The most important problem facing digital libraries is how to pay for them. Traditional libraries get general support from universities, while economists are always after some way to charge the users. And numerous other organizations ranging from telephone companies to startups thought they could make money at this. Libraries are caught between general budget limitations and the high prices of commercially published journals. See, for example, Business model issues in the development of digital cultural content. Perhaps because this is such an important issue, I've run on quite a bit on the references here: there is a lot published and I keep running into new things to recommend.
Perhaps the most vocal debate is whether digital technology could dramatically reduce the cost of providing traditional scholarly journals, if we had some different way of organizing the business. Hear, for example, Publish or be damned, a BBC radio discussion of science publishing.
Among the best experiments going on are JSTOR and PLOS. The assignment is to hand in up to three pages discussing the business models of JSTOR and PLOS and suggesting how you think publication of scholarly materials should be supported in the future, and whether you think either JSTOR or PLOS has a long-term future.
In addition to the organizational sites
www.jstor.org and
www.plos.org,
some directly relevant papers are:
http://www.educause.edu/ir/library/pdf/ERM0164.pdf
(a view of JSTOR)
http://www.plos.org/news/announce_moore.html.
http://www.biomedcentral.com/news/20030822/02.
http://www.jcb.org/cgi/content/full/162/2/171
http://www.snaponline.org/public/articles/index.cfm?cat=76 (the
publisher viewpoint)
If you want to read more,
other papers in the area include (some are quite long):
http:// cogprints.ecs.soton.ac.uk/archive/00001699/index.html
Steve Harnad is very prolific; but he's consistently advocated the
free-publication position.
http://www.sims.berkeley.edu/~hal/Papers/dlib.html Hal Varian
is a key researcher in the area but it's hard to find a paper from him
with no math in it.
http://www.dtc.umn.edu/~odlyzko/doc/tragic.loss.long.pdf
(for some reason the pages are backwards).
Andy Odlyzko is the person writing about networks and digital libraries who
has the most data in his papers. His different papers usually don't repeat.
This is one of the early ones and is now a sort of classic.
http://www.econ.ucsb.edu/%7Etedb/Journals/jpricing.html
Ted Bergstrom's papers, an economist with tables of cost per page and cost
per citation for different journals (an economist with real data instead of
equations; they'll probably force him out of the profession).
http://www.cdlib.org/inside/groups/jsc/ebooks/ebookfinalreport.pdf
http://www.nybooks.com/articles/546
http://www.arl.org/scomm/scat/getz.html
Should you want to read even more about this, my short view is in
http://www.lesk.com/mlesk/ derek02/derek2.html, and
a good overall summary of many issues is found in
http://www-personal.umich.edu/~jmm/papers/peak-harvard97/, but this one
is a harder slog. There's a lot of data in the reports of the PEAK project
if you later find yourself needing it. The web pages of Varian and Odlyzko
both have multiple references. Historically, Don King did much of the work
on the value of information and the paper at
http://www.si.umich.edu/PEAK-2000/king.pdf
is good, although not a substitute for the many books he wrote.
And the best known book in the area is Information Rules by Hal Varian
and Carl Shapiro. There is a full book about JSTOR entitled
JSTOR: A History
and written by
Roger C. Schonfeld.
You might well complain that many of these references were written before the dot-com crash and don't reflect the current cynicism; right, but the data and arguments are generally valid. At least I haven't assigned George Gilder.