[Below is a reply to the following query (identity removed)]:
> Thank you for your email yesterday on electronic archiving.
>> It arrives, by curious coincidence, precisely at the time we were
> scanning some of our articles with the intention of putting them on our
> website, and subsequently forwarding them to your CogPrint archive.
>> Now I am somewhat confused. You write of new software for doing this,
> which should be freely available shortly, but not just at this moment.
> Does this imply the use of a format other than pdf or html (?) or is
> this, among other things, a way to keep the information involved in an
> article down to the Kb as opposed to the Mb range, but still using
> these conventional formats ?
>> Perhaps your new software sidesteps the following question, but in case
> it doesn't - bearing in mind I do not wish to court copyright problems,
> is it a recommendation to use html, where the published format is lost,
> or is indeed a pdf version without risk (despite retaining the
> published format and publishers's logo etc) [Your previous instructions
> for sending eprints for archiving mentions adobe-pdf files, but my
> random selection of a few cogprints to view only turned up html
(1) It doesn't matter whether you self-archive your papers in CogPrints
(a centralized Cognitive Science Archive) or in an Eprints Archive
set up using the eprints.org software at your institution. The beauty
of OAI-compliant interoperability is that these two forms of
self-archiving are now equivalent. And both CogPrints and Eprints
accept papers in a choice of formats: (1) HTML, (2) PDF, (3) TeX,
(4) PS, (5) ASCII, etc. It need merely be screen-readable.
The only relevant question is speed: The Eprints software is now in
beta-2 stage, and the operational version should be released within a
month. That will then need to be installed at your institution, if
that is the route you decide to take.
CogPrints is ready to receive the papers right now. So I would suggest
archiving whatever is ready in CogPrints until the point when your
institution Eprints Archive(s) is/are up and running, and then continue
either way, as you choose.
(2) About copyright, please see the CogPrints copyright FAQ
as well as the "H/O Strategy" for legal self-archiving now:
The H/O strategy described in the URL above is sufficient for the
current and future literature. For past literature (if it is covered by
a copyright transfer agreement you signed that does not allow online
self-archiving), I suggest asking your publisher for permission in the
first instance, and if granted, archive what you like, including the
publisher's PDF version if you prefer; most publishers should grant
permission as there are no significant sales after about a year
For those past papers whose publishers refuse, the strategy is to scan in
the text and figures, reformat them, and do some substantive revision
-- enough to make it into a substantively revised second draft. This
can be done by adding more references, more data, enhanced figures, and
of course more text.
There is no legal metric for how much enhancement turns a paper into a
revised new edition, but I would say that virtually no risk of even the
threat of legal action is involved; at worst, after a threat one can
elect to remove a paper if one wishes to capitulate, but it is
noteworthy that not one of the 130,000 physics papers archived in the
Los Alamos Archive since 1991 has been challenged or removed to date! The
conflict of interest in such a challenge to a publisher's authors is so
blatant that it would be very bad public relations for publishers to
make it, and I believe that publishers are aware of that. Making that
conflict of interest explicit and public would only hasten its
resolution in the direction that is obviously optimal and inevitable
for research and researchers, and clearly feasible and within reach.
As with the H/O preprint + corrigenda strategy, the revised, enhanced
draft should have a file linked to it indicating the corrigenda -- but
this time the ones that were made to turn the final published draft
into the revised, enhanced draft.
The draft + corrigenda strategy is a mild inconvenience for now, but it
is needed only for the minority of cases where the other options above
are not available, and it will not have to be done for long, as the entire
literature will be irreversibly freed by self-archiving. For the many
potential readers who could not otherwise access your paper at all
(because their institution has no subscription, license, or
pay-per-view for the journal it appeared in), the draft + corrigenda
option will be infinitely better than (and more convenient) nothing --
and, as I said, will be sufficient to quickly propel us all to the
optimal and inevitable outcome of it all.
Stevan Harnad harnad at cogsci.soton.ac.uk
Professor of Cognitive Science harnad at princeton.edu
Department of Electronics and phone: +44 23-80 592-582
Computer Science fax: +44 23-80 592-865
University of Southampton http://www.cogsci.soton.ac.uk/~harnad/
Highfield, Southampton http://www.princeton.edu/~harnad/
SO17 1BJ UNITED KINGDOM
NOTE: A complete archive of the ongoing discussion of providing free
access to the refereed journal literature online is available at the
American Scientist September Forum (98 & 99 & 00):
You may join the list at the site above.
Discussion can be posted to:
september98-forum at amsci-forum.amsci.org