Re: Central vs. Distributed Archives

From: Stevan Harnad <>
Date: Tue, 9 Sep 2003 03:33:07 +0100

On Mon, 8 Sep 2003, Eberhard R. Hilf wrote:

> the physics ArXiv has a linear increase of the number of papers put in per
> month, this gives a quadratic acceleration of the total content (growth
> rate of Data base), not linear.

Maybe so. But slide 25 of (slide 25)
still looks pretty linear to me. And it looks as if 100% was not only
*not* reached at this rate 10 years after self-archiving started in
physics in 1991, but it won't be reached for another 10 years or so...

> Total amount by now may be at 10-15 % of all papers in physics.

(10-15% of the annual output, I assume.)
I count that as appallingly low, considering what is so easily
feasible (though stunningly higher than any other field!)...

> Linear growth of input rate means the number of physicists and fields
> using it rises, while in each field (and physicist) a saturation is
> reached after a first exponential individual rise.

Interesting, but the relevant target is 100% of the annual output
of physics (and all other disciplines) -- yesterday!

> Never there will be a saturation such that all papers will go this way,
> since in different fields culture and habits and requirements are
> different. --

I couldn't follow that: Never 100%? Even at this rate? I can't imagine
why not.

Cultural differences? Do any of the cultural differences between fields
correspond to indifference or antipathy toward research impact -- toward
having their research output read, used, cited? Unless the cultural
differences are specifically with respect to that, then they are

Requirement differences? Are any universities or research funders
indifferent or averse to their researchers' impact? Unless they are,
any remaining requirement-differences are irrelevant.

Habit differences? Well, yes, there are certainly those. But that is
just what this is all about *changing*! Are any field's current
access/impact practises optimal? or unalterable for some reason? If
not, then habit-change is (and always has been) the target!

And the point is that the rate of habit-change is still far too slow --
relative to what is not only possible, but easily done, and immensely
beneficial to research, researchers, etc. -- in all disciplines.

> [That is why it is e.g. best, to keep letter distribution by
> horses at a remote island (Juist) alive since the medieval times].

That I really couldn't follow! If you mean paper is still a useful back-up,
sure. But we're not talking about back-up. We are talking about open
online access, which has been reachable for at least a decade and a half
now, and OAI-interoperably since 1999. What more is the research cavalry
waiting for, before it will stoop to drink?

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):

Discussion can be posted to:
Received on Tue Sep 09 2003 - 03:33:07 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:03 GMT