background image
DigiCULT 9
data (LOM).
10
One major success in the enhance-
ment of heritage resource discovery is the strong
uptake of the Open Archives Initiative Protocol for
Metadata Harvesting (OAI-PMH) in the sector.
11
This includes subject gateways
12
or resource discov-
ery networks such as the UK RDN;
13
large players
such as the Library of Congress (e.g. for the American
Memory Collection),
14
as well as highly specialised
communities such as institutions and scholars work-
ing on Renaissance emblem literature.
15
Also Minerva
Europe, the Europe-wide co-ordination initiative in
the area of heritage digitisation, shows a clear inter-
est in the OAI-PMH for digital resource discovery. It
may well become a basis for a future European cul-
tural heritage portal.
16
A major reason for the success of OAI-PMH rests
on its mandatory minimal requirement of unquali-
fied Dublin Core (DC). However, this also leads to a
downside. DC has been designed to support simple
cross-domain resource discovery. When rich metada-
ta is cross-walked to DC, the hierarchical structure of
the original encoding is lost, i.e. its full richness can-
not be expressed. But, overall, OAI-PMH has shown
itself to be an efficient and effective way of metadata
exposure and exchange.
R
ESOURCE
D
ISCOVERY
: W
HAT
M
AKES
IT
A
C
OMPLEX
P
ROCESS
?
R
esource discovery involves the searching, locat-
ing and retrieving of information resources on
computer-based networks, in response to queries of
a human user or an automated mechanism. It also
involves presenting the information to the user in the
most appropriate form, and the capability to manage
the found resources at either the find or the retrieve
level. Therefore, the discovery process should support
a query and browsing interface as well as other meth-
ods of user interaction (e.g. for personalisation).
The resource discovery requirement may be tran-
sient or may be based on a more permanent system
that regularly notifies the user about relevant infor-
mation. Advanced systems will utilise profile infor-
mation on the user in order to provide the most
relevant information efficiently. However, ideally a
system should also allow for serendipity, i.e. discover-
ing valuable resources the users were not looking for
intentionally.
17
Interactive resource discovery is a complex proc-
ess that includes multiple phases, is iterative and high-
ly dynamic. In particular, it is an intellectual process,
which involves (re-)formulating queries and examin-
ing candidate resources. It is also a learning process, in
which users will successively develop and refine crite-
ria of information granularity as well as their under-
standing of domain specific terminology, concepts and
discourse. In this respect, the resource seeker can be
compared to a traveller, who in communicating with
local residents may use a phrase book, but if he does
not learn their language he will not be capable of
`navigating' properly within their culture.
18
I
T
'
S
THE
U
SER
, S
TUPID
!
M
ost of the time we are all fast-moving electron-
ic travellers who use simple phrase books, and
want useful information as quickly as possible. This
is why almost nobody less than 0.5 per cent uses
advanced search options; why most people want to
have a simple entry field into which they will type on
average 1.3 words; and why most people will not look
at more than one page of result list.
In consequence, general search providers will need
to concentrate most of their energy on doing as good
a job as possible with little input from the users, hid-
ing compound queries behind point-and-click, and
fitting more information on to search-result screens.
19
In consequence also, nobody should expect heritage
organisations to compete (in any sense of the word)
with powerful search engines such as Google. Let the
problem of large numbers of hits, information sources
of doubtful relevance, and no guarantee of authentic-
ity and reliability be the problem of the fast-moving
e-traveller.
The appreciation of highly valuable structured
heritage information resources as well as elaborat-
ed concepts and different perspectives of the Arts &
10
IEEE Learning Object
Metadata (LOM),
http://ltsc.ieee.org/wg12/
11
OAI, http://www.
openarchives.org; for an
overview on the use of
the OAI-PMH in the
heritage sector see Muriel
Foulonneau (ed.): Open
Archives Initiative Protocol
For Metadata Harvesting.
Practices of cultural heritage
actors, September 2003,
http://www.oaforum.org/
otherfiles/oaf_d48_cser3_
foullonneau.pdf
12
The DigiCULT
Website's `Resources' sec-
tion also features Gateways
to Cultural Heritage
Resources, currently over
70 annotated links: http://
www.digicult.info/pages/
resources.php
13
The UK RDN is a
collaboration of over 70
educational and research
organisations, including the
Natural History Museum
and the British Library,
http://www.rdn.ac.uk
14
The Library of
Congress's experience as
an early adopter of the
OAI-PMH is described
by C. R. Arms (2003):
Available and Useful: OAI
at the Library of Congress,
http://memory.loc.gov:
8081/ammem/techdocs/
libht2003.html
15
See the DigiCULT
special publication:
Digital Collections and the
Management of Knowledge:
Renaissance Emblem
Literature as a Case Study
for the Digitization of Rare
Texts and Images. February
2004, http://www.digicult.
info/pages/special.php
16
Cf. Foulonneau 2003
(footnote 11), pp. 43-44.
Footnotes 17-19 continued
on page 10.