Re: The archival status of archived papers

From: Stevan Harnad <harnad_at_ecs.soton.ac.uk>
Date: Tue, 3 Dec 2002 22:54:32 +0000

On Tue, 3 Dec 2002, Bob Parks wrote:

> Stevan Harnad writes:
>
>sh> It cuts both ways. Yes, authors should not start archiving willy-nilly
>sh> every raw draft and every afterthought.
>
> The word DRAFT implies correction and updates. In economics, where
> working papers and revisions of them are extremely common, one would
> NOT want a previous DRAFT of a paper available since it might be wrong,
> and had been corrected, and it would injure both the reader and the
> author to have the incorrect version present. Ginsparg and I traded
> a lot of notes about this when EconWPA was born - Paul kept a revision
> date, but not the prior drafts and I think that arxiv still works that
> way.
>
> I think that if you make the previous drafts remain forever, authors
> will be reluctant to post DRAFTs at all. OED defines draft as
> "A preliminary sketch or rough form of a writing or document,
> from which the final or fair copy is made. "
> which begs change to the draft.

OED's definition was written before the Internet. One can certainly
write on a draft, prominently "This is just a temporary draft, and will
be revised." One can even add "Please do not quote or cite." But if you
put that on the Web, not only will some people nevertheless quote,
cite (and link), but the draft is not necessarily ever removable, in
principle, even if the original draft and website are removed and the
URL no longer works. Once it is publicly posted, there will not only
be fugitive saved versions downloaded earlier by users, but copies
may well have been mirrored, cached, harvested, and reposted willy-nilly,
persisting till doomsday. This is simply a fact of PostGutenberg life --
and a confusing fact, detrimental to scholarly reliability and rigor,
unless we devise a means of version-identification, tagging and control
that does not rely on people simply looking the other way if someone
writes "this is just a draft."

Just as anything I ever said can be repeated by hearsayers (and taped,
and re-played), so everything I ever wrote can be. If I made only one
copy, and showed it to only one person, I am fairly safe (if I can trust
the person not to repeat any of it to anyone), but it's all down the
slippery slope from there.

Will the self-archivers of preprints be discouraged from self-archiving
after having this very banal fact about human communication and human
nature pointed out to them? (I rather doubt it.)

>sh> But they should not feel
>sh> constrained in doing corrections and updates whenever they are needed
>sh> too. Authors should know, though, that from the moment they place a draft
>sh> into a public open-access archive, it may be read, cited, and pointed to
>sh> -- that specific draft -- in perpetuum. That is part of what it means
>sh> to have archived something publicly.
>
> On this I would disagree.

Disagree in the sense that you don't believe it may be read, cited,
and pointed to in perpretuum (even if it is withdrawn)? Or in the sense
that you don't believe it ought to be?

> But the point is to further open access
> to research, and putting in a constraint that all prior DRAFTs are
> available in perpetuity will make for less archiving of working papers.

I think deletion should be a (discouraged but available) option, but
with a persistent tag for the deleted (null) text, as a place-holder for
would-be citers who did read that draft and do want to refer to it (even
against the author's request, and even backed up only by hearsay). The
best corrective for this is to self-archive the updated, corrected draft
too, and have the archive pointers always point to that too, even the
null (deleted) ones. (But I really do think that if someone's draft is
so raw that they would rather it could vanish without a trace, then
perhaps it was too early a draft to post publicly in the first place,
and should only have been emailed to a few trausted colleagues!)

> Maybe not for other professions, but certainly in economics, business,
> and political science (subjects about which I have knowledge).

I am sure this was the practise and expectation in paper days, when
drafts were sent only to specific trusted colleagues, but it is a fact
that public posting on the Web is (like publication) another ball-game
(a bit more like guassian roulette).

> The persistent URL should, as with arXiv, point to the most recent
> draft and penultimate drafts should be in the trash.

That is an option that should be available, but its use should be
strongly discouraged. Better to selectively email the potentially
embarrassing drafts, intended to be forgotten, and self-archive only the
ones one feels one can live with being seen publicly (and potentially
remembered and referred to forever). It is, after all, something of an
antidote to unwelcome citing and quoting to be able to point to the
extant draft and say: "See, it said 'temporary draft, to be revised, do
not cite or quote'...."

(Ceterum censeo: This is all irrelevant to the issue of open access,
which is mainly about open access to the research literature after peer
review. How early a draft one wishes to make openly accessible before
peer review is a matter for the author to decide. But open access should
in general be thought of as being forever.)

Stevan Harnad
Received on Tue Dec 03 2002 - 22:54:32 GMT