Page 113 - Building Digital Libraries

CHAPTER 5

It’s this development of a shared web, or knowledge graph, of information
that continues to push cultural heritage organizations to explore ways to
not only utilize but reconcile legacy metadata in order to take advantage of
these semantic web concepts.

Not Just a Library Standard

Traditionally, libraries have had very few vendor choices when it comes
to software purchases or hiring outside consulting. The main reason for
this difficulty has been the standards that libraries have become reliant on.
MARC is a good example. No other profession utilizes MARC or RDA as
a descriptive framework. As a result, only a handful of “library” vendors
support the creation and use of legacy library data. Libraries inevitably have
few ILS choices—and larger organizations essentially have to choose from
only two or three larger vendors. So while these “library” standards have
served libraries well for many years, they have also perpetuated a retard-
ing of software development within the library community. Today, library
“innovation” is essentially pushed by a handful of large vendors providing
large, monolithic systems. To a large degree, the library community’s con-
tinued reliance on MARC and RDA has stagnated library development over
the current generation. Non-MARC metadata schemas (either in XML,
JSON, etc.) have the potential to unyoke the library community from tra-
ditional “library” vendors and enable the information sciences community
to use the technology and innovations developed for web-native standards
and best practices. While the library and information sciences community
has created XML-based schemas like MODS, METS, and Dublin Core, and
non-XML schemas like IIIF, the fact that these schemas are in open and well-
understood “languages” allows libraries to look outside the traditional library
vendors to a broader development community. For the first time, libraries
can truly start to leverage the larger open-source community, and not simply
the handful of open source developers that happen to work for libraries.

JSON

While library metadata formats and data models are primarily reflected in
XML and XML Schema documents, the use of XML as a communication
format between applications has become less and less common. To some
degree, this is probably why library communication standards like SRU/W
have failed to gain significant traction as a replacement for older standards
like Z39.50. Though XML provides a technology-neutral method of trans-
ferring data between systems, the processing of XML documents carries
with it significant overhead. Additionally, the technology needed to transfer
XML content on the client side (when used in web browsers) has largely
remained stagnant. The current XSLT and XQuery standards may continue
to be developed, but as of this writing, support on the client side remains
98