Forum
Future R&D for Digital Asset Preservation
DPC Forum with Industry
5th June, 2002. Prospect House York Road London SE1 7AW
Since its inception the DPC has aimed to gain industry awareness of
its key messages and of the future needs and opportunities that lie ahead.
This forum is part of that process. During the day representatives from
the private and public sector will be speaking. They will address long-term
trends and the research and development issues involved in the implementation
of continuing access and preservation strategies by industry and government.
Issues covered will include preserving TV and broadcast archives and
research and development in the public and private sector.
Meeting Report
Introducing the first Digital Preservation Coalition (DPC) Forum focussed
on developing a dialogue with Industry, DPC Secretary Neil Beagrie welcomed
guests and members of the DPC. He then placed the question of
digital preservation and opportunities for industry participants, firmly
within an international context. The US National Digital Information
Infrastructure and Preservation Program (NDIIPP): a $175m national programme;
the National Science Foundation Cyberinfrastructure initiative; the Information
Society Technologies programme of the EU 6th Framework; and,
in the UK, the Research Grid and the work of the DPC, will be central
to bringing political, technical and organisational impetus to bear on
the challenges of digital preservation. With accelerating development
of digital content, the issue of maintaining long-term access will
be of concern to an increasing ranging of sectors and individuals.
Forging strategic alliances with industry in the context of preservation
was a vital component of any initiative in this area, said Neil Beagrie.
He outlined how the DPC's constitution was shaping its emerging links
with industry. Two important principles governed the DPC's work: firstly,
the DPC will support the development of standards and generic approaches
to digital preservation, which can be implemented by a range of hardware,
software and service vendors; in short, he continued, the goals of the
DPC are vendor-neutral. The second principle is that the DPC is a coalition
of not-for-profit organisations including industry associations; it is
committed to promoting and disseminating information so that all can
learn from the transferable lessons and outcomes. The DPC is actively
interested in broadening its industry links including those to individual
companies. There is a major potential role for industry, concluded Mr
Beagrie; this event therefore marked an important step in realising that
potential.
Philip Lord, formerly of SmithKlineGlaxo, and now a Digital Archiving
consultant, spoke of his own experience in industry and of some of the
major drivers for industry in the field of digital preservation. He particularly
emphasised the importance of the US Federal Drug Agency (FDA)'s regulation 21CFR Part 11 on the
maintenance of good electronic records for the pharmaceutical industry.
Legal and regulatory issues were, he said, clearly of major importance,
as were the various voluntary drivers, such as contractual and IP obligations,
operational efficiency considerations and the need to preserve for future
reuse. Rarely were records preserved, he said, for historical or sentimental
reasons.
Mr Lord talked of the many challenges that faced industry in this area:
the heterogeneity of data sources and systems, geographical dispersion
(spanning legal and regulatory jurisdictions), the lack of suitable preservation
systems and services, management issues, cost, and the lack of expertise
in this area. Mr Lord reported that little progress had been made so
far within the sector, although a few companies, he reported, were leading
the way.
David Ryan from the Public Records Office reported that digital
preservation was a core part of the PRO's e-business strategy. Research
and development in this area was focussing, he said, on proprietary format
migration, emulations and simulation, open format export and migration
and product reviews. The importance to users of interaction with
preserved digital information, rather than simply using that information
as an historical record (as is more likely the case with printed data)
was made, and this gave an added importance to the work of the PRO in
its e-preservation activities. Mr Ryan concluded by saying that there
was an overriding need for the archives community to establish credibility
with the key stakeholders in e-preservation, including industry, the
media and the general public. The community as a whole needed to prioritise
so that the most urgent tasks and challenges were tackled first.
The film and sound archives of the BBC contain some 1.75m film and videotape
items and around 800,000 radio recordings from the late 1940s onwards.
A ten-year preservation strategy had, Adrian Williams from the
BBC reported, recently been approved and would cost around £60m.
A key part of this strategy was a programme of digitisation for both
access and preservation. He reported on the European Commission Presto
project, which involved 10 partners, lasted 24 months and cost around
4.8 million Euros. Findings from this project suggested, for example,
that digitisation and mass storage is about 50% more expensive, but is
expected to double usage of an asset; and moreover, that the value of
an item must be four times the preservation cost to be financially viable.
He concluded by suggesting that Europe requires a "dedicated preservation
factory" given the scale of the task facing national broadcast archives.
There was substantial audience interest in the approaches to cost and
business modelling described in the Presto project.
Julian Jackson noted in the headlong rush to put photographic
images into digital form, little thought seemed to have been given to
the problem of the longevity of digital files. There is an assumption
that they will be lasting, but that is now under question. He addressed
general issues surrounding preservation and obsolescence in digital images.
He surveyed the techniques of refreshing, migration and emulation and
emphasised the crucial role that metadata and metadata standards have
to play in these preservation processes.
Paul Wheatley of the CAMiLEON project spoke of some of the practicalities
of digital preservation and emphasised the need for long-term strategies.Existing
methods have many drawbacks. Mr Wheatley described advanced techniques
of data migration which can be used to support preservation more accurately
and cost effectively.
To ensure that preserve works can be rendered on computer systems over
time "traditional migration" has been used to convert data into current
formats. As the existing format becomes obsolete, another conversion
is performed, etc. Traditional migration has many inherent problems
as errors during transformation propagate through future transformations. Mr
Wheatley described how the Camileon project had developed new approaches
to extending software longevity ("C--") which had been applied in experiments
and demonstrated improvements over traditional migration. This
new approach is named "Migration on Request".
Migration on requesting shifts the burden of preservation onto a single
tool, which is maintained overtime. Always returning to the original
format enables potential errors to be significantly reduced. Mr
Wheatley also described how preservation quality emulators were being
produced and strategies of migration on request and/or emulation were
being applied.
The need for public-private partnerships in the field of digital preservation
is crucial, said David Bowen of Audata Ltd. He went on to outline
what industry was currently doing in this field - - e-mail, document
and database migration, as well as promoting standards, while software
suppliers are also improving backward compatibility (Word, Wordperfect,
RTF), and increasingly adopting and promoting standards themselves too.
Mr Bowen called for R &
D partnerships, like the Testbed Digitale Bewaring in the Netherlands,
which is leading to the sharing of results and advice, and sound record
creation and metadata practices. Particularly important, he concluded,
was the need for software suppliers to be brought into the growing public-private
partnerships that are developing.
The final session of the day was a discussion session. Key themes that
emerged included:
- the importance of archiving software and technical documentation. It
was felt by participants from all sectors that this is a major gap
and there was an urgent need to develop appropriate repositories;
- the need to develop case studies and tools for modelling costs. It
was felt this is a major area that should be covered in future DPC
forum;
- the necessity of developing national funding for the preservation
of intangible heritage assets. It was noted there is no "Superfund" or
legislation which allows digital heritage to be gifted in lieu
of tax to or purchased by, the nation;
- further work by the Digital Preservation Coalition to establish
contacts with industry and to build on the dialogue commenced at the
forum.
It was felt that it was important, as David Bowen had said, to include
software and hardware suppliers in future developments as their actions
could be crucial, in particular in providing the tools and products for
end-to-end solutions which were needed in this area. Once again the importance
of using both migration and emulation strategies was emphasised, as was
the question of the criteria for choosing what should be preserved; we
are not in a position to judge easily what will be in demand in the future.
Therefore sampling could be crucial importance as one strand in our overall
strategy.
Some delegates from industry felt that there were gaps of responsibility
between the organisations, and that it was therefore important for the
DPC to coordinate and facilitate activities in this area.
The forum ended on a note of optimism that the first steps in the dialogue
with industry had been taken, and with a number of concrete action points
which Lynn Brindley, Chair of the DPC, promised would be followed up
in the coming months.
End of Meeting Report
Programme and Presentations
|
