----------------------------Original message----------------------------
Here are some tentative answers to queries about Usenet and the
archiving of elecrtronic journals:

> Date:     Tue, 7 Apr 1992 08:14:00 EDT
> Sender: "Publishing E-Journals : Publishing, Archiving,
>        and Access" <VPIEJ-L@VTVM1.BITNET>
From: Gail McMillan <GMCMILLA@VTVM1.CC.VT.EDU>
>
> What about Usenet?
>
> Chuck Bacon [crtb@HELIX.NIH.GOV] wrote:
>
> >>"Usenet is an anarchic system for remote access to all kinds of rapidly
> >>changing data. It was not designed for archival use. There is almost
> >>nothing in even the deepest and oldest archives on Uunet, for instance,
> >>which is more than ten years old.
> >>
> >>"Involvement by the management of some of the archiving companies, like
> >>Uunet, would do the library community some good (would somebody go but
> >>Rick Adams).  Electronic archiving should beocme an important industry;
> >>it is not, at present . The safest and best archives still seem to be
> >>on paper and film."
>
> Having experienced it only once, accessing Usenet seemed very difficult
> from my campus mainframe (VM at VPI&SU). In theory, I understand that
> if an e-journal is 'mounted' on Usenet, then one does not have to
> subscribe to the e-journal to have access to issues and articles to
> read, print, or download to diskette.
>
> Is Usenet more difficult to access than INFO on VM1? Is it someone's
> responsibility to see that every issue of Psycoloquy, for example, is
> available? I'm asking from an archival point of view but also as a
> librarian who wants unrestricted and timely access to electronic
> journals. Is there a point in time (even a theoretical one) when back
> issues will not be as readily available as current issues on Usenet?
>
> Are any libraries advocating access through Usenet rather than more
> controlled, online-on-campus access to electronic journals?
>
> Gail McMillan Serials Team Leader University Libraries VPI&SU.
> Blacksburg, VA - (703) 231-9252.

These are very important questions, and are even becoming urgent ones.
I am currently advocating that libraries set up Usenet or Usenet-like
interfaces to make the existing electronic journals immediately
available to their readership through public-access terminals like the
ones that allow readers to search library book and journal holdings. The
Usenet interface is not ideal, but it's a good first approximation, and
it already exists. A lot of different customized interfaces would
probably have more trouble getting started, and then there would be a
nonuniformity problem.

What I recommend is that libraries get a Usenet feed and restrict what
they make available to the electronic journals on Usenet as well as the
few higher level scholarly discussion Usenet groups in various
disciplines. Of the 3000 or more groups on Usenet, fewer than 50 would
qualify, perhaps considerably fewer. Once the Usenet software is set
up, it is then easy to add to it, in the same format, "local" journals
that do not actually originate from Usenet. So, for example, single
subscriptions to non-Usenet electronic journals (that are distributed
either through Bitnet's Listserv, or as independent mailing lists)
could be "fed" into the local Usenet server and would then appear
and be available like all the other Usenet groups. This is a first step
that libraries could take immediately, and it would greatly accelerate
the evolution of electronic journals.

Archiving is another matter, and one in which the libraries could again
take an active hand. The point about paper/film being the safest
and best archives, above, is, I think, completely wrong. Tapes and disks
are just as safe, as long as they are backed up by multiple copies at
multiple sites (as paper and film are) and as long as they are
systematically transferred to any new storage media that develop as
archiving evolves.

Insofar as archiving is concerned, Usenet itself is very poor; the
software saves the current material for a while and then just erases
it. There may be some permanent central archive, but I don't know where
it is or how it is accessed. Moreover, most of the material on Usenet
is not WORTH archiving in perpetuum. So here too the libraries can take
an active role, arranging for local permanent archiving (on tape and
disk) of the chosen few Usenet journals, with some sharing of the load
at a few selected remote sites. This way, the electronic journal archives
will be multiple and distributed.

A question is asked about PSYCOLOQUY: This journal is published both
ways. On Bitnet it is distributed by listserv@pucc.bitnet as "psyc" and
on Usenet, it is gatewayed from phoenix.princeton.edu as
"sci.psychology.digest." The Bitnet version is also archived by
listserv, and back issues are permanently searchable and retrievable
using the listserv software. As mentioned, Usenet itself does not
provide an archive. However, we are also permanently archiving
PSYCOLOQUY on princeton.edu, where it is publically searchable and
retrievable by anonymous ftp (directory: pub/harnad). In addition,
several libraries that are making PSYCOLOQUY available along the lines
described above are also independently archiving it. Finally, secondary
services like GOPHER are also archiving it. This redundancy is
desirable and important. Finally, we are working on software for
archiving and retrieval that will allow full-text, real-time searching
using ordinary informal English queries (with the help of Bellcore's
"Latent Semantic Indexing") rather than just by ftp or Boolean search.

Stevan Harnad
Co-Editor, PSYCOLOQUY
Department of Psychology
Princeton University
Princeton NJ 08544
harnad@princeton.edu
609-921-7771