background image
DigiCULT 63
lighted the fact that standardised preservation moni-
toring software would be needed, which could also be
achieved around 2008. However, he mentioned that in
an RTD proposal a work package to create a carrier
monitoring system was taken out because the review-
ers regarded this as not innovative. He added: `At the
Germanische Nationalmuseum, a brand-new RAID
server crashed completely, even though it has a the-
oretical lifetime of some 10,000 years. (In Prague,
whole libraries were flooded. Not to speak about Iraq)'
Reagan Moore (Director, Data Intensive Comput-
ing Environments, San Diego Supercomputer Center,
USA) wrote that preservation environments up to the
demanding tasks of long-term digital preservation and
access would need a support community for sustaina-
bility (funding, domain expertise). As a major require-
ment, however, he added: `For sustainable preservation
environments, the collections must be shown to pro-
vide information and knowledge that would other-
wise be lost. This means that the preserved collection
must be capable of supporting knowledge generation/
data mining to identify those items that address today's
questions. This in turn means that the preserved col-
lection is useful if it can be processed, new organi-
zational structures can be applied, and discovery can
be supported for purposes different than the original
guiding preservation policies.'
Regarding the technological set-up, Moore suggest-
ed building on `data grid technology, which provides
the infrastructure needed to manage technology evo-
lution, and automate processing of collections', as well
as being able to `support the long-term data manage-
ment'. Major current RTD limitations or gaps he saw
in `the equivalent of a digital ontology... that charac-
terizes the structure of a digital entity and defines the
operations that can be performed on the structures.
We then need a display ontology that characterizes the
operations needed for display. Preservation (emulation)
then consists of mapping from the operations that can
be supported on the digital entities to the operations
required for display. Migration is transformation of the
ontologies onto new relationship encoding formats
(such as RDF).'
Considering these RTD issues, Moore described
some major steps or breakthroughs required to close
existing gaps as follows: `The challenge is that the
encoding formats for text, spreadsheets, slides are quite
complex. We need better languages to describe order-
ing of operations that are needed to interpret bits,
transform to ascii, transform to text structures, and
transform to display mechanisms. The goal is generic
software that supports self-describing objects in terms
of their internal structures and the allowed operations
for manipulating those structures.
Another major step is the development of con-
straint-based data management systems. This is the rec-
ognition that the consistency constraints imposed for
authenticity, the access control constraints, the viewing
transformation constraints, will also evolve over time.
We need to be able to dynamically apply new con-
straints to a preserved collection, as well as to be able
to define the constraints under which the collection
was originally created.' The timeframes Moore gave
for when this is likely to be achieved are included in
the table.
Selected projects and publications
ERPANET... the Electronic Resource Preser-
vation and Access Network, has established an
expandable European Consortium, which makes
viable and visible information, best practice and
skills development in the area of digital preserva-
tion of cultural heritage and scientific objects. The
project also serves as a virtual clearinghouse and
knowledge-base in this area.
InterPARES... the International Research on Per-
manent Authentic Records in Electronic Systems,
aims at developing the theoretical and methodo-
logical knowledge essential to the long-term pre-
servation of authentic records created and/or
maintained in digital form.
A practical and information-rich online tuto-
rial... Digital Preservation Management: Implementing
Short-term Strategies for Long-term Problems. Cornell
University Library/Research Department (2004),
DPC/PADI What`s new in digital preservation..., a
joint service of the Digital Preservation Coalition
(DPC) and Preserving Access to Digital Informa-
tion (PADI) that provides a quarterly summary of
selected recent activity in the field of digital pre-
servation compiled from several sources.
A research agenda for digital archiving and long-
term preservation... Invest to Save. Report and
Recommendations of the NSF-DELOS Wor-
king Group on Digital Archiving and Preservation
(2003), Margaret Hedstrom and Seamus Ross et
DCTHI7_271104.indd 63
06.12.2004 8:38:35 Uhr