Re: The Green and Gold Roads to Open Access

From: Stevan Harnad <>
Date: Thu, 7 Oct 2004 15:57:46 +0100

Brian Simboli's points, below, have already been discussed many times
in this Forum, but for those for whom the token just might at last drop
this time, I will try again, from a slightly different angle:

On Thu, 7 Oct 2004, Brian Simboli wrote:

> I think the overlay journal concept is much more within practical reach
> than people realize. Why is there this unspoken assumption that green is
> any more practicable than, say, the overlay concept?
> Witness:

Overlay journals such as the journal above are merely an efficient new way
of implementing conventional journals online. The idea is that instead
of submitting your manuscript to the journal's website, you deposit
it on your own website and just send the journal the URL. The editor
looks it over, and if it is suitable for refereeing, sends the referees
the URL. If it successfully passes peer review (the usual way), it is
accepted and published -- which in this case merely means adding the
journals "accepted, peer-reviewed, published" certification tag to the
final accepted, revised, peer-reviewed postprint.

Now this overlay method has nothing to do with whether or not the journal
is an Open Access (OA, gold) journal. A Toll Access (TA) journal could
implement peer-review this way too, and some of the American Physical
Society (APS, green) journals have been doing this, though they also
generate an APS-style edited PDF at the end, which is also archived
in the APS archives, and printed as the print edition.

But overlay journals have nothing in principle to do with OA -- though
of the few that exist, most are OA journals (gold).

And self-archiving (green) is about providing OA to the articles
published in the 95% of journals that are TA. That means self-archiving
the peer-reviewed postprints, not just the unrefereed preprints. For the
postprints, there is nothing for a journal to "overlay" on -- or rather,
the "overlay" is already there, in the form of the metadata tag naming
the journal in which the article was published.

So Brian Simboli's cavalier suggestion that "overlay journals" are more within
reach than (green) self-archiving is merely another one of those "let them eat
cake" suggestions to researchers (hungry for access/impact), and in fact merely
a variant on "Waiting for Gold":

(The concept of "overlay journals" should not be confused with other
hypothetical or barely tested proposals, such as replacing in-advance
peer review with post-hoc opinion-polling, or with "multiple peer
review" of the same article by many journals [a particularly profligate
and unrealistic suggestion, considering how hard it is and how long it
takes to get even one qualified expert to referee a paper even once!],
or other post-hoc performance indicators such as "Faculty of 1000",
citations, or commentaries, which are really supplements to peer review
rather than substitutes for it.)

> Why not devote precious dollars to this? Or the brunt of the dollars?

Why not devote no dollars, and merely self-archive the articles
that are already published in the journals that already exist? Why
is counterfactual armchair speculation more gratifying than concrete,
within-reach action? (But this question should of course be directed
to the content-providers themselves, the researchers, who are also
the beneficiaries of OA, not just at the well-meaning but entirely
misguided librarians who are trying to guide them!)

> Also, I am told that has been willing to expand its subject
> coverage. Why not use that as a repository for final, refereed versions
> articles?

Arxiv happens to be a central OAI-compliant OA archive. There are also
many other distributed institutional OAI-compliant OA archives. All
of them are open to all embryological stages of research papers, from
the unrefereed preprint to the per-reviewed postprint and beyond. It is
journals that provide the dynamic, interactive, answerable peer-review
service in between whose outcome (when successful) is then certified by
the journal's name -- its quality-control tag and its track-record.

In other words, neither ArXiv nor any of the other OA Archives is or has ever been
just for unrefereed preprints, waiting for "overlay journals." They are, and
always have been, for both preprints and postprints, and OA in particular -- whose
objective is toll-free online access to the full-texts of all 2.5 million articles
published yearly in the world's 24,000 peer-reviewed journals -- is concerned
first and foremost with Open Access to the peer-reviewed, published drafts, not to
the unrefereed preprints.

Now Brian for some reason does not like the green road -- of
self-archiving one's final, refereed articles -- if it is done in
an institutional OAI archive; there he recommends "overlay journals"
instead. But he *does* like the green road -- of self-archiving one's
final, referees articles -- if it is done in a central OAI archive
like ArXiv.

Brian gives no reason for this piece of whimsy, but no matter, let us
take him at his word: He is *for* the green road if it is central, only
against it if it is institutional. (I'm guessing it's because he sees the
latter as competing for library funds and time, and he'd rather see all
that off-loaded somewhere else.)

Either way, Brian is in contradiction with himself: He is both for and against
green. The rest (in the words of the wag) is just haggling over the price.

A much more reflective and informed case has been made for in favour of
distributed, institution-based green self-archiving -- and was announced
in this Forum a few days ago:

    Delivery, Management and Access Model for E-prints and Open
    Access Journals within Further and Higher Education. Alma Swan,
    Paul Needham, Steve Probets, Adrienne Muir, Ann O'Brien,
    Charles Oppenheim, Rachel Hardy, and Fytton Rowland (2004).

> I'm not asking hypothetical questions. I think the questions are quite
> important for a manager at something less than a Harvard or MIT, whose
> dollars, and staff time, are already stretched. I know where I'd put the
> emphasis.

Offloading institutional self-archiving onto remote central archives
(always still green, if it's self-archiving at all) is fine in the few
fields where the central archives (like Arxiv, CogPrints, and PubMed
Central) exist, and there is someone else ready to foot all the centralised
cost, but it is futile where they central archives do not exist and/or no one
wants to foot all of the cost and responsibility (which is in most
fields). More important, it is useless merely to provide an archive, be
it central or institutional. The real challenge is getting the archive
filled. And in this, institutional archives have a great advantage
because they share with their own institutional authors the benefits
of maximizing access and impact to their institutional research output
(and they share also the costs of *not* maximizing it, when other,
competing institutions do).

Institutions, in other words, are in the position and have a vested
interest in mandating, monitoring and maximizing the usage and impact
of their own research output (OA), across all of their institutional
departments: Central archives and disciplines have no such interest,
or means.

Research-funders too have the interest and means to mandate OA
provision by their fundees, and they could mandate it institutionally
or centrally, but mandating it institutionally has a far greater
probability of propagating the effects of the mandate -- the practice
of self-archiving -- beyond just the particular research projects and
fields the funder is funding, to all of each fundee's institution's
departments and disciplines, and to competing institutions as well
(researchers/institutions compete with rival researchers/institutions
for research impact).

That is why CogPrints, a central archive I started in 1998, is still
growing so slowly
and why even the biggest and oldest of central OA archives,
ArXiv is still growing only linearly, after almost a decade and a half
whereas distributed institutional are poised to grow exponentially
as soon as institutions and research-funders mandate self-archiving:

> I also know from hard experience how quickly publishers can pull the
> plug on provisions not in their financial interest.

Librarians' acquisitions/pricing experience is simply steering them
false here. It would be so good if Brian were to have a look at and
have a think about this FAQ this time, rather than simply reiterating
his doomsday prophecy, reflexively, and unreflectively, as he keeps doing:

Stevan Harnad

A complete Hypermail archive of the ongoing discussion of providing
open access to the peer-reviewed research literature online (1998-2004)
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-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 Thu Oct 07 2004 - 15:57:46 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:37 GMT