Re: The Green and Gold Roads to Open Access

From: Stevan Harnad <>
Date: Sat, 14 Feb 2004 18:35:37 +0000

Suhail A. Rahman wrote:

> 1. We have a problem with accessing the scientific literature due to
> access tolls that make it unreachable for all

On this we all seem to agree. Let us call it the Immediate Access Problem.

> 2. To alleviate this problem the OA initiative says there are two
> roads: self archiving & OA journals. Stevan Harnad strongly believes
> in self archiving

Correct. But it needs to be emphasized that my "belief" in and advocacy
of OA self-archiving are based on solid evidence and reasons, not on
faith or disposition: The Immediate Access Problem is a pressing present
problem (for those who lack the access), not a leisurely future one.

There are fewer than 1000 (5%) OA journals at present
(compared to about 23,000 TA journals: 95%), and, as Suhail correctly
points out, not all authors can afford to pay the publication charges
even for publishing in those 5% OA journals..

Hence the OA Journal ("Golden") Road to OA is too small, slow
and uncertain as the *sole* solution to the Immediate Access
Problem. Fortunately, there is also another, parallel Road, a larger,
faster and surer one, the ("Green") Road to OA: that of continuing to
publish one's articles in conventional TA journals but also providing
OA to those same articles by self-archiving them.

OA self-archiving has been practised and tested for far longer than OA
journal-publishing (since at least 1990), it has been proven to provide
reliable, lasting OA; it has successfully done so for at least three
times as many articles per year as are being published per year in
OA journals, and it is growing faster than OA journal-publishing (5%)
is growing, even receiving an official green light from at least 55%
of journals.

Most important, OA self-archiving, despite the fact that it
already provides at least three times as much annual OA today as OA
journal-publishing, and despite the fact that it is growing faster, is --
*relative to its true, full, immediate potential to provide OA* -- the
far more under-utilized of the two Roads to OA! For it already has the
capacity to provide immediate OA to 100% of articles, virtually overnight.

And the two reasons why this immediate 100% OA through self-archiving is
not yet being provided by researchers are (1) that far too few researchers
and (2) far too few research institutions have understood how and why
to do it. Suhail himself is one of those researchers who has not
understood self-archiving, and he gives ample evidence of this failure
to understand it below:

> 3. I embraced OA journals warmly in the early 2000's but ignore self
> archiving for two reasons. a) It isn't needed if OA journals gain
> popular support. b) We can meanwhile start by self archiving our
> articles at home

(a) and (b) are not only incorrect, but incoherent:

(a) With only 5% of journals OA, it makes absolutely no sense to just
sit and wait for "OA Journals to gain popular support" -- if we do indeed
agree on the premise (that there is an Immediate Access Problem) -- when
the option of providing immediate access to our own research is within
our own hands.

(b) What on earth is the difference between self-archiving and
"self-archiving articles at home"? Self-archiving one's own TA articles
on one's own institutional website *is* self-archiving!

> 4. I find out that OA journals are not really toll free, they offer
> different kinds of tolls. I am also afraid that these tolls may lead
> to a publishing hegemony in the future.

Fine. Then that sounds like yet another reason *not* to sit around waiting
for "OA Journals to gain popular support" in order to solve the Immediate
Access Problem but to self-archive instead, right now!

> 5. I argue for a different kind of OA initiative whereby existing
> journals find a mechanism to offer OA without the necessity for
> author tolls.

Now we are asked to sit around and wait, not for the *actual* OA
journals to provide OA, few as they are (5%), because they cover their
costs out of author charges that are unaffordable for some authors,
but for hypothetical *alternative* OA journals that cover their costs
in some hypothetical alternative way (0%). This 2nd-order waiting for an
unspecified and hypothetical alternative -- in preference to doing the
immediately doable -- does not sound like a very rational way to go
about solving the Immediate Access Problem.

> 6. I give give one possible mechanism, which is embargoed TA for 1
> year plus author distribution of embargoed papers (already being done
> by JCEM).

While sitting around waiting for the hypothetical alternative OA journals
that will cover their costs in some (unspecified) hypothetical alternative
way, we are also to sit around waiting for the solution to the *Immediate*
Access Problem in the form of *Delayed* Access (itself likewise almost
virtually nonexistent at this time) with authors meanwhile emailing
their articles one by one to those who email to ask. (This sounds like
a needlessly and arbitrarily gerrymandered version of self-archiving, where
the web is put to its natural use!)

Do all that waiting, but on no account self-archive your own articles
in the meanwhile, not even the 55% of them that are published in journals
that have given it the official green light! (Is it becoming a little
clearer why I find Suhail's logic so hard to follow?)

> 7. Stevan finds this unacceptable, as self archiving is ignored entirely
> (as we know it)

But I have made it crystal clear precisely *why* and *how* these
recommendations are not only unacceptable but arbitrary and incoherent!

> 8. I defend ignoring self archiving because I still do not at all
> understand how it would fit into my scheme of scientific publishing
> without tolls at all - Neither author nor user.

But the objective is *not* to "fit into Suhail's (unspecified) scheme of
scientific publishing without tolls at all -- neither author nor user"!

The objective is to solve the Immediate Access Problem! Immediately.
Rejecting OA Journal Publishing (because some authors cannot afford the
author charges) *and* ignoring or rejecting self-archiving is basically
offering *nothing* in place of *something* -- and for no reason at all.

> 9. Stevan takes the road that self archiving and OA publishing are two
> independent things and hence we have no right to ignore self archiving.

This is not about rights. You have the right to do or not do, say or
not say, whatever you like. But if you make public recommendations, you
have to be prepared to support your reasoning, and your reasoning is so
incoherent here that even on the most charitable assumption one simply
cannot make any sense of it.

Based on the *only* substantive point you make -- which is that some
authors cannot afford to provide OA to their articles by publishing them
in an OA journal at this time -- the only rational conclusion is that
then those authors should not provide OA to their articles by publishing
them in an OA journal at this time! They should instead publish them in
a TA journal and provide OA to them by self-archiving them.

To ignore that obvious alternative is tantamount to ignoring the Immediate
Access Problem, the very premise that we supposedly share. If we do not
share that premise, then simply say: "I don't think there is an Immediate
Access Problem" -- and then be prepared to defend *that*. But instead
affirming the Access Problem, denying the OA Journal solution (because
of unaffordability) *and* ignoring or rejecting the OA self-archiving
solution simply adds up to one thing: self-contradiction.

> 10. I disagree simply because a prerequisite for self archiving is
> publication in the first instance (unless Stevan is implying self
> archiving prior to publication).

*Of course* a prerequisite of self-archiving is publication in
the first instance! Self-archiving is the self-archiving of one's
own *TA journal articles*! I have referred Suhail several times in
these exchanges to the explicit definitions in the self-archiving FAQ
but unfortunately Suhail does not seem to have consulted the clear,
simple definitions therein.

And, yes, an additional bonus of self-archiving is that one can (if one
wishes) self-archive all stages of one's article, both pre-peer-review
preprints and post-peer-review (revised, accepted, published)
postprints. But the preprints are merely an optional bonus. The raison
d'etre of OA self-archiving is to solve the Immediate Access Problem,
and that (to repeat) is: Immediate Toll-Free Access to the 2,500,000
articles published in the world's 24,000 peer-reviewed journals (not
merely, or primarily, to the unrefereed preprints of those articles!):

    What is an Eprint?

    "Eprints are the digital texts of peer-reviewed research articles,
    before and after refereeing. Before refereeing and publication,
    the draft is called a "preprint." The refereed, published final
    draft is called a "postprint." Eprints include both preprints and
    postprints (as well as any significant drafts in between, and any
    postpublication updates). Researchers are encouraged to self-archive
    them all. The OAI tags keep track of all versions."

    What should be self-archived?

    "All significant stages of one's work, from the pre-refereeing preprint
    to the peer-reviewed, published postprint, to postpublication updates
    should be self-archived. The OAI tags keep track of all versions."

> 11. I conclude by asking Stevan if he believes that conventional self
> archiving is either required or even viable in the *presence* of author
> tolled OA journals, and if so, why?

Suhail keeps causing confusion by systematically misusing the word
"tolls," which was coined to designate *access* tolls and hence the OA/TA
distinction itself. With his terminology, OA journals are TA journals,
because he insists on calling the author publication charges "tolls"
too, thereby making OA/TA distinction incoherent (to no useful purpose).

(Maybe we should have called OA "FIPATRAFTO" after all! and TA
"non-FIPATRAFTO" to prevent this sort of word-play! ).

Never mind. Call them what you wish, the answer to your question, in
longhand, is this:

Yes, of course self-archiving (what on earth is "conventional
self-archiving"?) of (conventional) toll-access journal articles by
their authors in order to provide toll-free access to those articles
for all their would-be users is viable in the (parallel) presence of
OA journals (i.e., journals that provide toll-free access to all their
articles to all users by recovering costs from the author/institution):

Those authors for whose articles one of the existing 1000 OA journals
(5%) is suitable, and who can afford the publication charge, should
provide toll-free access (aka: OA) to their article by publishing it in
that OA journal. (This is called BOAI OA Strategy 2, aka: taking the
golden road to OA.)

Those authors for whose articles none of the existing 1000 OA journals
(5%) is suitable, or who cannot afford the OA journal publication charge,
should publish in a conventional TA journal (95%) and provide OA to their
article by self-archiving it. (This is called BOAI Strategy 1, aka: taking the
green road to OA.)

Jointly, these perfectly compatible and complementary measures provide
a unified solution to the Immediate Access Problem.

    "The Green and Gold Roads to Open Access"

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 Sat Feb 14 2004 - 18:35:37 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:21 GMT