Electronic repositories for publications

["William C. Keel" <keel@bildad.astr.ua.edu>]

If this doesn't veer too far off-topic, I'll point out that a model
of what it takes to run such a thing exists in the NASA Astrophysics
Data System (ADS) search and retrieval facility:
http://adsabs.harvard.edu/abstract_service.html?nosetcookie=1
which I used heavily in compiling some earlier messages on impacts
and Nemesis. Some key aspects in making this work are:

- NASA figured it was interested enough to fund not only the initial data
  ingestion but upkeep (several FTE's worth)
- it's hosted by an eductaional institution, not a for-profit company
- many of our major journals are managed by socities rather than
  commercial publishers. _Nature_ abstracts are not available on-lin
e,
  while many full texts from other journals are. Actually, the major
  discipline-specific journals are now scanned back to their inceptions
  in the last-but-one century.

It has especially useful features letting one trace the citation chain
up and down in time.

On the not-yet-published side (should we call that the Holtz period
in honor of his admonitions), physics, astronomy, and math are
hosted at the Los Alamos preprint server
http://xxx.lanl.gov/
(to the frustration of those working where blockers barf at the xxx).
This system uses volunrary submissions of submitted papers and
online compendia, and has a software setup that could usefully be
copied somewhere. Again, though, one shouldn't underestimate the
people-time needed to keep such things going, but they are fantastically
useful resources. I suspect we've benefitted from that fact that
"it is the fate of astronomy to become the first all-digital science"
(from Larry Smarr; it didn't clearly work out that way, but we were 
close) and the fact that by 1970 practically all cited work appeared
in only about 8 journals, so scanning and indexing was a manageable
if multiyear task. For most journals, there's a period after publication
when full-text access still requires either institutional or personal
subscription, as a form of copyright and reveneue protection.

To summarize, compiling such a thing in paleontology would be quite
labor-intensive but obviously worthwhile. It would probably be a key
aspect to have institutional support for both funding and hosting.

And a happy genuine turn of the millennium to all,

Bill Keel
Astronomy, University of Alabama