Re: The Green and Gold Roads to Open Access

From: David Goodman <>
Date: Mon, 22 Mar 2004 13:00:39 +0000

The use of codes in your reply is _exactly_ what I am protesting about.

  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.

All this was fine when we were speculating remote from the mainstream, but not when we are reaching the point of public acceptance. That your definitions are not generally understood or accepted is demonstrated by the previous 2 weeks worth of confused discussion about what is and what isn't open access.

David Goodman

-----Original Message-----
From: Stevan Harnad []
Sent: Sun 3/21/2004 9:26 PM
Subject: Re: The Green and Gold Roads to Open Access

On Sun, 21 Mar 2004, David Goodman wrote:

> In our discussions of OA, I feel there is a need for better terminology
> to distinguish between the arXiv-like database or repository model,
> in any of its modifications, and the two types of journals those paid
> at the reader end, and those paid for at the journal-production end.

Open Access (OA) can and should be defined at the article level, not
at the journal level. An article is OA if its online full-text can
be immediately and permanently accessed (downloaded, stored, printed,
processed) toll-free by anyone webwide. That is the BOAI definition of

    "The literature that should be freely accessible online is that which
    scholars give to the world without expectation of payment. Primarily,
    this category encompasses their peer-reviewed journal articles, but
    it also includes any unreviewed preprints that they might wish to
    put online for comment or to alert colleagues to important research
    findings. There are many degrees and kinds of wider and easier access
    to this literature. By "open access" to this literature, we mean
    its free availability on the public internet, permitting any 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 BOAI also lists the two ways of providing OA for articles:

BOAI-1 is the self-archiving of articles published in Toll Access (TA)
journals in OA Eprint Archives (whether central discipline-based Eprint
Archives or distributed institutional Eprint Archives, but preferably
OAI-compliant ones, so that they are interoperable and harvestable by
OAI services such as OAIster:

BOAI-2 is the publishing of articles in OA Journals that do not cover their
costs by charging users access-tolls.

Implicit in all this is the distinction between and OA an TA article, and
between an OA and TA journal. (The distinction between and OAI-compliant and
a non-OAI-compliant archive is functionally important, but not essential to the
definition of OA; the distinction between central and institutional archives
is also not essential, though there are reasons why institutional archiving
is a more promising way to scale OA up to 100%)

> For journal-production-end journals, I particularly dislike author-paid,
> as it is not the intention of any of the proponents that the author
> personally will pay for the publication of the article. I think
> sponsor-paid also bad, as the research sponsor pays through indirect
> costs a good deal of the convention system's costs. (And of course
> the same goes for university-paid, researcher-paid, and so forth).

Journal cost-recovery models are not relevant to the definition of OA.
OA is a property of articles -- in particular, whether or not they are
accessible toll-free. If they are accessible toll-free, they are OA. This
may be because they have been published in and OA journal, or because they
have been published in a TA journal and also self-archived.

David may have his preferences about journal cost-recovery models, but
not only do these have nothing to do with articles that are OA because
they have been published in TA journals and made OA by self-archiving,
but they have something to do with articles that are OA because they have
been published in OA journals only inasmuch as the OA journal must make
them accessible toll-free (to be an OA Journal). Apart from that, how the
journal chooses to recover its costs is not essential to the definition
of OA.

And it cannot be repeated often enough: The Open-Access Problem and the
Library Serials Crisis are *not the same problem* (although there are
some connections). They should not be conflated.

> For the current type of journal, library-paid is not really correct,
> as the library pays from money it receives from elsewhere, (and as it
> has been proposed earlier that the library might pay the costs of the
> new system). Reader paid or user-paid is also not right, as the reader
> or user almost never directly pays.

Again, David may have his preferences about journal cost-recovery models,
but none of this has anything to do with the meaning of OA.

It is true, though, that TA refers mainly to institutional library tolls,
not individual user tolls. By the same token, publication charges are more
likely to be institutional than individual ones.

> For the various database or repository models, I particularly dislike the
> term "archive", because this is widely used in another meaning, though
> a closely related one: an ultimate reference copy--which would be only a
> part of such a system. Database is a very general term, and has been used
> by the aggregators like Ebsco to mean their databases of journal articles
> republished from the original journals, which is certainly not the intent.

Too late. The OAI -- Open Archives Initiative (OAI) -- has adopted the word "archive" and it
accords reasonably well with what people mean by the term. Moreover,
"self-archiving" is now also in wide use, and self-explanatory
to researchers. "Self-depositing" might have worked, but it is less
transparent than self-archiving, and too late in any case.

It would be nicer if we were not also proliferating needless synonyms
(like "database" and "repository"), but none of these represent "models"
insofar as OA is concerned. An OA Eprint Archive is simply a (preferably
OAI-compliant) website in which articles can be deposited and stored
online so as to make them accessible toll-free.

> I am not making suggestions, just hoping for them.
> Arbitrary numerical, color, or place-name designations are out of
> bounds--we need meaningful names, not code.

We have meaningful names: Open Access (articles and journals), Eprint
Archives (institutional and central), and self-archiving. No new names,
colors or codes are needed. (What is needed is more OA!)

> To distinguish the pseudo-open access as used to mean open access to
> part of the journal: I think full open access and partial open access
> are sufficient, and non-pejorative.

Full vs. Partial OA is a contradiction in terms. An article is (fully)
OA if it is OA, otherwise it is *not* OA. A Journal is (fully) OA if all
of its articles are OA. Otherwise it is not an OA Journal (though some
of its articles may be OA).

The relevant journal distinction does have both a name and a color code:
A journal may be TA and/or OA. (These are not mutually exclusive! Many
of the OA Journals listed in the Directory of Open Access Journals
[DOAJ] still charge subscription tolls for
their paper editions; only their online edition is accessible
toll-free.) A TA journal may be "green" (already officially endorsing
author self-archiving) or "white" (not yet officially endorsing author
self-archiving). An OA journal (whose full contents are accessible online
toll-free to all users) is "gold."

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 Mon Mar 22 2004 - 13:00:39 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:24 GMT