services against collections

As more and more content is made available in full-text it becomes
increasingly important to provide services against collections, not
just pointers to it. In my opinion, the "next generation" library
catalog is not really a catalog at all. Instead it is a set of
collections and services used to enable and facilitate the mission of
a library's hosting organization.

Library catalogs, by definition, are indexes to the things a library
owns (or licenses). As indexes catalogs are essentially lists of
words complete with pointers to where items can be found. These words
are authors, titles, notes, subjects, and control numbers. The
pointers are call numbers (or URL's). The primary purpose of an index
is to facilitate search. Secondarily an index provides analysis
against the index as a whole. There are this many items in the
collection about that. Shakespeare wrote one thing, another thing,
and a third thing. Most of the items are dated after this time. Etc.

Instead of learning about the content of the index, I suspect our
users are more interested in the actual things pointed to from the
index. They want to find items of interest, acquire them, and put
them to use. Read the item. Compare the item to similar items.
Extract the most pertinent information from the item. View the item
in a new and different context. Annotate the item. Share the item
with others. Incorporate the item into a bigger idea. Use the item to
confirm or deny an idea.

These are possible services against collections. They are ways of
saving the time of the reader. They exemplify growth opportunities
for libraries. Everybody has access to indexes and content. I can
carry the totality of the WorldCat index around on my iPod. Providing
services against content is a growth opportunity for libraries. Yes,
we have traditionally provided services against content, but we need
to provide more services against it and figure out ways to provide
these services in a networked environment.

For example, a library could provide textual analysis tools against
sets of full-text literature. Such analysis could take the form of
concordances -- systems counting word occurrences -- allowing people
to rapidly get a feel for a text. List the top 1000 most frequently
used words. Truth appears 339 in this document, and love appears 2
times. Art occurs 75 times and science occurs 0 times. I have done
this with the Alex Catalogue of Electronic Texts. Each item is
associated with a concordance for just this sort of examination. Try:

   * Jane Eyre - http://tinyurl.com/2woeb9
   * The Republic - http://tinyurl.com/3drf7r
   * Tarzan of the Apes - http://tinyurl.com/2nvfnu

As image analysis becomes more mature we can provide similar services
against pictures (or even scans of manuscripts). Increasingly data
sets, even in the form of simple spreadsheets, are becoming a part of
the information landscape. Here is another opportunity for growth.

Services against collections represent a growth opportunity for
libraries. With regards to library catalogs, these services need to
include the means for people to get their work done quicker and more
efficiently. Textual analysis is one example. Document delivery --
"Send it to me!" -- is another.

--
Eric Lease Morgan
University Libraries of Notre Dame