The Green and Gold Roads to Open Access
harnad at ecs.soton.ac.uk
Mon Sep 20 09:32:06 EST 2004
The Green and Gold Roads to Open Access
> The Challenge. There exists a rapidly expanding stock of scientific
> knowledge. Yet, access to this pool of knowledge is often difficult.
The challenge is not about scientific knowledge in general. Much of
scientific knowledge is in royalty-based books and textbooks (and human
heads), and this is not about access to books or textbooks (or human
heads)! It is only about online access to articles in peer-reviewed journals.
Nor is is only about science; there are journals in all the scholarly
disciplines. It is about all articles in all peer-reviewed research
journals (of which there are about 24,000, across disciplines and
around the world, publishing about 2.5 million articles per year).
> A primary reason for this is the relatively high price of scholarly
> journals, their printed and their web-based versions.
It is even more general than that: The problem is that there is *any*
price at all charged for access to this special literature, which is
written purely for research usage and impact, and given away by its
The access/impact problem would still be there if journals were sold
at-cost, zero mark-up.
> It can be argued that this situation is both inequitable and
> inefficient, because:
> 1. Scientific research is often publicly funded (from public
> revenue)? and hence there is an expectation that the outcome of this
> research should become an, often global, public good;
Research-funders (not just science-funders) mandate that the outcome of
the funded research must be made public, through publication: "Publish or
Perish." In the online age, this very natural condition on the receipt of
research funding needs to be extended to: Publish and Provide Open Access
to your Publication (by self-archiving it, for free for all, on the Web).
But it is not only funded research that has the lost access/impact problem.
Even unfunded scholars write articles so that their work should be read,
used, cited and built upon. And most scholars careers depend on the uptake
of their work.
> 2. Knowledge is nonrival, that is, the costs of sharing it, once it
> is available, are zero or very low;
This principle is irrelevant because it applies indifferently to (1)
books/textbooks -- which are forms of knowledge that authors do not
necessarily always want to give away -- and to (2) peer-reviewed journal
articles, which they do always want to give away.
> 3. While some more applied knowledge is sometimes context-specific,
> basic scientific knowledge often is of wider, global applicability,
> a fact that makes restricted access to it particularly inefficient;
This too is an irrelevant distinction. The only relevant distinction
is between documents that the author writes in order to *sell* (for
salary, fees, or royalty) and documents that the author writes in order to
*give away* (to maximise their uptake and usage).
Both pure and applied research, if it is published in peer-reviewed
journals, is written for uptake and usage, and hence can and should be
made Open Access (OA).
> 4. Today's scientists "stand on the shoulders of" earlier scientists
> and the results of the publicly available work of these scholars;
True, but again irrelevant. What has changed is not the cumulative, public
nature of science and scholarship, but the advent of the Internet, and
hence the possibility of making peer-reviewed journal articles accessible
online for free for all would-be users.
> 5. Those, who do not find access to existing knowledge, are often
> those who might benefit most from it, viz. the poor of this world
> whose problems are often being neglected by mainstream research? yet
> this at a high cost for all, including richer countries and their
> research communities;
Most articles in peer-reviewed journals are not of interest to, nor are
they written for, the general public. They are written to be used and
built upon by other researchers, peers, in advancing science and scholarship
-- for the benefit of the general public.
It is a strategic mistake to make the case for OA in general only or
mainly on the basis of that small portion of journal articles that
might be of direct interest to the general public (such as certain
research findings in clinical biomedical research). The case for OA is
far more general than that, and it is mostly about peer-to-peer access
from researchers for researchers, worldwide. Public access follows from
that quite naturally too, but if accessibility to the general public
were the sole or principal reason for OA, then OA would be sought for only
a very small portion of the scientific and scholarly journal literature.
> So how can the current access problems be overcome? Open-access
> Scholarly Journals Initiatives have been undertaken to demonstrate
> that scientific knowledge need not necessarily be published in forms
> that make access expensive, or even impossible. It could be provided
> free of charge, through open access to it, without detrimental effect
> on scientific knowledge production and preserving the peer-review
> process that is key to validate scientific results.
But fewer than 5% of journals (c. 1200) are
"Open-access Scholarly [OA] Journals,"
and OA is not merely, or primarily, about trying to persuade the remaining
95% of journals to become OA journals, and to adopt the OA cost-recovery
model (author-institution publication charges instead of user-institution
subscription charges). If that "golden" road to OA were the *only*
road to OA, OA would still be very far away, and might never arrive.
There exists a second, far wider, faster, and surer road to OA: the
"green" road of author/institution self-archiving of articles published
in non-OA journals. Over 90% of journals have already given their green
light to author self-archiving, http://romeo.eprints.org/stats.php but
only 10-20% of authors have as yet self-archived.
The solution is simple: Although research-funders and
researchers' institutions cannot mandate that journals
convert to gold, they can mandate that their researchers
self-archive. Such a self-archiving mandate has already been
recommended by the UK
and the US
and is currently under consideration in a number of other countries as well
Swan & Brown's (2004) JISC/OSI author survey reported asked authors
"how they would feel if their employer or funding body required
them to deposit copies of their published articles in... [OA
archives]. The vast majority... said they would do so willingly..."
Swan, A. & Brown, S.N. (2004) JISC/OSI Journal Authors Survey
Swan, A. & Brown, S.N. (2004) Authors and open access
publishing. Learned Publishing 2004:17(3) 219-224.
So there are two roads to OA, gold and green one, and the green road
is the fastest and surest, and the one that is already fully within reach.
The GPG Forum questions here unfortunately dwell nstead almost exclusively
on the gold road, and treat it as if OA = OA publishing. It does not. And
this misconception is now the single biggest retardant to the growth
Harnad, S., Brody, T., Vallieres, F., Carr, L., Hitchcock, S.,
Gingras, Y, Oppenheim, C., Stamerjohanns, H., & Hilf, E. (2004)
The green and the gold roads to Open Access. Nature Web Focus.
The GPG Forum should focus on the two roads to OA in proportion to their
actual size as well as their immediate potential for growth. On both
counts, it is today the green road that calls for (and repays) most of our
immediate attention and effort.
> With open access, fees to meet the publishing costs - when required
> - are paid up front when articles are accepted by a journal, rather
> than by the readers. Access to the journal is then provided for free.
This is confusing OA itself with OA publishing (gold).
It is not with OA itself that costs are covered by author-institutions
instead of reader-institutions, but only with the OA publishing
cost-recovery model (the golden road to OA).
With the green road of self-archiving (of non-OA journal articles),
nothing changes except the access (and impact). And it is access to the
*article* that is provided for free (by its authors/institution).
The current status quo must always be kept in mind: 5% of journals are
OA and 95% of journals are non-OA, and it is not easy to create/convert
new OA journals.
> In most cases, the authors do not cover the publishing fees
> directly, since they might be able to charge the fee to the budgets
> of their research projects so that, ultimately, research sponsors,
> not the authors, pay for publishing. Today, about 5% of academic
> publishing follows the open-access model. But the model is quickly
> gaining ground, including among both for-profit (BioMedCentral -BMC)
> and not-for-profit (Public Library of Science - PloS) publishers.
Unfortunately this is incorrect. The creation/conversion of gold journals
is *not* quickly gaining ground. At 5% it is gaining ground extremely
slowly -- and far more slowly than the creation/conversion of journals
to green, which is already over 90%.
Likewise faster than the growth of gold (but still not nearly fast enough)
is the growth of the actual self-archiving by authors for which the
green journals have already given their green light. That is at about
10-20%. But that is what the proposed UK and US self-archiving mandates
are intended to accelerate.
> We can deliver open access through archives or repositories,
> not just through journals. OA repositories do not perform peer
> review, although they can host and disseminate works that have been
> peer-reviewed elsewhere. They can be organized by discipline or by
> institution. They can contain preprints or postprints or both. They
> are built on open-source software and very inexpensive to launch and
At last the green road to OA is mentioned, though a bit obscurely: Of
course OA archives don't perform peer review! They are archives, not
journals; they are there to provide OA to the articles in peer-reviewed
journals, self-archived by their authors, both before and after peer
review (by the journals). Journals provide peer review, not archives.
Archives provide access (OA).
> For example, they do not charge processing fees, as many
> OA journals do.
Of course OA archives don't charge processing fees! They are archives,
not journals, there to provide access to peer-reviewed articles, not
to peer review them or process them. Journals process articles, not archives.
Archives provide access (OA).
> Repositories can provide OA even to works published in conventional journals.
Provide OA "even to"? But providing OA to works published in non-OA journals
is OA archives' *main* function and raison d'etre !
Just as the gold and green roads should not be mixed up, and OA should
not be equated only or mainly with the golden road, so there may be many
reasons why institutions may wish to have online repositories, but let's
not mix up all those other reasons with the only one under discussion here:
OA provision for peer-reviewed journal articles!
> In fact, more than 80% of conventional journals already allow their
> authors to deposit their postprints in OA repositories.
To be exact, 91.31% of the 8860 journals surveyed to date (and that includes
just about all the core journals) are green, of which 69.2% are postprint green and
22.29% more are preprint green. That leaves 8.51% gray (no green light
for wither preprint or postprint yet).
> Most OA repositories are interoperable in the sense
> that users can search them without knowing which repositories exist,
> where they are located, or what they contain.
That is because they are compliant with the OAI metadata-tagging standard
which has made distributed institutional self-archiving possible:
This means that peer-reviewed journal articles can be openly accessed, searched
and retrieved from distributed archives as if they were all in one global virtual
> Questions for Debate
> What are the main pros and cons of open-access scholarly publishing?
Why are we asking only about the pro's/con's of OA publishing (gold),
rather than those OA itself (both gold and green)?
The pro of gold is that it provides OA; the con is (1) that the cost-recovery
model has not yet been tested long enough to know whether it is sustainable
(2) only 5% of journals are gold, and (3) converting the remaining 95% would
take far too long, if it was possible at all.
The pro of green is that it provides OA and over 90% of journals are
already green; the con is that although self-archiving is growing faster
than gold, it is still not growing fast enough (currently 10-20%).
The pro of OA (green plus gold) is that OA maximizes research access
and impact. There are no cons, although at least 31 groundless
worries have been delaying this optimal and inevitable outcome
for over a decade:
> Thinking in particular of scholars in developing countries (and the
> fact that research grants may not be as easily available for them than
> for industrial-country scholars), could they face a new disadvantage?
They certainly face a disadvantage if we keep on implying that OA =
OA-publishing (gold), with its associated publishing cost. But the green
road to OA is free, and developing countries can not only take that road,
but lead the developing countries along it!
"Access-Denial, Impact-Denial and the Developing and Developed World"
> What sources will be available to pay these fees when authors cannot
> get their funder or employer to pay them?
This is all a red herring, because OA does not equal OA-publishing. It
would be a big pity if we spent another decade waiting for and worrying
about gold instead of taking the green road of self-archiving that is
waiting, wide-open, for us.
> Will all open-access journals be able to waive processing fees in
> cases of economic hardship, as PLoS and BMC do?
Will this Forum spend all its time on 5% issues like this, when
95% of the literature still awaits OA?
> Should the international aid community maintain a fund/facility to
> help meet these costs?
What the international community should do is mandate self-archiving
of the articles in the 90% green journals, not fret about funding for
the publication charges of the 5% gold journals!
> Is the open-access model of publishing more likely to be successful
> in some than in other fields? What would determine the likely success?
Let someone do a research project on the disciplinary distribution and
hypothetical possibilities of the 5% of journals that are gold today. But
let all other researchers go ahead and self-archive their research to
make it OA by following the green road today.
> Could the open-access model of knowledge management
> be applied beyond scholarly academic publishing?
Knowledge management? Is the question: "Is an author-end cost-recovery
model applicable to anything else besides journal articles?"
Esoteric monographs, maybe. But let's wait to see how it does
with journals; it's only been around for a few years...
> Further Information The Budapest Open Access Initiative
> [http://www. soros. org/ openaccess] defines open access as: "free
> availability on the internet, permitting users to read, download,
> copy, distribute, print, search, or link to the full texts of these
> articles, crawl them for indexing, pass them as data to software,
> or use them for any other lawful purpose, without financial, legal,
> or technical barriers other than those inseparable from gaining
> access to the internet itself. The only constraint on reproduction
> and distribution, and the only role for copyright in this domain,
> should be to give authors control over the integrity of their work
> and the right to be properly acknowledged and cited."
Maybe this will be easier to grasp and remember:
Immediate, permanent, online, full-text access, toll-free, webwide.
More information about the Jrnlnote