 |
|
|
Federated Foxholes:
Moving Search Beyond Library Science to Content Science |
|
|
|
 |
|
|
|
6 October
2003 |
|
|
|
|
While the
Library Journal points out that federated search
engines are opening up new doors for users to access
library content collections in ways to which they've grown
accustomed via public Web search engines,
Information Today reminds us that federated search is
hardly a magic solution for consistent and unified content
location. At best federated search delays the day on which
information professionals will have to face some harsh
realities about their place in managing premium content
collections. Use it for what it's worth, but be ready to
move out of your federated foxholes while you can. |
|
The disciplines that comprise
library science are long-standing and filled with thousands of
intelligent, service-oriented professionals with deep insight
into the value and structure of institutional-grade content
collections, people who have years of training in finding the
virtual needle in the haystack amongst dozens of publication
databases -and plying a trade in deeper trouble than ever
before. As detailed in a
recent Library Journal article, the likes of Google and
other popular Web search engines have created a content culture
that pits information professionals against users who have been
trained by Web searches to expect instant gratification and
good-enough results to solve the problems that get them from
one problem to the next. They are also besieged by search
engine technologists who are determined to replace human
intelligence and more than a century of content organization
methodology with algorithm-driven approximations of content
retrieval wisdom. In the eyes of many professionals (and more
students than we'd like to admit), libraries are fast becoming
comfortable places to sit more than places to do serious
research.
To service these trends, interfaces to
collection databases oriented towards non-expert users are
starting to proliferate, many based on the concept of federated
search that allow results from multiple aggregators and other
sources to be aggregated and rationalized into a common,
coherent display. Without having to deal with the arcane
details of multiple professional search interfaces, users can
begin to approximate the ease to which they have become
accustomed with "civilian" content sources. Federated search is
hardly new - aggregators such as
Dialog and
LexisNexis
have been providing federated collection results for years, and
WebFeat
has been pioneering user-level federated access via Web
interfaces at academic institutions and corporations since
1998. But players such as
MuseGlobal
are notching up federated search for libraries to a new level
of sophistication that marries user friendliness to very
powerful advanced search and organization capabilities for even
the most rarified of content collections. Yet
as pointed out by Information Today this week, federated
search remains a mixed blessing at best: legacy content formats
and protocols, limited de-duplication of search results and
non-universal authorization interfaces hamper library content
approximating the ease of public Web search engines -
regardless of the quality of the underlying content and
organization methodologies.
While it solves many important issues,
federated search is at best a
foxhole on a battlefront that is moving the world of
premium and institutional content collections away from
traditional management methodologies. As they say in politics,
"Follow the money." The money trail is leading
information professionals towards inevitable changes that will
require them to be experts in content science as much as
library science. Here are a few major trail blazes that need to
be followed in the broader world of content and related
technologies when considering how heavily to invest in library
science-oriented search solutions:
- Follow the research money. In
the early days of search engine technology, much of the
research was funded to help universities and major research
institutions manage their content collections more
effectively. Investments in these search technologies were
soon outstripped by the requirements of a highly competitive
marketplace for commercially-oriented Web solutions that
consumed most of the major funding - and the talent that fed
off of it. While the dot-com bust freed up some academic
talent for library science solutions, U.S. government
spending bridged much of that talent into "war on terrorism"
solutions through the worst of the economic downturn. Now
that consumer-oriented and busienss-oriented search engine
competition is heating up, the best and the brightest
solutions have resumed their migration away from library
sciences (witness
Google's recent acquisition of Kaltix, a promising
leading-edge outgrowth of Stanford University's leadership in
search technology). Information professionals are right to
insist on preserving the value of the human element in
content research, but when the funding for solutions to
support them is moving away from library science, so must
they.
- Follow the institutional budgets.
Institutional investments in Web search infrastructure
continue to grow in the face of overburdened library budget
managers who find themselves increasingly having to choose
between inflated collection acquisition costs and expensive
technology investments. Inevitably this places pressure on
the breadth and quality of those collections - and the value
of investing in technology to improve access to them.
Publishers of professional-grade content oftentimes find
themselves caught in the same squeeze, having relied too
heavily upon aggregators to define the context of their value
propositions in an increasingly narrow marketplace. The
empirical philosophy of "investing in what works" will
continue to drain money away from library-oriented premium
content collection management as long as organizations
continue to improve their operations to mine more
cost-effective results from other content collection schemes.
- Follow the commercial models.
When institutional librarians have a firm control over the
technology to access premium collections, their role in
negotiating commercial agreements with content vendors remain
largely unquestioned. But as more professional content gets
integrated directly into workflow-oriented portal solutions
with other content sources, institutional librarians may find
themselves holding less authority in determining content
value and the return on total content investments. Newer and
less centralized methods of enabling access to premium
content such as Digital Rights Management will further weaken
the need to provide content vendor-supplied search
capabilities to access their underlying collections, further
decreasing the importance of purchasing searching
capabilities specifically for particular aggregated
collections - and the need to federate them. Centralized
purchasing of content by institutions probably will exist
always at some level, but it will be part of a much broader
commercial content model far wider than traditional library
retrieval management.
Federated search is a very useful tool
for now, especially when commercial collections are combined
with institutional and public content sources. But improving
search results beyond the federated approach is likely to press
professional content collection management well beyond the
abilities of today's library management and commercial
structures. Are information professionals ready to come out of
their foxholes and charge towards this new front line? To your
battle stations, everyone.
-
John Blossom
To top of page
 |
