Moreover, answers to these questions were becoming urgent since media were
decaying and actions had to be planned to ensure that contents would be still
accessible in the future. For the first question, there existed no definite answer;
little by little the concept of an ideal digital media was abandoned and the
operative concept of migration began to make its way.6
The second question was a very tricky one, since the choice of the format had a
direct relation to the quality of the digital preservation. The intention is that the
digital copy of an analog program will have the same quality as the original;
however the equivalent copy uses a huge amount of digital storage space, which
is an expensive item, mainly when it comes to store thousands of hours of
moving images. Compression formats reduce the digital storage space; a very
high-definition compression-rate as used in the Digital Betacam consumes about
85 megabits per second7 (the necessary rate for an equal definition would be
around 200 megabits per second, and even this rate is not widely accepted for
cinema purposes)In the broadcast domain, the production and delivery
standard is MPEG 2, with a rate of 8 to 20 megabits per second. There is a strong
tendency orienting production towards the MPEG 4 format, which improves
quality with lower resolution, diminishing the necessary storage space. Finally
the browsing quality, which permits one to access directly a hard-disk and view
very quickly any image, ranges from 500 kilobytes to 1 megabyte.
All these questions are directly related to cost, and cost is related to technology
and to the cost of a certain technology at a given moment. Migration has a cost
that has to be calculated through time, it may fluctuate if, for example, in the
next years a media appears with a longer life span warranty. Digital formats are a
big issue today since storage space is an expensive part of an archiving process;
but these costs may lower very sharply within few years and the discussion about
digital rates may become less important in the future. But there are not only cost
issues. Since archiving is placed at the end of the production cycle, when
production becomes completely digital, an archive has to archive what
production produces, so if formats change, and if SD (standard definition
television) becomes HD (high definition, with more scan lines per screen), then
an archive has to be prepared to deal with these new items.
However decisions have to be made at a certain point and here comes the third
question archives owners ask. How to calculate the cost of a preservation
process, of a large storage system, of a migration plan and what are the benefits
of such an operation? Last but not least, how to do it? Where should tapes be
sent, how to organise the process, how to select the more urgent programmes to
be preserved, how to restore degraded contents, how to index, which browsing-
software to use, how to select or annotate documents, etc., etc?
Solutions exist for every question, but most answers are partial and often derived
from requirements of other activity sectors and not only addressed to audio
visual archives. No specific and integrated proposal has been developed up to
date. Archives have developed, generally independently, their own experience
and launched, when possible, preservation and digitisation programmes, trying
to comply with all the technology and method gaps implied in the process.
Specific actions towards audiovisual archives
Since the early eighties the European Commission has launched ambitious
Technological Research programmes to induce cooperation between different
countries and develop the necessary technology in some identified domains.
Cultural heritage has been one of the chosen domains, mainly under the
perspective of its preservation and accessibility for the citizen.8 Many countries
have also developed their own National Research Programmes in order to
enhance research in different domains. The global objective of both kinds of
initiatives is to improve Europe's position in the technology domain and to
ensure that the results will be applied to industry and services.
No particular action existed in the research domain concerning audiovisual
archiving problems, so the cultural heritage interest within the Information
Society Technology (1ST) section permitted the concept of projects that would be
archive oriented, bringing some partial answers to the numerous questions
arising from the digital society. The first actions concerned a specific domain
within archiving involving restoration. Archives often preserve old programmes
that may have suffered technically through time; scratches, dust and instability
are common defaults found in film as well as other disturbances found in video
like colour fading, flickering and noise. For production use it is important that
these time disturbances can be corrected, especially when broadcast technology
improves image quality, making defects in images all the more obvious.
When the first restoration projects were launched, there already were tools that
permitted the correction of specific defaults, but the main objective of the new
projects was to permit an integrated and simplified approach to restoration.
Integrated, because several restoration tools were put together in the same
system; and simplified, because they were archive restoration oriented and
responding to problems often found by operators and implemented in a practical
mode. Projects like Aurora, Brava and Diamant were developed since 1996 and
have brought efficient answers to the restoration problems with new restoration
algorithms and effective systems for Tape to tape restoration or Disk to disk
restoration.
What is difficult in preservation?
If a transversal analysis had been made in 2000 concerning the existing
technology and facilities for preservation of analog media, the conclusion would
have been that the technology already existed, and so did the service providers
that could put it into action. However, the reality was that very few archives had
launched at that time important preservation or digitisation programmes. Two
major impediments prevented action: the technology choice already discussed,
concerning storage systems, formats and rates, but mainly the high financial
BEHOUD
6 The migration concept implies that you choose a digital format and system on which to encode your pro
grammes and you plan transfers of your contents to new systems according to their life expectancy. A
normal migration frequency today is every four or five years.
7 http://www.sony.com.sg/pro/bp/Digitalbetacam/
58
DANIEL TERUGGI DO WE NEED A PROJECT LIKE PRESTOSPACE?
8 http://fp6.cordis.lu/fp6/home.cfm
59
