Re: Royal Society Offers Open Choice

From: Stevan Harnad <harnad_at_ecs.soton.ac.uk>
Date: Sat, 24 Jun 2006 23:20:57 +0100

Jan Velterop has both priorities and event-order exactly backwards,
and I suspect he may not even be aware of it.

The priority is Open Access, now. This is an *immediate* and *direct*
research priority as well as a public-good priority, because it is the
public that benefits from research impact and progress, and it is for
that reason that the public funds research. Hence OA's first priority
is OA, 100% OA, not OA publishing, or publishing reform. Moreover, the
order of events, leading to OA publishing through publishing reform, is,
almost certainly: Mandating OA self-archiving --> 100% OA --> possibly
subscription cancellations --> possibly substantial subscription
cancellations --> transition to OA publishing. The order of events is
almost certainly not, instead: transition to OA Publishing --> 100% OA.

Let me repeat: the priority of OA is *immediate* and *direct*: In
particular, there is *zero* evidence at the present time that there
is any *other* problem (such as self-archiving causing subscription
cancellations) that *first* has to be solved *before* we can have
immediate 100% Open Access.

Still more particularly, it is simple *false* to say that we cannot have
immediate 100% OA until we first solve the problem of subscription revenue
losses for publishers, for there is *zero* evidence of subscription
revenue loss for publishers as a consequence of self-archiving, whereas
there is overwhelming evidence of the benefits of OA self-archiving to
research, researchers, and the public that funds them.

There is also overwhelming evidence that merely inviting or recommending
self-archiving does not generate rates of self-archiving above its
spontaneous baseline level of 15%. The only way -- and the sure,
demonstrated way -- to achieve 100% self-archiving is to mandate it.

And *that* is the issue on the table: mandating self-archiving. Not
protecting publishers from hypothetical risk, but mandating
self-archiving, for its demonstrated benefits to research.

Now, in weighing Jan Velterop's remarks below, please do keep this logic
in your mind, because alas those in Jan's position -- indeed anyone
whose primary allegiance is to what is best for publishers' bottom
lines rather than what is best for research, researchers and the public that
funds -- is bound to have great difficulty in keeping this logic in mind,
being preoccupied with their own conflicting interests:

    (1) 100% OA has been repeatedly demonstrated to benefit research,
    researchers and the public that funds research.

    (2) Immediate, 100% OA has been repeatedly demonstrated to be
    achievable in practise, rapidly and reliably, by mandating OA
    self-archiving.

    (3) There exists no evidence whatsoever to date that OA self-archiving
    reduces subscriptions.

    (4) Publishers are nevertheless lobbying against mandating OA
    self-archiving (with no supporting evidence) on the grounds that it
    might threaten their business.

    (5) In place of mandating OA self-archiving, publishers are now
    lobbying for mandates to pay publishers their asking price for
    providing paid OA.

    (6) The asking price is being set at a time when subscriptions are
    paying all publishing costs and there exists no evidence at all that
    self-archiving reduces subscription revenue.

    (7) If research funders and researchers are able and willing, right
    now, to mandate and provide for paying publishers' asking price,
    all is well.

    (8) But if research funders and researchers are *not* able or willing,
    right now, to mandate and provide for paying publishers' asking price,
    then publishers are delaying and deterring a demonstrated benefit for
    researcher, researchers and the public on the basis of no evidence
    of any actual cost (let alone substantial cost) to themselves.

    (9) The rational and practical thing for research funders and
    institutions to do under these conditions would be to act on what
    has already been demonstrated to be true: Mandate OA self-archiving,
    generate its demonstrated benefits for research, researchers and
    the public, and thereby *test*, at very the same time, whether
    it induces any subscription decline -- and if so whether that
    decline is substantial enough to require restructuring publishers'
    cost-recovery system.

    (10) With the objective evidence that cost-recovery needs to be
    restructured will come the funds for paying for it -- because
    institutional subscription cancellations mean corresponding
    institutional subscription savings, out of which institutions can
    then pay for their researchers' publishing costs using the same
    money that is currently being spent on subscriptions -- instead of
    extra money taken from what is currently being spent on research.

Jan does not see it this way because his first allegiance is to making
sure publishers make ends meet, and because he is convinced that they
will not be able to make ends meet if self-archiving is mandated,
even though there exists to date absolutely no evidence in support of this
conviction. The conviction, in turn, warrants -- not for Jan, who, I believe,
supports the self-archiving mandate despite his reservations, but for many
other publishers -- trying to prevent research funders from mandating
OA until and unless they can agree to pay *in advance* for the hypothetical
subscription shortfall (of which there is as yet not the slightest sign).

The demonstrated and readily reachable immediate benefits of OA to
research, researchers and the public are hence set aside, and hypothetical
risks to the publisher's bottom line are instead given the priority, with
the insistence that if OA is to be mandated, it is OA publishing that needs
to be mandated (and funded), not OA self-archiving.

I add only one other point to reflect upon, before turning to Jan's specific points:

Institutional subscriptions today are not paying only for online access,
but also for the print edition (among other perks). Is the publishers'
"realistic" asking price for author-institution-funder-paid OA meant
to be covering the costs of supplying the paper edition to all those
institutions too? (I will take up this theme again in replying to Ian
Russell of the Royal Society in the next posting).

On Sat, 24 Jun 2006, Velterop, Jan Springer UK wrote:

> Stevan Harnad on Saturday 6/24/2006 on the AMSCI Forum list:
>>
>> "... if mandated SA does generate substantial institutional subscription
>> cancellations, then those very same substantial institutional
>> subscriptions cancellations will generate the institutional windfall
>> savings out of which PA costs (again determined by the market and not
>> by a-priori fiat) could be paid without taking any money away from
>> research funding."
>
> I'm afraid Stevan fails to appreciate three things here:
>
> 1. Access to scientific literature and the formal publishing of articles
> are not optional, but essential parts of doing research, so the cost
> of access and publishing is an essential cost of doing research, and in
> that regard entirely comparable with the cost of laboratory equipment,
> reagents, et cetera;

But at the moment, those costs are being paid by researchers' institutions
in the form of library subscriptions. So why are we talking about extra
sources for money that is already changing hands, and covering costs? On
the hypothesis that self-archiving would cause institutional incoming
subscription cancellations, the money that is already being spent is freed
up to pay for institutional outgoing publication costs. And cancellations
are not even happening yet; nor is it clear they will, when the author's
refereed final drafts are made free online, as the mandates propose. (What
about the market for the print edition, for example?)

So why are we being asked to contemplate the "essential cost of doing
research" as a portion of the cost of "laboratory equipment, reagents, et
cetera" when (1) subscriptions are *not* being cancelled yet as a result
of self-archiving authors' final drafts, (2) there is no evidence yet
that they will be, and (3) if they ever were, the cancellation windfall
seems a much more natural candidate for redirection to pay for OA publication
than the laboratory equipment budget?

> 2. If the cost of essentials is seen as 'taking money away from
> research funding, then money is already being 'taken away' from research
> funding because subscriptions are largely paid out of the overhead that
> institutions take out of research grants (often more than 50%);

Fine. If and when those subscriptions should ever show empirical
evidence of declining toward unsustainable levels as a consequence of
self-archiving mandates, *then* is the time to talk of redirecting funds
to pay the essential costs of publishing, and the natural candidate
source for paying those costs will be those very same (cancelled)
subscription savings, including whatever portion of them is currently
being paid from research funding overheads. But right now subscriptions
are doing fine, and there is no sign that self-archiving does now or
will if mandated diminish them. So it seems like a good time to do the
empirical experiment: mandate self-archiving and test what happens to
subscriptions. (We already know what will happen to research impact.
Meanwhile, leave the lab equipment funds alone!

> 3. Shifting payment patterns from subscriptions to open access
> via institutional self-archiving mandates (the 'windfall' argument)
> is unnecessarily disruptive and as such only delays open access as
> it inevitably causes entirely predictable and understandable doubt
> as to the real intentions and ulterior motives of the OA 'movement'
> (which often seems more about money than about access), and consequent
> defensive attitudes amongst publishers and scholarly societies, and even
> amongst researchers themselves.

Do you mean that delaying OA self-archiving mandates until funders agree
to "redirect" payments toward paying for OA publishing in advance, at
the publishers' asking price, and while funds are still tied up in subscriptions,
without any evidence they will be released as a result of self-archiving
mandates is the *nondisruptive* way of reaching OA? Whereas waiting to see whether
there is any need to redirect, and waiting at the same time for the release of the
funds to be redirected, is the *disruptive* way of reaching OA?

Jan, please stop and reflect, as here you have gotten so entangled in your own
conflict of interest that you have really turned reality upside down:

    (a) OA advocates want one thing, and one thing only: Free online
    access to research.

    (b) They have one motive, and one motive only: To maximize research
    impact.

    (c) And the only defensive attitude of the OA movement is about
    attempts to delay and disrupt reaching 100% OA.

    (d) OA self-archiving mandates will indeed reach 100% OA,
    demonstrably, reliably, and rapidly.

    (e) The ones delaying and disrupting the adoption of OA self-archiving
    mandates are (some) publishers.

    (f) OA advocates are *not* talking about "shifting payment
    patterns from subscriptions to open access via institutional
    self-archiving mandates (the 'windfall' argument)". (That is all
    just counterspeculation -- in response to the publisher speculation,
    based on zero evidence, that mandating that authors self-archive
    their final drafts will cause catastrophic cancellations.)

    (g) What OA advocates are talking about is mandating OA self-archiving, now.

    (h) It is publishers who are thinking "more about money than access" --
    and that is why they are trying to block the OA self-archiving mandates.

    (i) Please don't conflate OA advocates -- who are largely from the
    research community -- with the library community and its ongoing
    struggles with journal affordability.

    (j) The journal affordability problem and the research access problem
    are not the same problem. OA is the sure solution to the latter,
    but not necessarily the former.

> Advocating open access should not be conflated with advocating
> cost-evasion (the ultimate free-ridership). Access and costs are two
> independent variables. Lower costs do not necessarily bring open access;
> and open access does not necessarily bring lower costs.

The public who has funded the research, and the researcher who has
conducted, written up and successfully revised the research according
to the dictates of peer review -- these are surely not "free riding"
if they insist on making their own findings accessible online to all
would-be users and not just to those who happen to be able to afford
the publisher's subscription.

The publisher is -- and continues to be -- allowed and able to sell the
print and online edition, in exchange for the value he has added (mostly
in implementing the peer review, which researchers likewise perform for
free). And there are no signs of any decline in the publisher's ability
to make ends meet in this way when authors self-archive.

If and when self-archiving ever demonstrably reduces subscription
revenue to unsustainable levels, *then* we can consider a transition to
another cost-recovery model. But calling author self-archiving -- or the
mandating of author self-archiving by authors' funders and institutions
-- "free-riding" is next to absurd. Both parties are doing their work,
and getting paid: The publisher, from the (undiminishing) subscription
revenue and the author, his institution and his funder, from the impact of
their research. Is maximizing the impact of the work one has funded and
conducted to be viewed as "free-riding" on the investment of a publisher
who is -- unchangingly -- selling subscriptions to it, as always? (Have you forgotten
that researchers are not getting and royalties from any of this? That all they seek,
and all they ever have sought, is maximal research impact?)

Free-riding would happen if publishers were no longer able to make ends
meet, and authors still expected them to keep implementing peer review,
for free. But of course, if that were ever to happen, or even to approach
happening, the transformation to the paid OA cost-recovery model would
occur quite naturally of its own accord. Today, in contrast, we are merely
talking about pre-emptive, counterfactual speculations about free-riding,
not actual free-riding.

> But we would be able to make a great deal more progress on an
> equal-revenue basis, were that advocated more widely. The amount of
> money now being spent, Academia-wide, on subscriptions, could, almost
> by definition for the vast majority of journals, also fund full open
> access. That's what we should be focussing on.

Jan, I think your are getting unconsciously blinkered by your own
hypotheses here. Construed one way, we are in complete agreement
(even though you criticise my speculation about paying institutional
OA publication costs out of institutional subscription cancellation
savings. What you just said above could be construed as saying precisely
the same thing! Please re-read the quoted passage from me with which
you began this commentary:

> Stevan Harnad on Saturday 6/24/2006 on the AMSCI Forum list:
>>
>> "... if mandated SA does generate substantial institutional subscription
>> cancellations, then those very same substantial institutional
>> subscriptions cancellations will generate the institutional windfall
>> savings out of which PA costs (again determined by the market and not
>> by a-priori fiat) could be paid without taking any money away from
>> research funding."

But you don't really want to say that, exactly, because we are not there
yet (and don't even know we will ever be there): Right now there are no
cancellations, even in fields that have self-archived for over a decade
and have already been at or near 100% self-archiving for several years now.

What you really mean by "equal-revenue" is: (1) Let's peg publishers'
revenues at where they are now, per published article, based on their
current subscription revenues, and (2) let's agree to pay them that same
amount per article up-front, for OA publishing.

That would be fine if (a) the subscription pot were available to redirect funds from
(it is not: it is all tied up in subscriptions today); and, perhaps more important (b)
it is not at all clear what products and services would still be needed and wanted
from a publisher in an OA publishing world: Will we want the print edition? (If not,
factor out that cost in the "equal-revenue" equation.) Will we want the publisher's
XML and PDF, or will we settle for the author's refereed final draft? (If the latter,
cut those costs out of the "equal-revenue" equation too.) Will we still want
copy-editing? Or will we just want peer review? For if it's just about peer review,
with all the rest offloaded onto the network of institutional repositories, with their
peer-reviewed, self-archived article holdings, the "equal-revenue" figure -- then the
publishers asking-price for paid OA today -- will look radically different.

It's always tempting to hope for an a-priori "sweet-heart deal"
between supply and demand, but it's usually the market that needs
to sort such things out. The only way to find out what products and
services institutions will still be willing to pay for in a 100%
OA world is to test it out: Mandate OA self-archiving of the vanilla
refereed final drafts, see whether it causes subscription cancellations
at all; if it doesn't, fine. If it does, try to cut costs and reduce to
the essentials that subscribers *are* still willing to pay for; and if
the only essential turns out to be peer review (with the print edition
and its costs, the publisher's copy-editing and mark-up and its costs,
the archiving and distribution of the publisher's version, etc., all no
longer generating sufficient demand to cover costs, hence jettisoned),
then those much reduced peer-review costs will only account for a fraction
of institutions' windfall savings, and can easily be paid up-front out
of the windfall cancellation savings.

And remember that this is all just playing the game: hypothetical
speculation and counterspeculation. The objective reality is as spelled at
the very beginning of this commentary: OA has already been demonstrated
to be good for research; self-archiving mandates have already been
demonstrated to generate OA; so mandate self-archiving. The rest is
an open empirical question, and anything else is mere speculation and
counterspeculation.

Stevan Harnad

AMERICAN SCIENTIST OPEN ACCESS FORUM:
A complete Hypermail archive of the ongoing discussion of providing
open access to the peer-reviewed research literature online (1998-2005)
is available at:
http://www.ecs.soton.ac.uk/~harnad/Hypermail/Amsci/
        To join or leave the Forum or change your subscription address:
http://amsci-forum.amsci.org/archives/American-Scientist-Open-Access-Forum.html
        Post discussion to:
        american-scientist-open-access-forum_at_amsci.org

UNIVERSITIES: If you have adopted or plan to adopt an institutional
policy of providing Open Access to your own research article output,
please describe your policy at:
        http://www.eprints.org/signup/sign.php

UNIFIED DUAL OPEN-ACCESS-PROVISION POLICY:
    BOAI-1 ("green"): Publish your article in a suitable toll-access journal
            http://romeo.eprints.org/
OR
    BOAI-2 ("gold"): Publish your article in a open-access journal if/when
            a suitable one exists.
            http://www.doaj.org/
AND
    in BOTH cases self-archive a supplementary version of your article
            in your institutional repository.
            http://www.eprints.org/self-faq/
            http://archives.eprints.org/
            http://openaccess.eprints.org/
Received on Sun Jun 25 2006 - 00:27:15 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:48:22 GMT