Page 155 - Building Digital Libraries

CHAPTER 6

the client side, and well-documented tools like node.js provide a rich set
of JavaScript processing libraries on the server side, repositories that make
use of this format immediately enable greater interoperability opportuni-
ties, simply due to the ubiquitous nature of the data structure and the wide
range of tools and developers that are familiar with JSON and its processing
libraries.

BIBFRAME

The final data primitive that I wish to highlight is the Library of Congress’s
BIBFRAME initiative. However, unlike the other metadata formats dis-
cussed above, BIBFRAME is still largely an in-development format that
is not yet used within digital repository software. At the same time, the
BIBFRAME development has been incredibly important and influential,
because it has pushed linked data and semantic web concepts to the fore-
ground of library metadata discussion—so it is worth highlighting here.

History
The BIBFRAME initiative started as a project to develop a new data model
17
for library bibliographic metadata. Essentially, BIBFRAME would be devel-
oped to replace MARC, and it would enable libraries to move beyond some
of MARC’s inherent limitations. At the same time, BIBFRAME was very
much a test bed for investigating how linked data and semantic web con-
cepts could be leveraged within library data. As of this writing, the Library
of Congress has released two drafts of the BIBFRAME schema, BIBFRAME1
and BIBFRAME2. BIBFRAME2 fills in many of the gaps found in the ini-
tial implementation, presenting a more complete set of concepts and data
model, but offers some challenges for libraries because the BIBFRAME2
model made some substantial breaks from current RDA practice.
Has BIBFRAME been successful? From a practical perspective, it might
be difficult to argue that the BIBFRAME initiative has met its goals. Started
in mid-2000 and developed to utilize RDA concepts, BIBFRAME was ambi-
tiously scheduled to become the predominant metadata framework found
within the library community by 2010. This goal has yet to be met, however.
In fact, outside of some very targeted research initiatives, very few systems
utilize any BIBFRAME concepts though we are starting to see projects in
Europe and within the LD4L (Linked Data 4 Libraries) community imple-
ment projects that utilize the BIBFRAME vocabularies. More interesting,
libraries are also looking at a wide range of other efforts, specifically within
the schema.org community, to evaluate different ways in which biblio-
graphic data could be modeled for use in a modern bibliographic system.
While efforts are ongoing to look at ways in which efforts in the schema
.org and BIBFRAME communities could be harmonized, as of this writing,
harmonization is still a goal that is yet to be achieved. So, where does that
leave BIBFRAME and the library community? This is still an open ques-
tion. Much like the development and implementation of RDA, many in the

140