On Mon, 31 May 2004, Joseph J. Esposito wrote (on liblicense)
http://www.library.yale.edu/~llicense/ListArchives/0405/msg00143.html

> 1.  Does anyone know of any library cancellations of journals because of
> the availability of some or all of the articles in such journals in self-
> or institutional archives?  I do not know of any such cancellations
> myself, but I wonder if I am once again embarrassingly underinformed.

I know of precisely the opposite, and it is very important to understand
that a flurry of journal cancellations by libraries would be precisely
the *wrong* way either to greet and encourage journals going green or to
encourage authors to act upon it by self-archiving. This is yet another
example of the very urgent need to unbundle the journal affordability
problem from the access/impact problem.

    Harnad, S., Brody, T., Vallieres, F., Carr, L., Hitchcock,
    Gingras, Y., Oppenheim, C., Stamerjohanns, H., & Hilf, E.R. (2004)
    The green and the gold roads to Open Access. Nature (web focus)
    http://www.nature.com/nature/focus/accessdebate/21.html

The case illustrating the opposite is physics, in which there has been substantial
self-archiving since 1991, with some fields already at 100% OA (Open Access)
for years now. There have been no journal cancellations as a result; rather, the
result was that the American Physical Society (APS), the publisher of the most
important journals in the field, was the first publisher to go green, many years
ago. It would have been a great shame, extremely short-sighted, and an even
greater retardant on OA, if the reward for APS's pains had been cancellation!

   "Evolving APS Copyright Policy (American Physical Society)" Blume (1999)
    http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/0471.html

An even stronger counterexample comes from the very area of physics
where self-archiving first flourished, and where it reached 100% first:
High Energy Physics (HEP):

    (We don't have the data for HEP alone, but they are a subset of nuclear
    and particle physics, which is now at approximately 50% OA overall:
    please compare the green bars in nuclear/particle physics alone
    http://www.ecs.soton.ac.uk/~harnad/Temp/self-archiving_files/Slide0032.gif
    with those for all fields of physics:
    http://www.ecs.soton.ac.uk/~harnad/Temp/self-archiving_files/Slide0031.gif )

The Journal of High Energy Physics (JHEP) was "born gold" in 1997, that
is, it was created as an OA journal in this field that had already been
self-archiving since 1991 and was already at or near 100% OA by 1997.
Within one year, JHEP achieved an ISI impact factor of 6.6
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/1814.html
and it just kept rising:
http://www.ictp.trieste.it/~ejds/seminars2002/Loriano_Bonora/Jhep_presentazione/impact01.htm

Nevertheless, by 2002, JHEP had difficulty making ends meet as an OA
journal, so it converted from gold to green, the title being taken
over by the Institute of Physics (IOP) and offered on the usual
subscription/license toll-access model.

    "JHEP will convert from toll-free-access to toll-based access" Harnad (2002)
    http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/1812.html

The first thing to note is that in that case institutional libraries did exactly
the *right* thing, which was to subscribe to JHEP in virtue of the fact that it
was an important, high-quality, high-impact journal.

The second thing to note is that all the *articles* in JHEP are and remain OA to
this day, because their authors continue to make them OA by self-archiving them.
This is what the authors of all the 2.5 million annual articles in the 24,000
peer-reviewed should all be doing, as of now.

In keeping with the importance of separating the article access/impact problem
from the journal pricing/affordability problem, I will not speculate here about
the role of pricing in the decision of so many institutions to subscribe to JHEP
after it converted from gold to green. Reasonable prices are always
desirable and advantageous. But the substantive point is that --
rather than speculating about and perhaps even encouraging library
*cancellations* as the reward for publishers taking the positive step of
going green so authors are encouraged to provide OA -- we should consider
JHEP as evidence of the possibility of peaceful co-existence between OA
via self-archiving and the continuing support of the journals that have
given it their green light.

> 2.  Assuming cancellations because of self-archiving are negligible or
> nonexistent, at what point, if ever, would one expect such cancellations
> to begin? Or are we to imagine that there will be no cancellations and
> that the widespread acceptance of "Romeo Green" standards will have no
> economic impact on publishers' revenues from libraries (and, thus, no
> impact on reducing libraries' expenditures)?

The purpose of OA provision by researchers is in order to maximize
access by -- and hence usage and impact from -- other researchers,
for the sake of research progress and productivity. The access/impact
problem was first brought to light by the affordability/pricing problem,
but they are *not the same problem,* and the solution to the one is not
the solution to the other, and vice versa.

At this point in the evolution of OA it would be far better if these
two problems were disentangled from one another. Libraries should of
course continue doing whatever they can to minimize journal prices and
maximize journal affordability. But they should not treat OA as one of
the means to this end, as that would only work to the disadvantage of OA.

Moreover, OA provision through self-archiving grows anarchically,
article by article, rather than journal by journal, mooting questions
of cancellation at this early point, where institutional OA provision policies
are what is urgently needed, not institutional green-journal cancellation
policies!

    "The Green Road to Open Access: A Leveraged Transition" (Harnad 2004)
    http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/3378.html

If you want to cancel, cancel the gray journals, not the green ones!

http://www.ecs.soton.ac.uk/~harnad/Temp/self-archiving_files/Slide0037.gif
http://www.ecs.soton.ac.uk/~harnad/Temp/self-archiving_files/Slide0036.gif
http://www.ecs.soton.ac.uk/~harnad/Temp/self-archiving_files/Slide0028.gif

> 3.  Assuming that cancellations or their threat do occur, how are
> publishers likely to respond?  Will they watch their businesses whither
> away?  Will they step back from "Romeo Green"?  Or will they migrate the
> value away from the articles themselves (which presumably are free to one
> and all through a well-tuned Google search) to other facets of their
> subscription services? This is the point that I am personally most
> interested in.

What would be most helpful at this point is not continuing passive
speculation about hypothetical future contingencies, should OA ever
prevail, but immediate action on implementing policies and practises to
ensure that OA does prevail.

http://www.eprints.org/signup/sign.php

Stevan Harnad