More than a decade ago I saw a demonstration of an online literature
course, with everything from literary works to images to historical
data to student assignments presented digitally. The work was impressive
and the possibilities it generated exciting, but it appeared to me that
there was a substantial obstacle blocking widespread implementation
of such courses: at the time, there was almost no scholarly information
in digital form and seemingly no low cost, mass-production mechanism
for generating it. The Brown professor who had put this innovative course
together had devoted many months to inputting the necessary works, both
primary and secondary; it seemed unlikely that many of his colleagues
would be eager to follow suit.
What a difference a few years makes, and how little we foresaw the arrival,
and the influence of the World Wide Web and Netscape! To oversimplify,
the creation of Netscape and its successors made the Internet easily
accessible for "the average guy" and awakened publishers, information
brokers and universities to the value of creating digital information,
either at its origin or through "re-formatting" (for example, converting
a work which had first appeared on paper to a digital form). For many
years, authors had been creating their works digitally via the computer,
but the publisher's end product was print; now, almost overnight, publishers
and authors alike had the power to publish directly on the Web.
Almost overnight, libraries went from a world in which online materials
had consisted mostly of large abstracting and indexing databases, licensed
from a handful of vendors, to an environment where full-text journals,
monographs and other materials began to be offered on the Web. In some
cases, these online collections were offered at little or no cost (such
as John Hopkins University's
Project
Muse, more than 100 full text online journals), but more frequently
publishers or information brokers offered online versions of their print
publications for a substantial license fee, usually based on the community
of predicted users. Until just recently, many publishers attempted to
protect revenue by requiring a library to retain its print subscription
in order to have access to the online version; happily, this practice
is dying off as publishers begin to understand that the digital journal
is where the real future lies. Nonetheless, the digital revolution presents
a number of complex issues for research libraries.
The first, perhaps not so complex, is cost. A decade ago, the Brown
University Library spent about $10,000 annually for full-text digital
information; today, the figure is approximately $600,000. In most cases,
the digital information we license does not duplicate information available
in print and offers faculty and students capabilities which print cannot
replicate; hence, it must be viewed as an add-on, not a trade-off. Thus
far, much of the money that the Library has used for digital licenses
has been found within the Library budget, usually through canceling
marginal journals; but the fact is, the demand for digital information
is putting a severe strain on the Library's ability to acquire scholarly
material in other forms. And the $600,000 we note above does not include
any of the hardware or software infrastructure that the University must
have in place to allow us to access the Internet.
Even in cases where a trade-off is theoretically possible, faculty and
student acceptance often runs counter to pragmatic cost effectiveness.
JSTOR is a project which was initially funded by the Mellon Foundation
and which now exists as a not-for-profit online "publisher." Working
with research libraries around the country, JSTOR has located and digitized
the full back runs of the most important journals in a number of scholarly
fields, from economics to history to mathematics. The journals appear
online exactly as they look in print, yet can be keyword-searched either
collectively or by individual title. The theory behind JSTOR was that,
once these backfiles were available online, research libraries could
remove the print volumes, thus relieving the strain on crowded book
stacks. Alas, although the JSTOR database is heavily used (at Brown
and elsewhere), faculty and students have resisted suggestions that
print backfiles be sent to storage. The reasons cited are various (from
concerns about the longevity of digital information to reproduction
quality of the digital version), but to this date few research libraries
have divested themselves of their print JSTOR backfiles.
What are we buying when we license full-text information? Well, technically,
nothing. A license is just that, a temporary right to access information,
according to agreed-upon rules. Licenses typically dictate who can access
material (e.g., only "members of the Brown community"), how libraries
can make it available - many licenses forbid interlibrary loan as we
have known it - and how long access is granted (licenses are renewable,
just like magazine subscriptions). Vendors can, and have, cut off access
to libraries that violate license provisions. Moreover, without making
special provisions, libraries cannot be assured that we retain the right
to access "backfiles" of digital information if we cancel our current
subscription - something which obviously was not an issue in the print
environment.
Some librarians are also concerned about the implication of "package
deals" on collection development. For example, this year the library
invested $60,000 on an electronic package from the major scientific
publisher Elsevier, that consists of more than 1,000 medical journals
- 700 of them new to Brown. .Now, in this package, Brown acquired not
only titles we would have selected, but a number that probably would
have been of less interest to us, had we been making a title-by-title
selection. Some of my colleagues are concerned about an apparent loss
of control; I am more sanguine about this, since one of the delights
of developing library collections is discovering that scholars use them
in ways we never expected, anyway. It's very likely that Brown students
and faculty will find useful material in many of those titles we would
have thought of little interest.
Finally, and not insignificantly, can we count on the longevity of digital
information? My librarian ancestors paid money for tangible items that,
placed on a shelf, could be expected to endure for the use of generations
to come. The great research libraries of this nation are great in part
because they represent the accumulated and published thought of many
centuries. A hundred years from now, will our descendants still find
available the digital record upon which we have spent our funds? How
can we ensure that they will? Many librarians, digital experts, and
others are working on this problem, which today still lacks an entirely
satisfactory answer
Well, with all these risks and uncertainties, why are research libraries
like Brown putting their money into digital information? For the same
reason librarians have always acquired information, because it is there
and our users require it. Digital resources today are not only vast,
but can be searched and manipulated in ways which save users time and
permit entirely new approaches to research. Moreover, digital information
does not become inaccessible when the Library closes - you can reach
it when you need it, from home or office. Today, many of the Library's
online users don't even reside in this country. Were 16th century librarians
a little overwhelmed by the potential of Mr. Gutenberg's moveable type?
Without doubt. Should they have rejected his product? Of course not.
We find ourselves in a similar situation!
Merrily E. Taylor, Joukowsky Family University Librarian
