Re: ALPSP statement on BOAI

From: Stevan Harnad <>
Date: Thu, 11 Apr 2002 22:58:48 +0100

On Wed, 10 Apr 2002, Sally Morris wrote:

> Stevan, why not wait to criticise our new questionnaire until you have seen
> it?
> > The full results of the study, Authors and Electronic Publishing, will
> > be available for sale very shortly and details will appear on our
> > website,

Because (a) ALPSP is selling the results, and I have no intention of
buying them to see them, and because (b) the advance conclusions you
announced in your own postings were sufficient to sustain my comments,
comments very much like the ones I made about the prior ALPSP
questionnaire. If you wish to put on your website anything that shows
that my comments have missed their mark, and how, I will be very happy
to read, and answer, if I can.

    "ALPSP statement on BOAI"

> Your earlier question was about our attitude to OAI. Briefly, the problem
> is this. While DISorganised self-archiving, or even institutional
> archiving, need not threaten the survival of the journals on which (as you
> agree) it parasitizes, organised cross-searchable archives are considerably
> more alarming.

Let me repeat this, so I am sure I understand: Institutional author
self-archiving is ok as long as it's DISorganized?

I'm afraid I don't know what that (or its negation) might mean! The Net
and the Web are, in a sense, fundamentally DISorganized systems, anarchic,
in fact: Everyone puts on their own content. Then services like Google
harvest the contents. And in the case of the OAI it is a protocol that facilitates, indeed
helps organize that content disclosure and harvesting.

So are you saying that it is alright to self-archive as long as the
Archive is not OAI-compliant?

It really will be useful to get a clear and unambiguous statement of
ALPSP policy. Because whereas I was personally unbothered (and am still
unbothered) by the hedges in the ALPSP copyright recommendations --
unbothered enough to list them as models that might be used by others
in the BOAI Self-Archiving FAQ -- I recall that
there were others, some time ago, when the ALPSP recommendations were
first publicized, who were not at all happy with the hedges,
and would have liked to see them clarified:

    "Re: ALPSP creates model Grant of Licence for journal articles"

The passage in question is:

    "You... retain the right to use your own article (provided you
    acknowledge the published original in standard bibliographic
    citation form) in the following ways as long as you do not sell it
    [or give it away] in ways that would conflict directly with our
    commercial business interests. You are free to use your article...
    mounted on your own or your institution's website; [posted to free
    public servers of preprints and/or articles in your subject

Let us call a spade a spade. It is very possible that sooner or later
the existence of free online self-archived versions of peer-reviewed
research papers may be preferred by users over the publisher's
toll-access versions, and this might in turn eventually "conflict
directly" with the publisher's commercial business interests (in the
sense that it might diminish subscription/license revenues, and might
eventually force cost-cutting, downsizing, and transition to
open-access publishing, which probably means scaling down to
little more than author-institution-end peer-review service-provision
instead of the provision of the article itself as a product to the

Anyone who denies this possibility is either being obtuse or untruthful
-- despite the fact that over 10 years of self-archiving in physics has
not yet had this effect.

Now let us continue in this spade-is-spade vein: Why are publishers not
fighting against self-archiving? There are 3 factors, and I will list them
in order of importance (although in reality, any one of them is decisive).

Factor 1. The awkward conflict of interest inherent in trying to oppose
self-archiving publicly:

To oppose author self-archiving of their own give-away research --
i.e., of papers that are given away by the author to the publisher as
well as to the reader, without seeking or receiving any royalty income
in exchange, purely for the sake of reaching as many potential users as
possible, and thereby maximizing the research's impact -- would be to
bring directly into the public eye a very unflattering fact, namely,
that there is a profound conflict of interest between what is best for
researchers, their institutions, and for research itself, hence
society, on the one hand, and what is best for publishers' current
revenues streams and current way of doing things, on the other.

This conflict of interest is not new. What is new is that the unprecedented
possibilities opened up by the online medium have now made it possible
to resolve the conflict in favor of research, through self-archiving.
Hence, to try to oppose self-archiving would simply be for publishers to
declare openly that what is best for their revenue streams takes
precedence over what is best for research.

I don't think any publisher would be eager to publicly imply anything
like that, for it would not only look very bad, but it would only serve
to hasten the exact opposite outcome (which is in any case optimal and
inevitable). So publishers all have to portray themselves as friends
(or at least not foes) of self-archiving, but... [and then the "buts,"
and not any direct opposition to self-archiving, have to do the work of
trying to hold self-archiving at bay] We will return to some of these

Factor 2. The legality of self-archiving:

Whereas a lot of legal-sounding things are said about copyright and
self-archiving in the same breath, there is in fact no legal way to
prevent authors from giving away their own give-away work, even if,
despite Factor 1, publishers were tempted to try to stop them from
doing it. It is not like napster, where consumers are stealing and
giving away the work of others; authors are giving away their own work
when they self-archive it. And at the time when authors publicly
self-archive their pre-peer-review preprints, there is not yet any
publisher at all! So if the publisher tries to impose an
over-restrictive copyright transfer agreement on the author at the
(later) time of acceptance of the final, peer-reviewed draft, the
author need only sign the agreement and append a list of corrigenda (if
any) to the already publicly archived pre-refereeing draft.

    "6. How to get around restrictive copyright legally"

But because of Factor 1, and because of the growing recognition that
publishers do not have any need or justification for asking authors to
transfer to them anything more than all the rights to sell or lease
their article, on-paper and on-line, more and more publishers are now
making the more convoluted preprint+corrigenda strategy (above) for
self-archiving unnecessary, by instead officially allowing the refereed
postprint to be self-archived too. (This is what the highly hedged
ALPSP-recommended text above is trying to do, but its hedging is so
thick that it obscures the view!)

Factor 3. The sluggishness of researchers in self-archiving so far:

The 3rd reason publishers do not try to oppose self-archiving despite
the potential risk to their eventual revenues is that it is a (sad but
true) fact that few researchers have yet begun to self-archive, and
that even what self-archiving there has been has not affected publishers'
revenues at all.

I think it is this third fact that ALPSP are both banking and hedging on,
with this incoherent constraint that "you may self-archive as long as it
does not affect our revenues."

Of course, as a legal constraint on a copyright transfer agreement, that
clause is ludicrous, which is why I wasn't bothered by it, and put it in
among the models. But this is where the "buts" come in: Authors, already
sluggish at best, are certainly not going to be reassured by it, if they
are worried about self archiving. ("So, if people use my self-archived
version, and that makes the publisher lose money, they're going to sue
me?" These are the silly ideas that such incoherent hedges put into
would-be self-archivers minds.)

Never mind. That is where open-access advocates come in. For every bit
of textual obfuscation that publishers formulate that might confuse
authors, we can dispel the smoke, trim the hedges, and show that the
way is clear.

It's just that all this willful obscurantism still leaves publishers with
egg on their face.

I invite Sally to wipe clean the albumen, and give an unequivocal ALPSP
recommendation on self-archiving and open access. The double-talk in
the current ALPSP policy recommendation, and the arbitrary distinction
between "DISorganized" and "organized" self-archiving is not useful or

> Thus the development of OAI is actually making a number of
> publishers, who were previously relatively relaxed, considerably more
> concerned. If search tools in effect allow a user to emulate the original
> journal without having to pay for it, then all the added value - which, as
> we have shown, authors and readers do in fact value highly - will disappear
> because it will no longer be paid for.

What on earth does "emulate the original journal" mean? We're talking
about authors providing free access to their refereed research papers
by publicly self-archiving them online, for use by anyone and everyone
whose institution cannot afford to pay the access tolls of the original

As to the value of the added-value: That remains to be seen. The peer
review is an essential added-value, and my own prediction is that that
will be what refereed journal publishing will downsize to: peer-review
service provision and certification, paid for on the author-institution
end at a cost of about $500 per (outgoing) paper published. The
reader-institutions are currently paying, collectively, $2000 per paper
incoming paper accessed. Hence there will clearly be more than enough
annual windfall institutional savings to cover the cost of the essential
added-value of peer review. But all other added-values are uncertain,
and competition with the free peer-reviewed version looks like a good
way to find out which of them users and their institutions will still
want to keep paying for, once the peer-reviewed draft, without any
extra options, is accessible for free.

(These optional add-ons include the paper version, the publisher's PDF,
copy-editing, mark-up, reference-linking, and publisher-archiving. They
make up the rest of the $2000 total collective cost per paper.
Alternatives include phasing out the paper version and the publisher's
PDF, minimizing the copy-editing, offloading the XML markup to the
authors, and leaving the reference-linking and archiving to the free
OAI services and the institutional OAI Eprint Archives.)

> Hence our emphasis on developing robust new economic models (and a migration
> path towards them) before, and not after, damaging or even destroying what
> is valuable about traditional journals.

Unfortunately, this apocalyptic talk about damage and destruction is
unhelpful too. We are talking not about damaging or destroying what
is valuable about traditional journals, but about providing free online
access to all of it to anyone and everyone whose institution cannot
afford to pay the access tolls, now that they are no longer necessary.
The "damage and destruction" is just a piece of hyperbole to refer to
that risk which has already been mentioned (and which no can or should
deny) to journals' current revenue streams, that might eventually
result from user preference for the free online versions.

But the free online access for all would-be users is already feasible
today, and indeed vastly overdue. It is likely that any revenue
pressure from the growing free-access use will hasten us toward
"developing robust new economic models (and a migration toward them)";
but where is the pressure toward developing these new models -- and
toward downsizing to the essentials and unbundling them from the
options -- going to come from WITHOUT the growing pressure from free

Free access is already within reach. That means all research impact
that has been lost because of access-barriers since self-archiving has
become possible to get around those barriers has been needlessly lost
research impact. How much more of that research impact should continue
to be sacrificed, while we wait for publishers to get around to
developing and migrating to new models? And how long? And why?

> However, perhaps this will turn out to be a non-problem given the widespread
> total ignorance of eprint and preprint archives which we have found outside
> the very specific world of physics!

(The alert reader may recognize here another old friend, namely Factor 3,

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):

Discussion can be posted to:

See also the Budapest Open Access Initiative:

and the Free Online Scholarship Movement:
Received on Thu Apr 11 2002 - 22:59:49 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:30 GMT