Page 217 - Building Digital Libraries

CHAPTER 9

chapter 6, a great deal of attention was given to the development of services
around a digital repository and the benefits associated with allowing the
harvesting of one’s repository’s metadata. The sharing of metadata goes a
long way towards promoting the open-access culture that the library com-
munity continues to cultivate—but it doesn’t relate directly to discovery. In
many cases, individuals will harvest and reindex metadata to create new
research tools or services, offering yes, a separate set of access points, but
providing the content within a different contextual framework. Ultimately,
allowing the sharing and harvesting of metadata requires organizations to
give up some level of control over the metadata and content as individuals
mix and mash their services with other materials.
Discovery is a different animal altogether. While metadata harvesting
does promote the discovery of materials within different contexts and ser-
vices, it’s not explicitly used, nor should it be solely relied upon, for outside
discovery. Successful digital repositories offer a multiplicity of discovery
avenues—allowing users the ability to choose the searching methodology.
Digital repository administrators need to consider what additional search-
ing protocols they are willing to support to provide the necessary access
points for both users and what we today refer to as “federated searching.”
In some cases, support for library protocols like SRU/W or OpenURL may
be provided by the digital repository software platform—but if not, what
and how is support added? Likewise, does the organization wish to sup-
port emerging search protocols from outside the library community like
OpenSearch, or create locally developed RESTful API, or leverage new open
standards like SPARQL? Ultimately, the question that repository owners
need to consider is how any of these services will benefit users and aid in
the long-term findability of the organization’s content. Likewise, how does
a digital repository administrator decide what legacy protocols to continue
to support, and when should these legacy protocols be deemed obsolete?
Repository administrators have a wide range of potential protocols and
search standards that could be supported by their repository software—
meaning that an evaluation of how the repository will be searched should be
conducted to meet this need. This chapter, however, will highlight many of
the protocols that are most often supported within the library community,
and it will describe the current crop of federated search software packages.

Unpacking Discovery?

Before we jump too far ahead in this chapter, we need to deal with the sticky
issue of discovery. Within the current literature, the terms “federated search,”
“metasearching,” “integrated searching,” “cross-database searching,” “parallel
searching,” “discovery systems,” and likely many other terms are all utilized
to represent the same set of concepts. For the purposes of this book, the
terms “discovery,” “metasearch,” and “federated search” will be used inter-
changeably to represent the same set of technologies and concepts. So, what

202