background image
DigiCULT 33
future Semantic Web developments which will tend
to be more resource intensive. The critical issue facing
KOS in Web environments is that existing standards
are based in the print world and are not concerned
with data interchange formats. KOS intellectual
resources can be exploited in searching. However, the
lack of standardised access and interchange formats
currently impedes the wider use of such resources
in the distributed Web environment. Programmatic
access requires commonly agreed protocols building
on lower-level standards, such as Web services. The
development of common KOS representation formats
and service protocols is closely linked. Progress needs
to be made on both dimensions if standards are to
be achieved. A service protocol should be expressed
in terms of a well defined but extensible set of KOS
data elements and relationships, with the relation-
ship type a parameter to the protocol commands.
This would allow the specialisation of the current
thesaurus relationships.
Users tend to be unaware of the relative effective-
ness of different search techniques on any particu-
lar collection and need assistance in search strategies.
Support for translating from user-specified free text
terms to appropriate controlled vocabulary terms is
particularly important, both in disambiguating hom-
ographs (a choice of KOS concepts) and in sugges-
tions where the user is having difficulties in locating
suitable controlled terminology. More work needs to
be done on simple `search box' interfaces, where KOS
are used behind the scenes to support free text search.
However, service protocols also need to be able to
support the development of innovative and responsive
Web interfaces that encourage different types of users
to take full advantage of the resources offered by KOS
for searching digital heritage collections.
A substantial amount of intellectual effort is
expended in the initial compilation of a complex
knowledge structure such as a thesaurus. Additional-
ly, the evolving nature of language and culture dic-
tates that, once compiled, frequent maintenance of
such resources is required in order for them to remain
relevant to contemporary audiences. Ease of dissemi-
nation, maintenance and use is essential for the fruits
of this effort to be fully realised. A balance needs to
be struck between maintaining a standard version
of a commonly used KOS for interoperability, while
also allowing local institutions to tailor and augment
it. This might, for instance, involve adding more spe-
cific local concepts as leaf hierarchies. In the long
run, there may also need to be agreement on intel-
lectual property right issues, and in some cases possi-
bly licensing models, if we wish to make available for
general use thesauri developed by a wide spectrum of
organisations.
A
CKNOWLEDGEMENTS
W
e would like to acknowledge the support of
the UK Engineering and Physical Sciences
Research Council (Grant GR/M66233/01). We also
wish to thank Helen Ashby, Ann Borda, Alice Grant,
Sarah Norville, Charlotte Stone and other staff from
the National Museum of Science and Industry for
their assistance, together with the J. Paul Getty Trust
for provision of the AAT, and mda and CHIN staff
for helpful advice.
28
CERES Thesaurus
Protocol and browser,
http://ceres.ca.gov/
thesaurus/
29
ADL Thesaurus Protocol,
http://www.alexandria.ucsb.
edu/thesaurus/protocol
30
Binding, C. and Tudhope,
D. 2004: KOS at your
Service: Programmatic
Access to Knowledge
Organisation Systems. In:
Journal of Digital Information,
Vol 4, Issue 4, http://jodi.
ecs.soton.ac.uk/Articles/
v04/i04/Binding/
31
SWAD-Europe Thesaurus
Activity: SKOS API,
http://www.w3.org/2001/
sw/Europe/reports/thes/
skosapi.html
32
NKOS Website, http://
nkos.slis.kent.edu/
33
Tudhope, D.: ECDL
2003 Workshop Report:
Networked Knowledge
Organization Systems/
Services (NKOS): Evolving
Standards, In: D-Lib
Magazine, September 2003,
http://dlib.ejournal.ascc.net/
dlib/september03/09inbrief.
html#TUDHOPE