Re: On the Need to Take Both Roads to Open Access

From: Stevan Harnad <>
Date: Fri, 5 Mar 2004 14:31:37 +0000

On Fri, 5 Mar 2004, Frederick Friend wrote:

> JISC in the UK is putting as much effort into supporting institutional
> repositories as it is into open access journals...
> The results of an author survey funded by JISC and OSI to be published
> on the JISC web-site next week show that a low percentage of authors
> have deposited pre-prints or post-prints in IRs, although a very high
> percentage would do so if required to do so by their funder or employer.

The low percentage was already known from the arithmetic on the number
of articles published annually

versus the number of articles self-archived annually

although this low percentage must not be misinterpreted as being lower
than the far lower percentage of articles published annually that
are published in open access journals!

The stress on the lowness of the self-archiving figure is made because
the *ceiling* on the number of articles that can immediately be made
open-access via OA self-archiving today (100%) is so much higher than
the ceiling (5%) on the number of articles that can immediately be made
open-access via OA journal publishing today -- for the simple reason
that fewer than 5% of journals are OA journals today.

The further finding of the JISC/OSI survey that "a very high percentage
would [self-archive] if required to do so by their funder or employer"
has also been anticipated for some time, and would seem to imply quite
strongly that the best way for JISC and OSI to support both OA journal
publishing and OA self-archiving is to direct their efforts now to
promoting mandatory OA-provision policies by research funders and
employers (via whichever of the two OA roads is suitable):

    Unified Dual Open-Access-Provision Policy:
    BOAI-2 ("gold"): Publish your article in a suitable open-access
            journal whenever one exists.
    BOAI-1 ("green"): Otherwise, publish your article in a suitable
            toll-access journal and also self-archive it.

The Declaration of Institutional Commitment to implementing such policies
will soon be released (and is already open for pre-signing); it is
to be hoped that JISC and OSI will support it vigorously, so both of those
percentages can be helped to rise:

> I am very concerned that the two branches of open access "do not go down
> separate tracks".

If both OA branches always explicitly promoted the unified OA Provision Policy,
they would not go down separate tracks. But Paul Gherman was not writing about
unifying BOAI-1 (OA self-archiving) and BOAI-2 (OA journal publishing)
(they are already unified, implicitly, if not always sufficiently
explicitly) but about unifying the OA movement and the Institutional
Archive ("Repository") movement.

I have replied to Paul's recommendation
trying to show how and why the 5 diverse and divergent means and ends
of Institutional Archives do *not* coalesce coherently in such a way
as to clarify or promote OA at this time; so it is in fact important
that they should *not* go down the same tracks. In particular, neither
the institutional agenda for the preservation of its digital holdings
(PRES) nor the institutional agenda for electronic publication (EPUB) --
2 of the 5 uses to which Institutional Archives could be put -- should
be conflated with the (unified) institutional agenda for mandated OA
provision for its researchers' journal article output (RES).

> [Separation of BOAI-1 and BOAI-2] has been happening because funding agencies
> do not perceive IRs to offer a sustainable high-quality service for the record
> of science.

Funding agencies had no perceptions about OA by either road, gold or green.
It was up to *us* to make the unified case for OA to them. Instead, there
has been a good deal of unilateral emphasis on BOAI-2 alone, and even the
little that was said about BOAI-1 was unclear and uninformed, hence misleading.
The relevant point to be made to research funding agencies is about
*research access provision* not about a "sustainable high-quality service
for the record of science"! The very phraseology -- sustainable high-quality
service for the record of science" -- is already massively tilted toward
BOAI-2 (OA publishing)! JISC and OSI should be promoting a unified, bilateral
perception of OA and not this one-sided and counterproductive one.

[We don't just *meet* perceptions: we unconsciously *make* them! Is it
not a coincidence that research funding agencies -- of all places --
should have come up with the librarians' characteristic "sustainability"
worry, when their concern with OA is and ought to be about how to increase
access to the research output they fund? It is not at all clear,
then, a priori, that it is the "record" that needs to be changed,
rather than merely the extent of *access* to it! All this can be stated
theory-neutrally. To start to speculate about sustainability and quality,
when the underlying issue is just access, is to start putting a lop-sided
load on the concept of Open Access that bends it in the direction of an
alternative publishing model -- when in reality OA means an alternative
*access* model, of which an alternative publishing model is merely one
of the two means of achieving it (and not even the fastest or surest
of them!).]

> The perception is [1] that IRs will not hold the best version of a
> research article, [2] that it will not be indexed as comprehensively as a
> traditional journal article, and [3] that its long-term preservation is not
> secure.

How interesting that the survey should be echoing back the very same 3
concerns that the library community has been voicing for several years
now! One possibility, of course, is that the library community was quite
right about this, and the funding community is simply drawing the same
correct conclusions. But the other possibility is that the case is being
unwittingly put to the funding community (which had not previously had any
view at all on the matter, not having given research access any thought
at all) in such a way as to create this very perception -- rather than
the much more neutral and balanced one that it would and could have had,
if the proposition put to them had been purely about *adding* more access
to existing access, rather than about an alternative publishing model!

For, seen from this unbiassed perspective, the view, to the funding
community, would look more like this: There is the present access
situation, TA, with the degree of access that it provides for the articles
reporting the research we have funded (accessible only to researchers
who are at those institutions that can afford the access tolls); and
then there is the alternative, OA, in which that degree of access is
increased, either because the author publishes the article in an OA
journal, or because the author publishes the article in a TA journal
and also makes it OA by self-archiving it.

(Does this unbiassed way of putting the proposition make it a little
more obvious that the concern that authors may self-archive the "wrong"
version is not objectively a greater worry than that they will not make
it OA at all, because there exists no OA journal to publish it in? What
are the respective percentages of potential access, over and above current
access levels, that would be lost owing to these two risks, do you think?)

So much for 1. If the proposition had been put clearly and neutrally, as
only about enhancing access-provision so as to provide 100% open-access,
it would be evident to the funders that we are talking about present-day
TA publishing *plus* whatever further access is provided by either OA
publishing or OA self-archiving, not present-day publishing *minus*
something, such as the "best version" of the article!

The exact same reasoning would apply, quite transparently, to 2 (indexing):
We have all the indexing of the present TA system *plus* whatever further
access is provided by either OA publishing or OA self-archiving, not
present-day publishing *minus* something, such as the "comprehensive
indexing" of the article! (And this is without even mentioning that
a world in which all full-texts of all articles are digital and OA is one
in which powerful online forms of indexing will trump existing ones
just as google is trumping existing forms of search!) But I rather doubt
that worries about indexing would enter the minds of either funders or
researchers in a neutral query about the desirability and feasibility
of OA provision unless they are implicitly planted there by the way we
put the proposition to them.

This is perhaps the clearest with 3 ("long-term preservation"),
the most irrational worry of all, and so obviously emanating from the
library community's own perception, rather than being a reflection of
the perception of the funding or research community: For, again, if the
OA proposition were put neutrally, as a question about maximising access
and impact, rather than as a question about changing publishing models,
what new preservation worries does it raise?

Are we worried about the preservation of the existing TA journal article
corpus, whether online or on paper? Well, fine, let us worry, and let us
work on its reservation, but *what on earth does that have to do with OA?*

Now let us talk about access: If access to the existing TA journal article
corpus is enhanced by authors supplementing the TA journal versions of
their articles by self-archiving an OA version, has that introduced a new
preservation problem? Or just enhanced access, as promised? (And even if
one is bent on worrying about how long that *extra* supplementary version
will remain accessible, it would seem that the facts about the 13 years
of self-archiving by physicists, with the 1991 OA versions still going
strong in 2004, might be cited in a value-neutral way of putting the
OA proposition to funders and researchers as evidence *against* worries
about preservation even for this secondary, supplementary version, rather
than as grounds for a blanket "preservation" worry about BOAI-1!)

I am not saying that these misperceptions all (or even mostly) arose
because of bias in the way the questions in the survey were put. I
have not seen the way the questions were put, and no doubt these
prima facie misperceptions occur quite naturally to those who have no
experience with these matters, or have not reflected on and read about
them. Each of the three (as well as the earlier one, about sustainability)
has arisen spontaneously in our own prior surveys, and in general crop
up often enough everywhere to have generated their own entries in the
self-archiving FAQ:

But even if everyone is prone toward these prima facie misperceptions
(about both BOAI-1 and BOAI-2) at first blush, surely JISC's and OSI's
mission should be, having polled the perceptions, to work to correct
them -- especially by educating the funding and research community about
feasibility and benefits of implementing a systematic institutional
open-access provision policy.

> One leading researcher described IRs to me as "anarchic", whereas
> journals - whether subscription or OA - are perceived to be organised,
> reliable and secure.

Self-archiving (BOAI-1) *is* anarchic! But if portrayed in an
unbiassed way, it is a means of *supplementing* TA journal access, not a
*substitute* for it! Portraying self-archiving (or taking at face value
the misunderstanding of self-archiving) as somehow replacing either TA or
OA journals is simply an error, and should be pointed out as an
error, rather than a basis for a rational worry about or objection to

> Whether these perceptions are true or not, we have to
> raise the status of IRs as part of the record of science.

Not IRs (they are many things to many people) but BOAI-1 -- one of the
two means of providing OA to journal articles. And not as part of the
"record of science" (with its attendant preservation burden), but as a
supplementary means of providing access to the articles (whose primary
record resides in the journal that publishes them).

Let us work on correcting perceptions rather than confirming them.

Stevan Harnad

NOTE: A complete archive of the ongoing discussion of providing open
access to the peer-reviewed research literature online (1998-2004)
is available at the American Scientist Open Access Forum:
        To join the Forum:
        Post discussion to:
        Hypermail Archive:

Unified Dual Open-Access-Provision Policy:
    BOAI-2 ("gold"): Publish your article in a suitable open-access
            journal whenever one exists.
    BOAI-1 ("green"): Otherwise, publish your article in a suitable
            toll-access journal and also self-archive it.
Received on Fri Mar 05 2004 - 14:31:37 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:22 GMT