access Archive.

This is (roughly) the OA policy that has since been adopted at Southampton:

and of course the self-archiving component (II) is the critical one, as
institutions cannot create or convert OA Journals, nor can they commit
their researchers to publishing in them, but they can certainly create
OA Archives and commit their researchers to self-archiving a copy of
all their research articles in them immediately upon acceptance for
publication (and encourage self-archiving the preprints even earlier).

At least 7 other institutions besides Southampton (2 in Germany, 2 in
France, 1 in Australia, 1 in Portugal, 1 in India) have already adopted
and implemented an institutional policy along these lines:

If this policy (or a suitable variant) is adopted as the Berlin
Declaration's official "Roadmap" for OA in February, then institutional
self-archiving and OA provision should shortly experience a dramatic
growth spurt worldwide.

Also to be present at the Berlin Declaration meeting are the
representatives of two important national research institutions --
France's CNRS and Germany's Max-Planck Institutes. These distributed
multi-disciplinary institutions are far bigger than any single
university, and if they adopt the implementation policy, all other
research universities and institutions will follow suit shortly thereafter

This is especially important in light of a set-back to OA progress
that has just occurred in the US: The NIH (in the earnest hope of
promoting OA thereby) adopted a flawed policy of *inviting* (rather
than requiring) NIH grant-recipients to make their findings freely
accessible online after a delay period of up to 12 months following
publication (rather than immediately) in PubMed Central (rather than
in each author's own institutional repository). One of the purposes
of Berlin 3 is to provide a much better OA implementation policy as a
model, thereby averting any worldwide cloning and proliferation of the
NIH's very inadequate delayed-access policy -- which is certainly neither
OA nor an implementation of the Berlin Declaration, and might have
locked in a 1-year access delay for years to come.

    "Please Don't Copy-Cat Clone NIH-12 Non-OA Policy!"

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

In contrast to the US NIH policy, the UK Parliamentary Select Committee's
formal recommendation (although it has not been adopted by the UK
government) is almost identical to the Unified Institutional OA Provision
Policy described above:

Research Councils UK are currently working on formulating a policy of
their own for implementing the UK Committee recommendation, but RCUK
will not present anything at the Berlin 3 meeting because its date
happens to fall exactly at the delicate time when RCUK are working on
finalising their policy, which has not yet been agreed upon.

(It would of course have been better if RCUK too could have attended Berlin
3 to present its own OA plans along with CNRS and MaxPlanck, but the
timing prevented it: I hope RCUK will announce soon after, and that its
announcement will be favorable, but I have no way of knowing yet what
its decision will turn out to be!)

There is one more theme to be noted in closing: One of the outcomes of
last month's 2-day Southampton Workshop on OA self-archiving in the UK --
"Open Access Institutional Repositories: Leadership, Direction and Launch"

had been a candidate alternative to the NIH delayed-access policy for
universities. It has become apparent across the years that the single most
important obstacle delaying 100% Open Access provision is *keystrokes*:

If there were a way to ensure that all the metadata (author, title, date,
journal-name, etc.) plus the full-texts of all university research article
output were duly deposited in the university's institutional repository --
by (someone) performing the relatively few keystrokes per paper required
to do this (Southampton's logs suggest it only takes 6 minutes per paper)
-- then 100% OA would only be one keystroke away: the keystroke that makes
the full-text (and not just the metadata) accessible webwide rather than
just accessible internally to the author's institution. (The metadata
are visible and harvestable worldwide in any case.)

All the issues that derailed the NIH proposal would be bypassed if
performing that last keystroke were simply left to the discretion of
the author (though strongly encouraged) in any case where there was
any reluctance or uncertainty -- but all the preceding keystrokes (for
entering the metadata and uploading the full-text into the university's
repository) were mandatory.

Several researcher surveys have now confirmed that although researchers
are beginning to realize the power and value of OA self-archiving, many
nevertheless state very explicitly that they will *not* self-archive
until/unless they are *required* to do so by their employers and/or
their research funders -- yet almost 80% say that if/when they
*are* required to do so, they will do so *willingly* (just as they comply
willingly with the requirement to publish-or-perish).

At Southampton University, it turned out that the practical benefits
of having all university research output deposited in the university
repository -- for the purposes of internal record-keeping, asset
management, CV-generation, and research performance evaluation, as well
as for external research assessment (e.g., the RAE), grant applications,
and research visibility -- were sufficient in themselves to motivate
making self-archiving an official university policy.

Whether or not the last keystroke was done to make the full-text visible
externally turned out to be a minor matter, affecting a minor number of
cases, as long as the rest of the keystrokes were done: The metadata are
then (1) all already harvestable and hence (2) all already generating
eprint requests to the author from would-be users around the world, (3)
92% of journals have already given full-text OA self-archiving their
green light, and meanwhile (4) the objective evidence of the power of
OA to enhance research usage and citation impact is growing rapidly.

So as long as those first N-1 keystrokes are done, nature can be trusted
to take its course toward the outcome that is optimal and inevitable
for research, researchers, research institutions, research funders,
and the tax-paying public that supports and funds the institutions and
the funders.

Stevan Harnad
Chaire de recherche du Canada
Centre de neuroscience de la cognition (CNC)
Université du Québec à Montréal
Montréal, Québec, Canada H3C 3P8
tel: 1-514-987-3000 2461#
fax: 1-514-987-8952
Received on Thu Feb 10 2005 - 16:10:16 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:47 GMT