In Open Access News, Peter Suber comments on John Harnad's critique
of the CERN plan for gold OA
http://www.earlham.edu/~peters/fos/2007_03_04_fosblogarchive.html#117330629987428697
Harnad, John (and others) (2007) Debating the future
of physics publishing. Physics World 29 (3): 22
http://cogprints.org/5448/
http://physicsweb.org/articles/world/20/3
Here are my comments on Peter Suber's comments. The full text of John
Harnad's critique appears at the end. (As will be seen, on some points
I agree with John Harnad, and on others I agree with Peter Suber.)
>> JH 1: 'Green' OA can achieve [OA] quite adequately, without transferring
>> the cost burden to researchers.
>
> PS: True. But under the CERN plan, there would be no burden to researchers
> either. Journals in particle physics would convert from TA to OA, and
> the institutions that formerly paid subscriptions would thereafter pay
> author-side publication fees. Authors themselves would pay nothing.
This has neither been tested nor even thought-through. There are about
10,000 research universities in the world, perhaps 3000 "core"
universities, and maybe 800-1200 mainstay institutional subscribers
for the average journal.
There are very few one-discipline research institutions like CERN.
Subscriptions are collective annual packages and commitments.
Institutional BioMedCentral-style "memberships" are not: Journals cannot
contract to accept N articles annually from a particular university
and universities cannot contract to submit N articles annually to a
particular journal.
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/3753.html
Even if annual quotas can be estimated annually from prior-year
averages, this does not scale to universities that did not previously
subscribe to the journal, yet publish articles in it.
Nor does it scale to universities that have many journals in many
disciplines, and cannot readily make special arrangements for a few
journals and institutional contributors to that journal.
I -- like Peter Suber and unlike my brother -- *do* believe that this
kind of redirection will be possible if/when *all* journals in *all*
disciplines at *all* universities are being cancelled because of 100% or
near-100% Green OA.
Doing that redirection now, however, pre-emptively, journal by journal
-- especially with no necessary match between subscription input and
publication output at the journal or field level at a given university --
may look like it makes sense to one-field institutions like CERN, but
it looks very different to the overwhelming majority of the c. 10,000
universities that exist -- or even the 800-1200 subscribing universities
that make up the mainstay of each journal.
CERN may be able to talk this "consortium" of 800-1200 into a
BioMedCentral-style "membership" agreement, but the question then is
whether it will last, or scale. There is a *huge* speculative element in
the assumption that this can all be done as smoothly as CERN anticipates,
in the short- and long term. And either way: what is the point of doing it
now, when what is urgently needed is much more OA, not top-down business
experiments in fields where OA is already well along its way?
>> JH 2: Journals must generate revenue by one or more of the following
>> mechanisms....
>
> PS: This short list oversimplifies the situation. The majority of OA
> journals charge no author-side fees and we don't know much about what
> business models they use instead. But we do know that some receive
> direct or indirect institutional subsidies, and some generate revenue
> from a separate line of non-OA publications, auxiliary services,
> membership dues, endowments, reprints, or a print or premium edition.
> None of these revenue sources appears on JH 's short list.
Peter is right that JH's list of funding sources and business models is
not exhaustive, but, as Peter says, we don't know much about the long-term
viability of these other business models either -- nor whether they
would scale to more journals or all journals.
Again, a needless push is being given toward an untested business model at
a time when (1) what is urgently needed is more OA in other fields, not
new business models, in a field that is already more advanced in OA than
most; and (2) for the reason mentioned earlier, it is not clear whether
the pre-emptive "redirection" plans scale even within the field in
question, rather than creating hardships for universities that don't
fit the CERN model.
>> JH 3: In most areas of physics...the choice boils down either to
>> "subscriber pays" or "author pays".
>
> PS: The DOAJ lists 199 peer-reviewed OA journals in physics (excluding
> astronomy), of which 13 charge no author-side publication fees. That's
> about 6.5%, even before the CERN plan takes effect.
Again, if that 6.5% is itself sustainable (rather than short-lived)
that still leaves JH's point applicable to 93.5% of OA journals --
and much higher, once we consider the percentage of all physics journals
(of which the OA ones are only about 10%). So it is probably quite
realistic to say it's a choice between subscriber-institution pays on
author-institution pays; and that there are no known, viable options
other than subsidy or volunteerism, which do not scale.
>> JH 4: Although some public funding agencies have expressed themselves
>> in favour of OA, none have indicated willingness to increase their
>> total funding to cover such extra expenses.
>
> PS: The European Research Council is willing, although I believe its
> willingness was only made known this week. In any case, the point is
> moot for the CERN plan, since the publication fees will be covered
> by the members of the CERN-assembled consortium.
This is a straight "redirection" of the core 800-1200 subscribers from
subscription charges to consortial "memberships." It might or might not
work; short- or long-term. What is sure is that it's not what's needed
urgently today; and it is likely it will cause problems for at least some
institutions and researchers.
And it will not advance the cause of Green OA or Green OA mandates.
(The ERC funding would mean a redirection of scarce research funds of
precisely the sort that researchers worry about.)
>> JH 5: There is also a mistaken notion that 'Gold' OA is more cost
>> effective, because electronic papers are much cheaper to produce and
>> distribute. But this has more to do with advances in technology than
>> the OA model itself.
>
> PS: Not true. Several kinds of savings can be traced to the OA model
> itself: OA dispenses with print (or prices the optional print
> edition at cost), eliminates subscription management, eliminates
> DRM, eliminates lawyer fees for licenses and enforcement, reduces or
> eliminates marketing, and reduces or eliminates profit margins. Also
> note that one of CERN's findings in June 2006 was that "sponsoring
> all journals ready for OA at the time of the enquiry would require an
> annual budget of 5-6 Million ¤, significantly less than the present
> global expenditure for particle physics journal subscriptions."
I agree completely with Peter that getting rid of the paper edition
and offloading access and storage onto the distributed IR network would
cut both costs and services in a way that merely going online would not.
But to realise those economies, the paper edition first has to be
abandoned, and the offloading network need to be in place and filled.
Cancellation pressure from universal Green OA might just give rise to
that outcome. But the pre-emptive redirection instead being contemplated
here sounds more like just re-baptising the current core subscriptions as
"consortial memberships" at the current asking price. So no economies,
just redirection.
>> JH 6: [T]he scientific quality of journals switching to the author-pays
>> model may be adversely affected.
>
> PS: For the case on the other side, see my October 2006 article, Open
> access and quality.
I agree with Peter that this is not a worry at all.
>> JH 7: The ideal of open access can largely be achieved, however, simply
>> by encouraging deposit of all publications in freely accessible
>> archives.
>
> PS: Agreed!
But that is the point! What is the point of going in this needless,
untested direction when what is urgently needed is to mandate (not
"encourage") Green self-archiving, in the rest of particle physics as
well as the rest of the disciplines and institutions, so we can have
100% OA at long last? Why the Gold Rush instead?
Stevan Harnad
------------------------------------------------------------------
Harnad J (et al.) 2007. Debating the future of physics publishing.
Physics World 29(3): 22
http://physicsweb.org/articles/world/20/3
Clarifying Open Access: its implications for the research community
John Harnad, Director
Mathematical Physics Laboratory
Centre de recherches mathématiques
Montréal, Canada
When referring to open access (OA) publishing (January, pp22--23),
it is important to distinguish between two different approaches,
sometimes called "Gold" OA and "Green" OA. In the former, a journal
charges nothing for reader access, at least in its electronic version,
while in the latter, the journal charges for subscriptions but allows
authors to deposit copies of their papers in freely accessible archives.
"Gold" OA is currently being advocated by Rudiger Voss and others at CERN.
Viewed however from beyond the confines of a huge, well-funded, particle
physics laboratory, this model may not be in the best interests of the
research community. If the objective really is to provide universal
access to scientific research, rather than merely finding ways to reduce
journal subscription costs, "Green" OA can achieve this quite adequately,
without transferring the cost burden to researchers.
Journals must generate revenue by one or more of the following mechanisms:
subscription charges, direct support from public or institutional grants,
advertising revenues, or page charges to authors. In most areas of physics
direct grants to publishers or advertising revenue are not adequately
available, so the choice boils down either to "subscriber pays" or
"author pays". Relying solely on revenue from paper subscriptions while
offering electronic versions for free is not a viable business model
since most libraries would simply cancel their paper subscriptions. Gold
OA journals therefore have little choice but to transfer the cost burden
from subscribers to authors.
This would adversely affect most researchers. While in experimental
particle physics the extra cost may be only 1-2% of research funding,
in other areas, such as theoretical and mathematical physics, it could
be as high as10-15%. Researchers without access to substantial research
support would be particularly penalized by such page charges; e.g, (as
pointed out by John Enderby) those from developing countries, and those
who receive little or no funding under a highly selective process of
grant attribution. Although some public funding agencies have expressed
themselves in favour of OA, none have indicated willingness to increase
their total funding to cover such extra expenses.
There is also a mistaken notion that "Gold" OA is more cost effective,
because electronic papers are much cheaper to produce and distribute. But
this has more to do with advances in technology than the OA model
itself. Moreover, the cost savings could be beneficially applied to
reducing subscription rates for "Green" OA journals, increasing their
viability. It is also erroneous to expect that savings from libraries
canceling paper subscriptions will somehow be passed directly to
researchers as compensation for the extra costs imposed on them; the
sources of such funding are generally completely distinct. Finally,
the scientific quality of journals switching to the author-pays model
may be adversely affected.
Given currently available resources, a large-scale switch to "Gold"
OA is thus not in the interests of the research community. The ideal
of open access can largely be achieved, however, simply by encouraging
deposit of all publications in freely accessible archives. Naturally,
such archives do not provide quality assurance through peer-review nor
guarantees of long term preservation. But the parallel existence of
"Green" OA journals with publicly accessible archives provides both,
while making the results of scientific research universally available.
Received on Thu Mar 08 2007 - 05:24:42 GMT