Re: Proposed update of BOAI definition of OA: Immediate and Permanent

From: Stevan Harnad <>
Date: Sun, 13 Mar 2005 21:10:59 +0000

On Sun, 13 Mar 2005, Jean-Claude Guedon wrote:

> a publication is a published article; this is indeed a common and
> legitimate use of the term. However, from a librarian or publisher
> perspective, a publication is a journal with the result that a
> librarian/publisher publication is made up of researchers'
> publications... quite confusing indeed.

Not confusing at all if we keep in focus what OA is about, and for:

Open Access is provided *by* researchers *for* researchers (and for
research progress and benefits).

OA is not provided either by or for librarians (although librarians are
a great help in advocating and facilitating it).

Nor is OA provided by or for publishers (although OA publishers have
made a strong and important commitment to and investment in it).

A researcher provides OA to his *own* publication by either having
published it in an OA journal or having published in a non-OA journal
and self-archived it.

Hence, yes, the journals in question are indeed "publications" too,
but what researchers write, read, use, build-upon, cite, and list
in their CVs as "publications" are not the *journals* they publish
in, but the *articles* they publish in them.

I don't think there is anything confusing in this at all.

> Let us push the question just a little further: what is a researcher's
> publication? is it a peer-reviewed article?


(OA is not concerned, in the first instance, with other forms of research
publication such as books, because books, unlike journal articles, are
not necessarily author give-aways, written solely for research usage
and impact rather than for sales revenue or royalties.)

> Does the peer-review have to be done by a journal?

What is a journal? (I would say that it is a peer-review service
provider/certifier, plus an access-provider -- on-paper and on-line --
and sometimes also an archiver and to a degree an editing and
mark-up service-provider).

In any case, the target literature for OA is the 2.5 million articles
published annually in the world's 24,000 peer-reviewed journals.

Do those journals have to be called "journals"? Not necessarily. One
could call them whatever one wishes. It is the fact that they provide
the peer-review service that is essential.

But given that the peer-review for the 2.5 million annual articles *is*
currently being provided by the 24,000 peer-reviewed *journals*, and
given that it is those 2.5 million articles that are the target of the
OA movement -- i.e., it is toll-free online access to *them*, and not to
something else, that is at issue here -- one is not free to substitute
for these real, existent journals and journal articles a hypothetical,
non-existent "branding mechanism."

Because we are trying to provide immediate Open Access to real, existing
journal articles, today, for sure, not to hypothetical "branded entities",
someday, perhaps.

(Peer-review/certification, by the way, is not a "branding" mechanism
analogous to affixing the "Coca Cola" label to certain dark-brown,
sugary beverages in order to make them sell better. It is the result
of a quality-control process in which an answerable, qualified expert
[the editor] selects further qualified experts, answerable to him, to
advise him and the author on what needs to be done to a submitted paper
if it is to meet the established quality standards of a particular
"journal." The substance is in the dynamics of the evaluation,
adjudication, recommendations, revision, perhaps re-evaluation, and then
the eventual certification, not in the affixing of the brand-name itself
that comes if all goes well!)

> Does it have to be published in a journal?

What is "it"? "It" may not have to be published in a "journal," but the
2.5 million journal-articles that are the targets of OA are.

> And, by the way, what does "publishing" really mean?

We can philosophize or even legalize about that if we like. But meanwhile
access and impact are being lost: Access to the 2.5 million articles
"published" in the world's 24,000 "journals." It is those "publications"
that are being written, read, used, cited and listed in researchers' CVs
as "publications" or (synonymously) "published articles." It is to 80%
of those that we still do not have OA, and for which a substantial amount of
research impact is still being lost daily, weekly, monthly, yearly.

Do we really want to do the semiotics of "journal," "publication" and
"publication" and speculate about hypothetical "branding mechanisms"
while all that research impact continues to be needlessly lost, and the
solution is fully within reach?

> Placing in the public eye for example through some web site,

Placing what on the web? An unrefereed preprint? Useful (and should be
strongly encouraged wherever desired and feasible), but certainly no
substitute for peer-review.

Or placing a peer-reviewed, published postprint on the web? But that is
not publishing, it is access-provision. The publishing was done in and
by the journal it was published in (if the peer-review outcome was
successful! Without that, it is merely vanity-press self-publication.

> or depending on an established institution that claims the mediating role
> of "publisher"?

It always helps if one's peer-review service-provider ("journal,"
"publisher") has an established track-record for quality, and we have
about 24,000 of them today, with new ones popping up (and new and
some old ones) vanishing near-daily too...

> In short,
> vocabulary cleaning would be useful here, and it would be useful in two
> ways: for the vocabulary structure itself, of course, but also as a way
> to provide an interesting terrain to negotiate the exact terms of open
> access.

I think we can safely update our definition of OA to make it explicit
that it must be immediate and permanent, and make do with what we call
"journal," "publisher" and "publication" today, without having to do
any more major semiological house-cleaning. But one is always in favor
of clear speaking and thinking...

Stevan Harnad
Received on Sun Mar 13 2005 - 21:10:59 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:49 GMT