Page 139 - Building Digital Libraries

CHAPTER 6

History
The history of Dublin Core traces back to the 2nd International WWW
Conference in 1994. During the conference, Yuri Rubinsky (of SoftQuad),
Stuart Weibel, Eric Miller, and Terry Noreault (of OCLC), and Joseph
Hardin of the NCSA (National Center for Supercomputing Applications)
had a hallway conversation focusing on the current difficulties of finding
materials on the Internet. Because of that conversation, OCLC and the
10
NCSA decided to team up and collectively look for a solution to this grow-
ing problem. In 1995, at Dublin, Ohio, OCLC and the NCSA led a joint
workshop called the OCLC/NCSA Metadata Workshop. The workshop was
to focus on three primary goals:
1. Deciding what descriptive elements would be needed to
promote the findability of all documents on the Web
2. Exploring how to create a solution that would be flexible
for past, present, and future online publication on the Web
3. Exploring how to promote the usage of such a solution if
it exists

After this meeting, the participants were able to produce an agreed-upon
set of descriptive elements; these were fifteen general descriptive terms that
could be universally applied to virtually any resource currently available on
the Web at the time. From these initial fifteen elements, the Dublin Core
Initiative was born.
Originally, the Dublin Core schema was defined primarily as a method
for describing web documents, like websites, through the use of meta tags
within a documents header. These tags would then provide a reliable mecha-
nism for search engines to harvest and index materials, since metadata
elements like titles, descriptions, authors, and even subject access points
11
could be easily identifiable. A number of tools were created to help users
who wished to mark up their web documents.
Figure 6.4 provides a simple illustration of how Dublin Core was to be
used on the Web. Prior to the Dublin Core specification, web page devel-
opers had few options when it came to tagging documents for indexing
by search systems. Dublin Core’s syntax made this possible, giving web
developers a standard set of tags that could be used to create uniform
metadata for documents published on the Web—while at the same time,
giving search providers a standard set of metadata from which to harvest
and index. However, in general, this concept wasn’t very successful, since
search engines tended to ignore meta tagging due to tag abuse. Approved by
ANSI (American National Standards Institute) in 2001, the Dublin Core
12
Initiative schema has since been accepted as an ANSI standard (Z39.85
2001) and an ISO (International Organization for Standardization) standard
(15836) and has been adopted for formal use by several national govern-
ments (Australia, United Kingdom, Canada, etc.).
The Dublin Core itself is made up of fifteen central, unqualified ele-
ments known as Unqualified Dublin Core. These elements represent the
124