Re: Eprint versions and removals

From: Stevan Harnad <>
Date: Wed, 11 Jun 2003 23:18:05 +0100

On Wed, 11 Jun 2003, David Goodman wrote:

> I do not disagree with Arthur, or with Steven, when they wish to
> separate informal prepublication from full publication.

This is *not* about separating "informal prepublication" from "full
publication". It is about separating issues of publication (and
preservation) from issues of *access* (to those publications)!

> However, many people are not using the servers that way.

What people, and what servers? I assure you that those servers that
contain published postprints are being used (by those who do not have
institutional access to the publisher's toll-access version) *exactly*
as the publisher's toll-access version *would* be used, if it were

And preprints are a bonus, giving results earlier, but with a risk. Let
us not focus on them as they are side-effects, not main-effects.

> This isn't new--in many fields
> informal or grey publication has been as far as things ever get.

This has nothing to do with the "gray literature" (to the extent that I
even know what that means: I suppose it is material that is *not* destined
for submission to or publication by a peer-reviewed journal, hence neither
postprint nor preprint, but perhaps resembling preprints, faintly).

But this is not about black (published postprint) vs. white (unrefereed
preprint) vs. gray (neither). It is about supplementary *access* to
the black, provided by its authors, for the access-denied, by
self-archiving (in schmarchives, if you like). The white is earlier
(and partial, somewhat risky) access to the eventual black, and the gray
is no kind of access to it (hence irrelevant to this Forum, which is
about open access to the peer-reviewed, published literature).

> In practice, material made available that way may or may not count as
> technical publication, but it counts in the way that matters:
> disseminating material to the public.

What material? And why are we talking about this at all? We are not
trying to find ways of changing what counts as "publish" in "publish
or perish." That stays the same. We are talking about *access* to what
is published.

> We all use whatever information we can get--hopefully we have been taught
> enough to judge its reliability on both external and internal characteristics.
> That's what graduate education is for. Making
> sure the material is there, no matter how it was disseminated, is what
> libraries should be for. As Arthur says, this takes the cooperation of the
> authors, and the agents the authors use for dissemination, however called.

I'm lost! How did we get into this? Publishing is publishing, and
remains publishing. Access is what is changing, because of the new
possibilities opened up by the online medium. (And yes, people know,
online, as well as they do on paper, how to distinguish refereed,
published papers from unrefereed, unpublished ones.) And access is what
is important, and urgent. The preservation of working drafts is a worthy
but very minor, non-urgent issue. Please let's not conflate it with
substantive, urgent matters: the fact that most of the published
refereed journal literature is stilll only accessible to those whose
universities can afford the tolls. It is *this* that can and should be
remedied as soon as possible. All these other issues are extremely minor

> Dissemination increases entropy, and is therefore irreversible. It takes
> work to destroy things. It similarly takes work to keep the disseminated
> things organized enough to find them and get at them again.

This is all very abstract. The concrete problem is that most of the
toll-access journal literature is not accessible to most of its would-be
users. And self-archiving it can and will remedy that (if only we don't
keep getting distracted or side-tracked by irrelevant side-issues!).

Please stop worrying about a handful of removed first drafts when we
have empty archives and an annual 2,000,000 postprints still to be

Stevan Harnad
Received on Wed Jun 11 2003 - 23:18:05 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:58 GMT