Re: Momentum for Eprint Archiving

From: Stevan Harnad <>
Date: Mon, 25 Nov 2002 19:50:40 +0000

on Mon, 25 Nov 2002 [identity removed] asked:

> For [newspaper name deleted] I am writing a piece about institutional
> repositories and using that as a starting point to briefly describe the
> transformation of libraries. I have found "For whom the gate tolls"
> to be delightful reading and helpful in preparing my interviews. Now that
> I have spoken to librarians at MIT and Caltech and other places around
> the US, I am still puzzled and was wondering if I can bother you briefly
> for a phone conversation or an e-mail...

Some replies follow below. I also suggest you look at the American
Scientist Forum Archives at:
or, more conveniently navigable, the same archive at:

and the self-archiving FAQ,
created for the Budapest Open Access Initiative:

> If it is true that preserving papers, data sets--in a permanent way--is so
> important, then why aren't all disciplines flocking to the repositories? It
> would see like an obvious solution. And a solution to the subscription price
> dilemma of journals.

You are quite right to ask that question! I asked the very same question
myself, soulfully, in the first paragraph of:

    Harnad, S. (1999) Free at Last: The Future of Peer-Reviewed Journals.
    D-Lib Magazine 5(12) December 1999

The longhand answers (26+ of them!) were in the "Gate Tolls" paper
you've seen
(Zeno's Paralysis section) and have since been updated for the BOAI
self-archiving FAQ:

The shorthand version is that it is taking time to explain to researchers
the direct causal connection between access and impact, and the optimal
and inevitable outcome; but the message is getting through.

And it is also true that lately things have been speeding up, on the
road to the optimal and inevitable. See:

> It does seem as if it is mainly the engineers and, in MIT's case, the
> economists who do find repositories interesting.

Actually, no, I believe the order of precedence, historically, was
physicists first, then mathematicians, then computer scientists, then
the rest. This question of discipline differences is the current topic
of discussion in the American Scientist Forum at the moment:

    "Discipline Differences in Benefits/Feasibility of Open Access?"

> Then there is an open archive for cognitive science.

Yes. CogPrints was founded in 1997. See Peter
Suber's Free Online Scholarship (FOS) Timeline:

> But some biomed researchers tell me they need the "impact" factor of
> the peer-reviewed journal to validate their work. But they want to be
> very widely read.

You asked why it's taking so long? Here's one of the reasons. If you
search google with the words:
    harnad "self-archiving" "self-publishing"
    harnad substitute supplement
you will get an idea of how often and how long I, for one, have
been telling researchers that the self-archiving of peer-reviewed
(published) articles is not self-publication! You get the peer-review
(from the journal's referees) AND the open-access when you self-archive
the outcome! Self-archiving is a supplement, not a substitute, for
peer-reviewed publication. (You read that yourself, in "Gate Tolls,"
yet here you are asking the same question!)

But for some reason, people keep on imagining that self-archiving one's
own papers is something that one does INSTEAD of publishing them in a
peer-reviewed journal, whereas in reality it is -- and always has been,
since it first began in the late '80s) -- something one does IN ADDITION
to publishing them in a peer-reviewed journal, as a way of maximizing
their accessibility and hence their uptake and their impact (to free
them from the access-blocking gate-tolls).

The reason for this persistent error -- I am curious whether your own
article will again just perpetuate the error -- is 2 of the 26 FAQ
items I mentioned:

    7. Peer Review:
    10. Copyright:

People keep on imagining that self-archiving is a *substitute* for
peer-reviewed publication (rather than just a *supplement* to it, so
as to maximize research impact) because they think that copyright
prevents authors from self-archiving their published papers. This is
just plain incorrect. Not only are more and more peer-reviewed journals
now explicitly allowing self-archiving in their formal copyright agreements
but, even with the most restrictive copyright agreements researchers
can achieve almost the same outcome by using the copyright-&-corrigenda

So, contrary to what your biomedical interviewees have been
telling you, there is no trade-off whatsoever between impact and
peer-review. Self-archiving allows researchers to have their cake
(peer-reviewed journal publication), and eat it too (maximize its
research impact by maximizing would-be users' access to it, by
self-archiving it to ensure open access).

> I know this is an old and worn 'anomaly' for you but I was wondering
> if you could explain if institutional repositories, which are open,
> are a good path in your view.

They certainly are. That is why I am trying to promote them every which
way I can:

by promoting free software (designed by Rob Tansley and Chris
Gutteridge) so institutions can create their
own OAI-compliant repositories:

by promoting scientometric engines (designed by Tim Brody and Les Carr)
that do citation-linking (Mike Jewell, Steve Hitchcock) and measure and monitor impact:

by urging universities, research assessors and funders to mandate it:

And that's why I spend so much time
and skywriting
about institutional self-archiving!

> What could explain the difference between the disciplines in their
> affinity to the more subversive publishing models?

I now regret ever calling my original self-archiving proposal
"a subversive proposal"
-- --
because self-archiving is not a publishing model at all (it has almost nothing
to do with publishing); self-archiving is, as its name implies, a means of
maximizing the impact of researchers' own findings, by maximizing their
accessibility, by self-archiving them. Nor is self-archiving particularly
"subversive": The researcher's objective is certainly not to subvert or
in any way harm journals or journal publishers. The objective is only to
protect one's own work from the harm caused by needless impact-loss owing
to access-barriers. Unlike trade authors (whose interest is in sharing
toll-gate receipts with their publishers), researchers give away their
work, not only to their publishers, but to all its would-be users.

The effect of this resolution of the fundamental conflict of interest
(between what's best for publishers' revenue streams and what's best
for researchers' research impact) may or may not be a downsizing and
transformation of peer-reviewed journal publishing along lines I and
others have speculated about, whereby PostGutenberg journal publishing
evolves into peer-review service-provision and certification:

But one thing is not a matter of speculation: that the optimal and
inevitable outcome for research and researchers will be open access.
And it is entirely within their own hands how soon the optimal and
inevitable becomes the actual. Self-archiving is -- rather like other
self-interested personal functions -- something one can only do for
oneself... (But see how digital librarians can help: )

As to whether there are any discipline differences in this, again, see:

    "Discipline Differences in Benefits/Feasibility of Open Access?"

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):

Discussion can be posted to:

See also the Budapest Open Access Initiative:

the Free Online Scholarship Movement:

the OAI site:

and the free OAI institutional archiving software site:
Received on Mon Nov 25 2002 - 19:50:40 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:43 GMT