Re: Outcome of Berlin 3 at Southampton UK

From: Stevan Harnad <>
Date: Tue, 10 May 2005 18:49:24 +0100

On Tue, 10 May 2005, [Identity Deleted] wrote:

> I have read your e-mail on the Berlin conference with some interest
> because I have felt for some time that we in the scientific community must move
> towards an open access model. However, I don't understand how the
> [recommendation] of the conference will be carried out.
> As you stated, only about 5% of the journals are open access.
> If one publishes in a journal that is not open access how can one make
> the text of ones report available on the web without violating copyright laws?

The way its author can make an article Open Access is to self-archive
it in his own institutional repository.

92% of journals have already given their green light for this:

There is no need either to reform or replace the current journal
publishing system, just to supplement it by self-archiving.

> I don't see how self archiving will ever work unless there is a
> central repository like arXiv. Right now the system is very haphazard.
> Perhaps Google Scholar will help. As I understand it, most publishers have
> been resisting the efforts of NIH to establish a central repository for
> biomedical papers.

The purpose of the OAI protocol (which all OAI archives, central and
institutional, including arxiv) share is to make all OAI archives

That means they all use the same metadata tagging scheme (author, title,
journal, date, etc,) and can all be harvested into virtual "central"
archives for seamless search and retrieval, etc. without the user having
to know in which particular archive any particular article happens to be

One such virtual central archive is OAIster:

Google scholar is another, but as it is not attuned to the OAI protocol it
is less functional as yet.

Another is citebase, and that is very functional, with citation linking
(apart from the fact that only 15% of the annual 2.5 million articles
published have as yet been self-archived anywhere, in a central archive
or an institutional one):

NIH made a *huge* and needless strategic error in requiring self-archiving in a
central archive (PubMed Central) -- while at the same time not requiring
but only "requesting" archiving at all -- and even that, not immediately
upon acceptance, but within 12 months of publication.

Publishers can and will object to a 3rd-party central archive. (They can
treat it as a rival publisher.) But as you see from the Romeo directory,
they cannot and hence do not resist authors self-archiving their very
own article output in their very own OAI Institutional Repositories,
in order to maximize their usage and impact, for the sake of research
progress and productivity.

These are all very strong reasons for *not* copying the flawed NIH model:

A far better model for an OA self-archiving policy has been provided by the
UK Parliamentary Select Committee on Science and Technology:

And almost that identical policy was the one recommended for the signers
of the Budapest, Bethesda, and Berlin Declarations on Open Access:

There have also been several recent very important JISC-commissioned studies
by Swan & Brown that (a) surveyed researchers internationally and across
all disciplines and found that 81% of them would self-archive *willingly*
if required to do so by their employing institutions and/or their research funders
(14% more reported they would comply, but reluctantly, and only 5%
would not comply). For this reason, and for many others having to do
with cost, efficiency, and functionality, the JISC studies strongly recommended
(b) institutional archiving followed by central harvesting, rather than central

(Arxiv is a special case and is so successful not because it is central,
but because physicists have been self-archiving for 15 years -- and were
already sharing their preprints even earlier, in paper. Citeseer, in
computer science, is twice as big an archive as arxiv, but it is a virtual
archive, harvested from distributed institutional web-pages -- and not
even OAI-compliant institutional archives!
Moreover, even in physics there is a lot more self-archived than what
is self-archived in arxiv, and this is harvested by Physdoc, another
virtual central archive harvesting from arxiv as well as thousands of
institutional websites: )

> With regard to the romeo site, I don't understand what a postprint is.


> Do the green publishers allow me to post a pdf of a published article on my
> website?

Some green publishers do, some don't. Who cares? It's not necessary for
OA. What needs to be self-archived is the final, refereed, corrected,
accepted draft.

All these things have been discussed extensively in the American Scientist
Open Access Forum.

Stevan Harnad
A complete Hypermail archive of the ongoing discussion of providing
open access to the peer-reviewed research literature online (1998-2005)
is available at:
        To join or leave the Forum or change your subscription address:
        Post discussion to:

UNIVERSITIES: If you have adopted or plan to adopt an institutional
policy of providing Open Access to your own research article output,
please describe your policy at:

    BOAI-1 ("green"): Publish your article in a suitable toll-access journal
    BOAI-2 ("gold"): Publish your article in a open-access journal if/when
            a suitable one exists.
    in BOTH cases self-archive a supplementary version of your article
            in your institutional repository.
Received on Tue May 10 2005 - 18:49:24 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:52 GMT