Publication versus Access-Provision

From: Stevan Harnad <>
Date: Wed, 19 Mar 2003 15:16:06 +0000

On Wed, 19 Mar 2003, Philip Hunter wrote:

> By 'publishing process' I meant the coordination of multiple versions and
> revisions of the same paper, within the same or different OAI archives.

As some of these versions will indeed be published papers (published
in refereed journals) and some will not (just unpublished preprints, or
revised drafts, or postpublication corrections or updates) I think it will
be less confusing to authors if these are not all called "publications,"
and if self-archiving them is not called "publishing" them. Let us reserve
the word "publishing" for what it has always meant to publish-or-perish
academic research authors, namely, publication in and by a refereed
journal. (I omit books here only because most are not self-archiving
candidates, whereas all refereed-journal publications are; that makes them
moot on this point.) Publishing is done by publishers or journals. And
authors publish in journals, not in archives.

So why don't we call the topic you raise "online version control" rather
than "publishing"?

> It is a publishing process to the user, in that multiple copies are
> deposited in the archives, and that version control is a feature of
> the process.

The publication process to the user (i.e., the depositing author) is
something transpiring between himself and the journal that accepts his
article for publication. It is that journal reference that is listed on
his CV, under "published" (as opposed to "unpublished") work. And it is
listed only *once*, qua publication. Several URLs may now be appended
to it in modern online CVs (such as the ones we are advocating: ) but those URLs
indicate multiple ways to *access* the publication in our new online age.
They are certainly not multiple publications.

In other words, in the online age, it is time to separate "publication"
-- which has come to mean acceptance and certification by a
quality-control authority, namely a peer-reviewed journal -- from
"access provision" -- which is simply where and how the published (or
unpublished) paper can be retrieved from the Web.

Having distinguished publication from access provision, all that
remains is the question of version control: For published papers,
the journal will have accepted a specific text. That is a milestone,
and needs to be tagged as such, in any viable online version-control
system, to distinguish it from prepublication drafts as well as from
postpublication corrections and upgrades. This is a technical matter
that can and will be sorted out once there is a large enough body
of self-archived preprints and postprints to make it worthwhile and
necessary. (But I do not believe for a moment that author concern over
version-control is what is holding back self-archiving -- nor that
version-control first needs to be perfected before self-archiving can
scale up: .)

> I think this is what Chris Gutteridge was aiming at by suggesting 'some
> negotiated automated process for insitutional archives uploading to the
> subject archive, or at least assisting the author in the process'.

This is the technical matter in question. It is worthy of discussion
among the technical implementers of the OAI protocol and of institutional
self-archiving software, but it is not a general author/user matter.

> This process has to be intelligible to the user, or there won't be
> significant take-up.

The author/user deposits and tags versions. That is and needs to be
intelligible to the user. But I do not believe that concerns about
version-control are the reason why not enough researchers are
self-archiving yet, nor that there is a problem here that needs to
be solved before there can be "significant take-up."!

> This is the territory for the development of third party services by service
> providers, rather than part of the base-level eprints/self-archiving idea.
> The absence of such services (I was suggesting) might have a bearing on the
> currently low-level of deposits in archives.

Why it is that researchers are not yet self-archiving
more, in view of its benefits and feasibility, is becoming
the "Big Koan" of the online age (for refereed research)
but I do not believe you have resolved it! OAI-services for
version-control are certainly welcome, if that is a better way to
implement it than at the archive level, but I am certain their presence
or absence has nothing to do with the answer to the Koan.

I think the answer to the Koan is that (1) researchers don't yet realize
the causal connection between access to their research and the impact
of their research and that (2) the publish-or-perish authorities don't
yet realize it either.

What needs to be done is to gather and tirelessly disseminate the objective
evidence about the access-impact causal connection -- until the token
drops for a critical mass of both these populations.

> I prefer to speak of submission to archives, since not all eprint archives
> expose all the materials which have been deposited. The act of depositing an
> eprint does not automatically imply (in practice) open access. It is
> important to be clear about what is actually happening in the world of
> eprint archives.

Depositing in an OAI archive means that the metadata will be accessible.
One of the metadata indicates whether the full-text is accessible (i.e.,
open-access). So being clear about the matter you are concerned about
above is not a "deposit" vs. "submission" matter but (like so many
things) just a meta-data tagging matter.

Here are a few more useful metadata tags:

(1) Unpublished
(2) "Submitted for Refereeing"
(3) Published (by journalname)

This retains "submission" for how it is usually used: "submitted to be
considered (peer-reviewed) for publication". Outcome not yet known.
Whereas "deposited" means placed in an archive: Otucome known; see
metadata tags for the specifics.

Stevan Harnad
Received on Wed Mar 19 2003 - 15:16:06 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:56 GMT