Re: Repository: publication or archive?

From: Stevan Harnad <harnad_at_ecs.soton.ac.uk>
Date: Thu, 25 Oct 2007 19:02:22 +0100

> On Thu, 25 Oct 2007, John Smith wrote:

> This confusion reflects the beliefs of two distinct parts of the
> repositories community.

Actually, I think the confusion is between beliefs and reality. Some
believe in (various) hypothetical future forms of publishing. The reality
is 25,000 peer-reviewed journals publishing 2.5 million articles a year,
85% of them accessible only to subscribing institutions.

So what is needed is that those *published* articles should be
self-archived (not "published") in their authors' institutional
repositories, so all their would-be users webwide can access them.

Hypothetical publishing reform can proceed apace. But please let's not
use that hypothetical future form of "publishing" as the name for the
present-day self-archiving of present-day published articles.

> One group want to keep the old publishing model
> with refereed journals each having their own collection of articles with
> the repositories being a supplement to this (offering alternative access
> to publicly funded information) and the other group see repositories as
> the beginning of new forms of academic publishing.

One group want OA now, to what we have now. The other group prefer to
keep waiting for some hypothetical future form of publication.

> Those in the first
> group argue that publication in a traditional journal (or similar vehicle
> like a conference proceedings) is the 'definition' of publishing and any
> other form of making publicly available is not publishing despite the
> standard dictionary definition.

No, it is academic assessors of research performance who (correctly)
deem peer-reviewed journal publishing today as the criterion for
counting as a published article, not merely self-archiving an unrefereed
preprint on the web (nor any untested third option).

> Many of those in the 'self archiving'
> community are in the first group and EPrints (for example) was begun by
> the 'self archiving' community hence the statement in the EPrints FAQ.
>
> It is all the fault of history (as many things are). In the past
> Publishers controlled all the activities related to academic publishing
> not just the 'making publicly available'. They organised the editorial
> boards, they organised the referring, etc. So Publishing (with a
> capital P) meant more than making available, it included all these
> related activities.

And, for the most part, they still do. And OA is about freeing access to
those articles from publishers' access-toll-barriers, not about freeing
them from peer-review -- or hypothetical peer-review reforms.

> The two groups described above are to some extent talking past each
> other. The first group are talking about Publishing and the second group
> are talking about publishing.

No, the first group are talking about actual publishing today and the
second group are talking about hypothetical "publishing" (TBA). And the
first group wants to free access to those actual publications today
(2.5 million per year), whereas the second group wants to replace them
by some hypothetical future form of "publication," some day.

> So putting an item in a repository is publishing (especially if this is
> the only form of public access to it) but it is not, in and of itself,
> Publishing because it does not provide those other requirements of
> formal academic publishing like 'recognition of work done' or 'peer
> review'.

Putting an unrefereed, unpublished paper on the web makes it publicly
accessible; it even makes it "published" in some legalistic/IP sense,
but it does not make it published in the academic publish-or-perish
sense. (And that reality has nothing whatsoever to do with the OA
movement,)

> These aspects need to be added by some other activities like post
> publication review (which might just be counting the number and quality
> of any citations the item gets), or it might be formally reviewed/refereed
> by a body other that the body that makes it public, etc.

Postpublication metrics, by all means. But RE-reviewing 2.5 million
already peer-reviewed publications yearly? at a time when the 25,000
journals can barely get the peers to review them competently and in time
for publishing them in the first place? This is no longer just untested
hypothesis: it verges on science fiction...

> As you may have guessed I belong to the second group and believe that
> placing an item in a repository is publishing and it can/will form part
> of a cooperative activity known as Publishing.

Try conveying that belief to the RAE, or your annual review committee...

Cheers, Stevan
Received on Thu Oct 25 2007 - 19:07:02 BST