Re: Journal Publishing and Author Self-Archiving: Complementary Or Competitive?

From: Stevan Harnad <>
Date: Wed, 24 Aug 2005 21:28:03 +0100

            Or: Publishers Will Be Publishers

Not to put too melodramatic a turn on it, but Matt Cockerill's comment
below, from BioMed Central Inc., OA Publishers, does put one in mind
of the custody battle between the two maternal wannabes in the biblical
tale: Who is the true parent of the (OA) infant? who really cares about
its welfare, rather than about something else?

But I think a more charitable interpretation is probably closer to the
truth, since there continues to be so much confusion about OA, even in
the minds of its champions: Matt has been so taken in by an abstract
(and arid) definition of OA that he has lost sight of what it's all for,
what it's about, what the malady is for which it is meant to be the cure.

So I will have to remind Matt -- and anyone else who has lost sight
of it, in the recent welter of mostly irrelevant postings inspired by
a simple attempt to support the RCUK OA Self-Archiving Policy Proposal
against ALPSP's attempt to delay or derail it (as the NIH and UK Select
Committee Proposals were delayed and in the end derailed).

The malady for which OA is the cure is nothing more nor less than this:

Toll Access (TA) is available to those researchers who are at those
institutions worldwide that can afford whatever fraction they can afford
of the planet's 24,000 peer-reviewed journals, which publish about 2.5
million articles each year. No one knows exactly how many researchers
there are on the planet, nor how many articles each can and cannot access
among those articles that they would need to access in their respective
fields. But we can safely say that *many* of them cannot access *much*
of what they need.

We can even roughly estimate *how* much they are missing, by comparing the
use made of (i) what is accessible only via TA with (ii) what has also
been supplemented (sic) by an OA version, self-archived by the author:
We already know that published articles (in the same journal and year)
that *do* have self-archived OA supplements are cited 50-250% more than
those that do not.

So let us (conservatively) say that the malady for which OA is meant to
be the cure is a 50-250% access-gap. If all would-be users had access
to all articles, there would be no access-gap, and no malady.

Now, to continue with the reality: there are two potential cures for the
malady (the 50-250% access-gap, remember), one being "gold" and one being
"green." The gold cure is for publishers to become "good" publishers,
making their contents OA (i.e., freely available to all users webwide)
of their own accord, and covering their expenses some other way than by
denying (online) access to those would-be users who cannot afford to pay.

The golden cure is providing OA today for somewhere between 5-10% of
the articles published annually, because that is about the percentage of
journals that have elected to become OA so far. The number is growing,
but it is growing slowly, because publishers are not eager -- for various
reasons I could discuss, but won't, here -- to convert to the OA (gold)
publishing model.

BioMed Central (BMC) is one of the "good" publishers who have adopted the
OA model and publish about 150 of the approximately 2000 of the 24,000
journals published today that are gold; the other 22,000 are TA. BMC
have published about 20,000 articles since they started publishing in
the year 2000 or so. (Total articles published every year number about
2.5 million, recall.)

No one yet has an accurate and realistic estimate of the annual growth
rate in the number of OA journals, relative to the total number of
journals, but projecting totals from the Directory of Open Access Journals
for the past 24 months, it is clear that 100% or even 25% is far from
sight and reach. So the golden road is a worthy one, but not one that is
likely to bring us 100% OA for many, many years to come, if ever. Nor
is there any direct way to accelerate it, because it is not possible to
*require* journals to convert to OA. (Nor would it be just or legal to
do so, no matter how beneficial it would be to the research community.)

Then there is the green cure, which researchers can and do provide
for themselves: The rough estimate of the percentage of annual articles
being self-archived today is 15-25% (with the JISC survey suggesting that
49% of authors have self-archived at least *one* article).

So there is more annual green OA than gold OA. Green OA is also almost
certainly growing faster than gold OA. But it is still not growing
nearly fast enough. The difference, though, is that green OA, unlike
gold OA, *can* be accelerated, because researchers can be required
(by their employers and their funders) to self-archive their articles.
(And the above JISC survey also found that 81% of authors say they would
comply with such a requirement willingly, and 14% more would comply

It is just such a requirement that RCUK is proposing -- and that ALPSP
(and other publishers) are opposing. (And that was the topic that precipitated
this latest flurry of postings on every topic under the sun, but very little
on the matter at hand.)

And now we have yet another publisher, an OA publisher, good and gold,
BMC: Is BMC supporting or opposing the RCUK self-archiving policy? On
the face of it, BMC is supporting self-archiving, for it offers its
Open Repository service to help institutions set up repositories to
self-archive in.

But unfortunately the planet is now littered with near-empty open
repositories because the problem is not setting up repositories but
*filling* them. Researchers do not self-archive spontaneously in
sufficient numbers. This is the sub-problem for which the RCUK policy
is the intended cure.

So, having established that BMC is not doing much for green simply by
offering a repository service, what is BMC doing to help fill those
repositories? In particular, is it supporting or berating the RCUK

The RCUK policy, to remind you, is to require authors to self-archive
their final drafts -- not the publisher's version: the author's
version. If the RCUK policy is adopted, we can safely project (based
on the JISC compliance estimates above as well as the actual compliance
figures for the only two institutions that have already adopted such a
self-archiving requirement, Southampton ECS and CERN) an increase of OA
in the UK from its current 15-25% level to over 90% within a few years
(and with the rest of the world almost certainly following suit soon

Is Matt writing from BMC to support and celebrate this potential victory
for OA, this potential closing of the 50-250% access gap, this potential
cure for the malady? Or is Matt just writing to promote a product, one that
is "better" than green OA, in fact the only product deserving of the
brand-name "OA"? And if he has a better product/cure, is he proposing
to deprive the patient of the green cure, in favour of waiting patiently
for the eventual arrival of that gold cure? Let the reader judge.

But please keep the following inescapable contingency in mind: If I am
trying to persuade authors to submit their articles to my OA journal
rather than to the TA competition, and I do this by persuading them of
the advantages of OA, what am I to reply if they ask: "But why can't
I just keep publishing in my old journals, and just self-archive my
articles to make them OA"?

Reply? "That's not really OA" or "That is non-optimal OA"...

Comments follow:

On Wed, 24 Aug 2005, Matthew Cockerill wrote:

> Quoting, I believe, Stevan's previous posting to the list:
> "Self-archiving the publisher's PDF is not necessary (except
> where encouraged by the publisher) because the self-archived
> version is a *supplement* to the official toll-access version,
> not a *substitute* for it. It is intended for all those potential
> users worldwide whose institutions cannot afford access to the
> toll-access version. The publisher's PDF (or XML) version remains
> the official version of record and should always be linked from
> your self-archived supplement."
> I do find this position perplexing. Stevan is fond of saying that
> self-archiving of the authors final version is a complete solution,
> addressing all concerns that there could possibly be about access to the
> literature. Nothing more is necessary, in his view. But he then follows
> that by, effectively, saying that access to the self-archived copy is a
> less-than-adequate replacement for the toll-access version, and gives
> second-class access to those second-class citizens of the scientific
> community who can't afford access to the official version of record.

Who is giving second-class status to what? All I said was: Self-archive the
supplement and link to the publisher's version.

> You can't have your cake and eat it:

The issue is the welfare of the baby, remember? All those would-be users,
currently lacking access to 75-85% of annual research output. The 50-250%
access gap.

> Either
> (a) the OA author's copy is a fully adequate, equally good substitute
> for the official version.
> In which case, as long as effective discovery is in place, there is
> no reason to subscribe to a toll access journal.
> or
> (b) the OA author's copy is better than nothing, but not as good as
> having access to the official version.

Matt, I am not in the journal business, either OA or TA. All I said was
self-archive and link to the publisher's version. I didn't tell anyone
to subscribe to or cancel anything. I'll leave you to reflect on how
much better the world's research community would be with the 50-250%
access gap closed by access to the author's version.

> If (a) is true, then ultimately subscriptions will surely decline,
> and if open access is to continue, publishers will need to evolve towards
> a mode of cost recovery other than toll access.

Matt is now joining ALPSP in speculating, which is fine with me, but
my concern is the baby (remember), which is currently suffering from a
50-250% access gap, which self-archiving can and will cure. And RCUK
is proposing to cure it. And ALPSP is opposing it, saying it will
make subscriptions decline. And there is no evidence to date for
subscription decline. And Matt...?

> If (b) is true, then OA self-archiving of the author's final version,
> while a hugely important step forward, is not an optimal, unimprovable
> solution. If full open access to the official version is possible,
> then that is clearly preferable and would better serve the needs of the
> research community.

Matt, everything on earth can be improved upon. I am focusing on
closing the 50-250% basic access-gap (for the baby, remember?) and you
are looking at maybe a 1% improvement on top of that, and repeating
BMC's 2-year old attempt to reserve the "OA" "brand" ("full open access")
only for its product:

    "Free Access vs. Open Access" (August 2003)

It's fine. Let's agree. 100% Self-Archiving is "a hugely important step
forward." Let's take it. Let's express unequivocal support for it (especially
when the RCUK resolve is still wavering in the balance under the ALPSP attempt
to delay or derail it). This is not the time to fuss (with Marie Antoinette)
about the icing on the cake, when there is still a 50-250% gap in access to the
basic bread!

> BioMed Central, along with a rapidly growing number of other open
> access publishers, is clearly demonstrating that full access to the
> official version *is* possible, and is no less economically viable than
> a toll-access model.
> Matt Cockerill
> Director of Operations, BioMed Central Ltd.

That's splendid, hugely important. But it is too slow! (Rapidly
growing? You show me a way to make UK OA (sic) grow faster than the RCUK
mandate will make it grow and you've got another customer!)

There is no way to require OA publishing to accelerate. It may never get
there. And meanwhile, researchers continue to lose their daily, weekly,
monthly access, usage and impact, needlessly, and 100% correctably. This
is not the time to divert attention to the respects in which 100%
self-archiving of author's drafts would still be "suboptimal"! Tell
me about that when we are safely, unstoppably speeding on our way there;
not now, when there is still a hugely important access-gap (50-250%),
for most researchers, to much of research!

Solomon would say: if you really care about the welfare of the baby, forget
about the cake-icing and support the bid for the basic bread.


Your Faithful Archivangelist,

Stevan Harnad

A complete Hypermail archive of the ongoing discussion of providing
open access to the peer-reviewed research literature online (1998-2005)
is available at:
        To join or leave the Forum or change your subscription address:
        Post discussion to:

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:

    BOAI-1 ("green"): Publish your article in a suitable toll-access journal
    BOAI-2 ("gold"): Publish your article in a open-access journal if/when
            a suitable one exists.
    in BOTH cases self-archive a supplementary version of your article
            in your institutional repository.
Received on Wed Aug 24 2005 - 21:37:51 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:59 GMT