> I agree that the self archiving can be done unilaterally, and with the Santa
> Fe protocol work, I have put in place here work to get us up to speed for
> self-archiving while assuring that we comply with Santa Fe.
Bravo!
> The harder one, in my view, is the copyright assignment, where there's a
> Prisoner's Dilemma problem. None of my faculty (and generically, anybody's)
> feel that it is in their individual self interest to combat the copyright
> problem. It takes a field-wide response. That was one of the major
> benefits of the Los Alamos archive: since everybody was using it, the
> publisher's threat not to publish users of Los Alamos was hollow. Self
> archiving, it seems to me, does not go as far in solving that problem.
But there is absolutely no need to combat anything! Look at this entirely
non-combative flowchart:
(1) Authors can self-archive their pre-refereeing drafts the same
day they submit them to the journal for refereeing. There is
absolutely no copyright problem there. Go to (2).
(2) The day they receive their acceptance letter for the final
draft, along with the copyright agreement, if it permits
online-self-archiving, they archive the final, revised draft too.
If the copyright agreement does not permit it, go to (3).
(3) If the copyright agreement explicitly forbids online
self-archiving, authors strike out that passage and sign (or write
in an explicit enabling passage, if it is absent). If the amended
agreement is accepted, go to (2). If the amended agreement is
contested, go to (4).
(4) Link the self-archived pre-refereeing draft (1) to a
self-archived file containing the full list of (nontrivial) changes
that turn it into (2) the refereed, revised, accepted final draft.
No confrontation. No Prisoner's Dilemma. And everything is 100% legal.
Having the updates in a separate file is an inconvenience, to be sure,
but it will put the open archiving initiative over the top, freeing the
literature (most of us would definitely rather consult this free,
though hybrid version, than none at all or one that has to be paid
for, for most of our daily uses). Eventually, as a consequence of this
(not-quite-optimal-yet) freeing of the literature, the copyright
restrictions will be seen to be futile and will be dropped, and we will
have arrived at the optimal and inevitable.
> I have seen your references to the Los Alamos Lemma before, and I think it
> is absolutely correct. It's an existence proof that is important in all of
> these discussions.
Yes, but the Lemma applies as fully to distributed archives as to
central ones. It is a Lemma about SELF-ARCHIVING in general, not about
central self-archiving in particular:
The Los Alamos Lemma
"If X is a purported obstacle to self-archiving, one that must first
be overcome before we can do it (e.g., copyright, preservation,
plagiarism, journal restructuring, independent refereeing), and X
did NOT detain the Los Alamos Archive, then X is NOT an obstacle to
self-archiving!"
It will be a question for historians why Physicists "twigged" on all
this so much earlier than everyone else. I suspect it's partly because
they're more serious about their research, and perhaps on average
smarter, too. But I don't think it's because their archive happened to
be central! They're smart enough to know that the entire Web is
interconnected; and now that the Open Archives will be fully
interoperable, it is as if they WERE all just one central archive.
http://xxx.lanl.gov/cgi-bin/show_monthly_submissions
And NO publisher ever ventured to threaten users (I assume you mean
self-archivers) of Los Alamos: the blatant conflict of interest would
have been too naked, and to great. (But if they had been so inclined,
wouldn't it have been EASIER rather than harder for publishers to
identify these villains if they all archived centrally rather than in
distributed open archives?)
So I think your conjecture is wrong. But nothing hangs on it in any
case, because I have now described a trivially easy way to get around
the copyright problem perfectly legally. The only reason no one thought
of such trivial loopholes before was because we kep thinking of
copyright in terms of theft-of-text, with the author, normally, just as
eager to prevent this as his publisher, because they both stand to lose
revenue from such theft.
But self-archiving by the give-away refereed-journal-paper author runs
entirely counter to the intended protective function of copyright, a
protection that for the trade literature extends to author and
publisher alike. For in this give-away literature there is a profound
conflict of interest between the author and the publisher (what I have
called the "Faustian Bargain"); for the author wishes to maximize the
access to and impact of his research by giving it away, whereas the
publisher wishes to restrict access only to those who have paid.
So it would never have occurred to an author of, say, a best-seller, to
give away his book for free on the Web, and, in order to avoid violating
the copyright agreement he has made with his publisher, to self-archive
the original submitted draft, and then self-archive a list of all the
corrigenda that will turn it into the final draft! This would be a
completely absurd and empty exercise for a book author, who is as eager
not to lose the potential royalties from his book to pirated copies as
his publisher is.
But with the refereed journal literature the picture is entirely
different: And it always has been! It is just that the technology and
economics of print on paper made it impossible for these give-away
authors to do anything about it: Either toll-gates were erected to cover
the substantial costs of paper publication, or there could be no
PUBLICation of the paper at all. So the author had to cut his losses,
and trade toll-gated impact for no impact at all.
Until now (the Web era). For now we must stop thinking in these old and
irrelevant terms and go ahead and do the only sensible thing with
research findings that we have no interest in making personal pennies
on, let alone allowing them to cripple our universities' serials
budgets: self-archive them publicly, for free for all.
So back the open archiving initiative, and stop worrying about
copyright: "self-piracy" is not only completely DISanalogous to the
piracy (by others) of the books, magazines, music and software that
copyright law is designed to prevent, but it is also something against
which the "self" WISHES no copyright protection in this case! And that
is why loopholes like self-archiving the preprint and then the addenda
in order to get around restrictive copyright agreements -- which are
in conflict with the interests of these give-away authors as well
as with the interests of research itself -- turn out to be so trivially
obvious and easy: no one would have thought of it before, because paper
publication costs could not have been met if the papers were all open
to "piracy" (and that is why we all cooperated in "copyright clearance"
protection against photocopying piracy).
But those days are now over, and we must not let the literature continue
to be held hostage to them.
--------------------------------------------------------------------
Stevan Harnad harnad_at_cogsci.soton.ac.uk
Professor of Cognitive Science harnad_at_princeton.edu
Department of Electronics and phone: +44 23-80 592-582
Computer Science fax: +44 23-80 592-865
University of Southampton
http://www.ecs.soton.ac.uk/~harnad/
Highfield, Southampton
http://www.princeton.edu/~harnad/
SO17 1BJ UNITED KINGDOM
Received on Wed Feb 10 1999 - 19:17:43 GMT