IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

Journal of Machine Learning Research (fwd)

Stevan Harnad harnad at cogprints.soton.ac.uk
Tue Oct 16 13:24:54 EST 2001

Please see the clarification below of the chronology (and hence the
causal sequence) of the new Kluwer policy on author/institution
self-archiving online.

Leslie Pack Kaelbling, who founded the alternative Journal of Machine
Learning Research because of editorial board and author dissatisfaction
with Kluwer policy (particular policy on free online access), points
out that whereas it is true that the new Kluwer policy, permitting
author self-archiving of their refereed papers online, is a reasonable
one, and would have made it unnecessary to found an alternative
journal, it was only under pressure from the threat to resign and found
an alternative journal, and eventually the actual execution of that
threat, that Kluwer policy finally changed.

There is a lesson in this for journal publishers: Do not try to prevent
author/institution self-archiving. Not only is it in direct conflict
with what is in the best interests of research and researchers, and not
only does it risk losing the editorial board and forcing it to start up
a rival journal, as happened here in the case of Machine Learning, but
it is futile to try to stop it in any case, because author/institution
self-archiving of refereed research can already be done, perfectly
legally, even if the author is forced to sign the most restrictive of
copyright statements. See:

  6. How to get around restrictive copyright legally

So I stand corrected: 

The petition/boycott/resignation/alternative-journal pressure serves a
further useful purpose inasmuch as it encourages publishers to update
their author self-archiving policies explicitly to realign them with
PostGutenberg reality. This is useful for helping to embolden those
authors who may be hesitating to self-archive their refereed papers
because of a (mis)perceived legal obstacle. The legal obstacle is
apparent, not real, in any case; but, human nature being what it is, it
could only help if the appearances were formally corrected.

Stevan Harnad

---------- Forwarded message ----------
Date: Tue, 16 Oct 2001 13:30:23 -0400 (EDT)
From: Leslie Pack Kaelbling <lpk at ai.mit.edu>
To: harnad at ecs.soton.ac.uk
Cc: peters at earlham.edu
Subject: Journal of Machine Learning Research

I just got, fourth hand, a message that you sent, probably to Peter's
[FOS] discussion forum. I tried to read and/or reply to it there, but
couldn't for some reason.

In any case, I wanted to make it clear that Kluwer *only* allowed
authors to put their papers into eprint archives (or their home pages)
a few days ago, just after the major resignation letter was
circulated. If, in fact, this had been their position all along, I
probably wouldn't have bothered starting JMLR.  We (the MLJ editorial
board) demanded for years that they allow free electronic
distribution, somehow, but we didn't get anywhere.

Now, as far as I'm concerned, they have a pretty reasonable position.

- Leslie

> Date: Tue, 16 Oct 2001 03:52:24 +0100
> From: Stevan Harnad <harnad at COGPRINTS.SOTON.AC.UK>
> Subject: Constructive response and policy from Kluwer
> Further to the publicity that has been given to the editorial board
> resignations from Machine Learning, please see the following message
> from its Executive Editor, Robert Holte, as forwarded by Peter Suber of
> FOS (Free Online Scholarship).
> Note that Kluwer's policy here is very reasonable, and that the key
> feature is that it formally allows author self-archiving. That is all
> that is needed. With that, the entire refereed corpus can be freed
> online. As the publishers point out, now that this
> I hope the research community will soon realize that they are simply
> barking up the wrong tree if they focus on trying to force journal
> publishers to give away their contents for free, or on founding new
> rival journals to wrest their contents from them an
> All the research community needs to do is realize it, and just go ahead
> and do it. And let journal and journal publishers continue doing what
> they are good at, and necessary for (implementing peer review, and
> providing the on-paper and on-line options).
> Stevan Harnad

NOTE: A complete archive of the ongoing discussion of providing free
access to the refereed journal literature online is available at the
American Scientist September Forum (98 & 99 & 00 & 01):


You may join the list at the amsci site.

Discussion can be posted to:

september98-forum at amsci-forum.amsci.org

More information about the Jrnlnote mailing list

Send comments to us at biosci-help [At] net.bio.net