Subject:  Re: Draft IFLA Manifesto on Open Access
From:  Stevan Harnad <harnad@ecs.soton.ac.uk>
Date:  Mon, 31 Mar 2003 02:31:16 +0100 (BST)

On Fri, 28 Mar 2003, Alex Byrne wrote:

 > IFLA MANIFESTO ON OPEN ACCESS
 > TO SCHOLARLY LITERATURE AND RESEARCH DOCUMENTATION
 >
 > IFLA (the International Federation of Library Associations and
 > Institutions) is committed to ensuring the widest possible access to
 > information for all peoples in accordance with the principles expressed
 > in the Glasgow Declaration (see http://www.ifla.org).

It is highly desirable and commendable to be committed to the widest
possible access to information. But in order to promote *open access*
it is essential to be far more specific about the *nature* of the
information. In particular, the IFLA Manifesto is doomed to fail and to be
ignored if it does not make a specific and explicit distinction between
information that its creator *does* wish to give away, and information
its creator does *not* wish to give away. (Notice that I said *creator*
and not *publisher*.)
http://www.ecs.soton.ac.uk/~harnad/Tp/resolution.htm#1.1

If this distinction is not made, and followed, the manifesto is
incoherent, and makes it seem as if henceforth even the authors of
best-selling novels must give their work away online, whether they wish
to or not. (I repeat, this concerns the meaning of "open access.") It is
also desirable that you specify that the IFLA is concerned mainly with
*writing*, because otherwise your manifesto gets into the sticky area of
software (not all of which is intended by its creators as a give-away)
and even patent information (likewise not a creator give-away).

So my first suggestion is to specify that insofar as "open access* is
concerned, the IFLA Manifesto is only concerned with author-give-away
information, and mainly with give-away writings. Then a very specific
body of such writing can be named, as it is in the BOAI:
Refereed journal articles
http://www.earlham.edu/~peters/fos/boaifaq.htm#openaccess

The reason this particular body of writing is the target is that it is
and always has been an author give-away, written purely for research
impact (i.e., to be read, used, cited, built-upon, applied), and not
for author income from sales of the text. Hence the tolls charged by
publishers for access to the text have always cost their authors,
institutions and funders a great deal in terms of lost impact. This loss
in research impact because of access-denying tolls is in turn borne also
by society and by research progress itself. Hence all would benefit from
open access -- *to this literature*; but this argument does not apply to
information as a whole.
http://www.ecs.soton.ac.uk/~harnad/Temp/self-archiving.htm

It is also important to stress -- for this specific literature, the
research reported in the refereed journals (about 20,000 worldwide,
2,000,000 articles annually) -- that the reason open access is necessary
is for the sake of the research itself, and its potential benefits to
researchers, institutions, and the society that funds it. Of course an
immediate consequence of open access to this research will be benefits
to other users too: developing countries, students and teachers, the
general public. But it is a strategic mistake to portray these
additional benefits of open access as the main reason for seeking it
(for the same benefits would arise from freeing access to the
non-give-away literature, not written for impact but for income, and
there they would not be compelling reasons). The main rationale for open
access to the refereed research literature has to be research impact
itself, which comes from usage by researchers, and is currently blocked
by access-tolls.

 > IFLA declares that comprehensive open access to scholarly literature and
 > research documentation is vital to the understanding of our world and to
 > the identification of solutions to global challenges.

Even this is too general. It is better than "all information," but
"scholarly literature and research documentation" is still too vague and
includes work written for income as well as work written only for
impact. The distinction must be made clearly and explicitly, and honored.

 > IFLA acknowledges that the discovery, contention, elaboration and
 > application of research in all fields will enhance human well being,
 > progress and sustainability.  The peer reviewed scholarly literature is
 > a vital element in the processes of research and scholarship.  It is
 > supported by a range of research documentation which includes preprints,
 > technical reports and records of research data.

At last we arrive at the right target literature, but it would be better
if we had gone there directly, explicitly stating the two-fold basis
for singling out this literature: (1) It is an author give-away,
written by its authors only to be used and applied, not for income
from access-tolls and (2) the access-tolls block usage and impact,
and hence the progress of research itself.

It would also be good to add a reminder of exactly what has *changed*:
(3) In the on-paper era, the access-tolls were necessary to cover the
costs of dissemination; in the on-line era they are no longer necessary
at all (because the only remaining essential cost is that of implementing
peer review).

These peer-reviewers provide their refereeing services for free (just as
these authors give away their writing for free). The only remaining
essential cost -- the IFLA is simply asking for trouble if it denies
or ignores it -- is the cost of *implementing* the peer review. This
cost is no more than $500 per paper, and can easily be covered by the
institutional windfall savings if and when access tolls vanish (the
average collective cost to licensed institutions, jointly, per paper,
is $2000 currently). But this requires admitting that if open-access
is achieved by eliminating access-tolls, then all 20,000 toll-access
journals need to become open-access journals, charging institutions
for peer-review of their outgoing papers, instead of charging them for
access to incoming journal-articles.
http://www.nature.com/nature/debates/e-access/Articles/harnad.html#B1
http://www.ecs.soton.ac.uk/~harnad/Tp/resolution.htm#4.2

But note that the IFLA Manifesto would weaken itself if it were merely a
a call to toll-access publishers to convert to open-access publishing.
They will not and need not heed the call (they have heard it all many
times before). But there is a second path to open-access that does not
depend on publishers' agreeing to convert to open-access right now, and
that is: author/institution self-archiving of all institutional research
output, pre- and post peer review. This can take place (and is taking
place) in parallel with toll-access publication. It is a way for these
give-away authors and their institutions to maximize their research
impact right now. It is growing. And publishers cannot and will not try
to block it, for blocking it is tantamount to declaring themselves (and
declaring refereed journal publishing) to be dedicated to blocking
research impact in order to maximize publishing revenues, an untenable
position in the online era. Hence publishers are in fact supporting
self-archiving in growing numbers (55% do so already).
http://www.lboro.ac.uk/departments/ls/disresearch/romeo/Romeo%20Publisher%20Policies.htm
http://www.stm-assoc.org/infosharing/springconference-prog.html

And it is likely that open-access through self-archiving will
peacefully co-exist with toll-access for a long time to come, allowing
a stable transition to open-access publishing if and when it becomes
necessary. Meanwhile, toll-access will be used by researchers at those
institutions that can afford it, for the journals they can afford,
whereas open access will be available for all those who cannot.

 > IFLA notes that the worldwide network of library and information
 > services provides access to past, present and future scholarly
 > literature and research documentation; ensures its preservation; assists
 > users in discovery and use; and educates users to develop appropriate
 > information literacies.

All true, but not relevant to the give-away/non-give-away distinction
that is at the heart of all of this. Libraries provide access, etc.,
to both kinds of writing.

 > IFLA advocates the adoption of the following open access principles by
 > all involved in the recording and dissemination of research ^Ö including
 > authors, editors, publishers, libraries and institutions ^Ö in order to
 > ensure the widest possible availability of scholarly literature and
 > research documentation:

It is a mistake, I think, to make a joint appeal to both (give-away)
authors and publishers. Separate appeals are needed. To authors:
self-archive your toll-access journal papers (and publish in open-access
journals where possible). To publishers: Support self-archiving and
consider converting to open-access where possible.

 > 1. Acknowledgement and defence of the moral rights of authors,
 > especially the rights of attribution and integrity.

A worthy but completely irrelevant point. The need for protecting
authorship and text integrity is not specific to either the give-away or
the non-give-away literature; it applies to both, and applies whether it
is on-paper or on-line, and whether it is toll-access or open-access. It
is a mistake to wrap open-access considerations in any of this. It is
not at issue.

 > 2. Recognition of objective and effective peer review processes to
 > assure the quality of scholarly literature irrespective of mode of
 > publication and without distortion to support extraneous purposes such
 > as confirmation of tenure or promotion of faculty.

Again, affirming the value of peer review is worthy, but not the issue.
Open access to the peer-reviewed literature is the issue. The peer
review comes with the territory. (It would also be better for the
library community not to make needless declarations on tenure
procedures: supporting open-access is enough. Tenure procedures need not
change, nor does peer review, for open access. Set aside the incorrect
notion that new open-access journals are being blocked because they
are ignored by promotion committees. They are not, and that is not
the problem.)

 > 3. Promotion of measures to facilitate publication of quality assured
 > scholarly literature and research documentation by researchers and
 > scholars in developing nations, from indigenous peoples and among those
 > otherwise disadvantaged.

This too is not an open-access issue. Founding or supporting
peer-reviewed journals in developing countries, and/or helping papers
from developing countries to appear in peer-reviewed journals in
developed countries is again worthy, but not an open-access issue, and
only confuses the issue. On the other hand, open access will certainly
benefit the developing countries, both by giving them access to the
peer-reviewed literature worldwide, most of which they cannot afford
now, and by increasing the visibility and impact of their own research
output.

 > 4. Protection under copyright of all scholarly literature and research
 > documentation for a strictly limited period determined by law for the
 > benefit of authors followed by succession to the public domain for the
 > benefit of all peoples.

A worthy goal, but not particularly helpful for open-access, which is
needed for 2,000,000 annual articles in 20,000 journals, which are
appearing as we speak. Open-access does not mean open-access a year or
many years later. That is not the way to maximize research impact.
http://www.ariadne.ac.uk/issue28/minotaur/

 > 5. Strengthening of fair dealing provisions in international copyright
 > agreements and directives, national laws, and publishing contracts and
 > licences to ensure unhindered access by other researchers and the
 > general public.

Fair dealing is not specific to give-away refereed research, and it will
not solve the problem of open access:
http://www.ecs.soton.ac.uk/~harnad/Tp/resolution.htm#5.2

 > 6. Assurance of the availability to all peoples of all scholarly
 > literature and research documentation which has been designated by its
 > authors to be made available through preprints, open access journals and
 > archives, or other means.

At last we come to the point of open-access: It now remains to specify
the two means: BOAI-1 (self-archiving) and BOAI-2 (open-access
journals).
http://www.soros.org/openaccess/read.shtml

 > 7. Implementation of affordable mechanisms to enable access to scholarly
 > literature and research documentation by the peoples of developing
 > nations and all who experience information inequality including the
 > disabled and otherwise disadvantaged.

Again this is too general, not specific to open-access, and seems to
apply to the older initiatives for lower-toll access. Nor is the
open-access problem unique to developing nations: Open access is needed
for all would-be users, everywhere, whose access to refereed research is
blocked by access-tolls.
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/2171.html

 > 8. Inclusion of provisions in law, contracts and licences to ensure
 > preservation in perpetuity of all scholarly literature and research
 > documentation in libraries and archives in formats and under conditions
 > which will ensure enduring availability and usability.

Preservation is another worthy general digital-library issue,
but not relevant to the specific cause of open-access:
http://www.eprints.org/self-faq/#1.Preservation

 > 9. Operation of effective systems by libraries and publishers to ensure
 > the preservation in perpetuity of all scholarly literature and research
 > documentation with authenticity and continuing usability guaranteed.

Far more important (for immediate open access) are effective systems by
libraries for generating open access for their own institutional
research output (by self-archiving it).
http://www.eprints.org/self-faq/#libraries-do

Note that Open Access might also extend to some esoteric research
monographs with no market and likewise written for research impact rather
than royalty income:

"What About the Author Self-Archiving of Books?"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/0450.html
"Journal Papers vs. Books: The Direct/Indirect Income Trade-off"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/0317.html
"University Library Publishing"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/0930.html
"Death of the Book"
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/2807.html
Harnad, S., Varian, H. & Parks, R. (2000) Academic publishing in the
     online era: What Will Be For-Fee And What Will Be For-Free? Culture
     Machine 2 (Online Journal)
http://culturemachine.tees.ac.uk/Cmach/Backissues/j002/Articles/art_harn.htm
http://cogprints.soton.ac.uk/documents/disk0/00/00/17/00/index.html
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):

     http://amsci-forum.amsci.org/archives/september98-forum.html
or
     http://www.cogsci.soton.ac.uk/~harnad/Hypermail/Amsci/index.html

Discussion can be posted to: september98-forum@amsci-forum.amsci.org See
also the Budapest Open Access Initiative:
     http://www.soros.org/openaccess

the BOAI Forum:
     http://www.eprints.org/boaiforum.php/

the Free Online Scholarship Movement:
     http://www.earlham.edu/~peters/fos/timeline.htm

the SPARC position paper and resources on institutional repositories:
     http://www.arl.org/sparc/core/index.asp?page=m0

the OAI site:
     http://www.openarchives.org

and the free OAI institutional archiving software site:
     http://www.eprints.org/