background image
least the next ten years. As with many other
institutions, political changes can affect the
amount of investment that we can rely upon
as the priorities of funders can change. In
terms of research, we have launched a project
this year to develop technology specifically to
accelerate the digitisation process and make
it cheaper. This work has a tremendous value
for institutions at all levels within a country;
for example, an organisation like the BBC can
get money to digitise its materials, but small-
er institutions have less reliable funds, even
though some of their materials are equally
important. Who will protect these collections?
Ideally, our research will provide a solution to
the problem of the great expense of undertak-
ing digitisation activities.
hese steps are INA's approach to
potential obsolescence of materials in
the future. In my opinion it is useless for us
to wait for the `perfect container' for audi-
ovisual materials, as we will never have a
perfect storage format. In fact, the best one
is the first one: film! If we waited for the
perfect medium, data would be lost; instead
it is evident that action must be taken now
to prevent these cultural materials from
becoming inaccessible. The only solution is
to accept the fact that any medium is not
eternal (although you should still choose
the best format available) and to build a
good migration plan. It is imperative to realise
that the technologies will also change. This
may well be the last opportunity that we
have to digitise this particular item so it
must be performed at the best feasible qual-
ity with no modifications. One must also
understand the value of the material and its
uses in the future to justify decisions on the
quality of digitisation. For example, the data
needed by digital video are 50 Mb per sec-
ond, whereas our master copies are only 8
Mb/s. Experts agree that at least 200 Mb/
s is needed to replicate the quality of cine
film although 50 Mb/s is still an excel-
lent digital quality. Balancing the quality
requirements of the materials with the time
and expense of creating digital versions is
a difficult area. Once an unmodified mas-
ter has been produced at the highest pos-
sible quality, then temporary versions can
be created from it, for example, applying
digital clean-up algorithms on an old tel-
evision programme. Yet another factor to
consider is that users' demands will change
as the years go by what is considered as
a natural requirement now may not always
be the case. For example, we supplied some
moving image material to users after per-
forming image restoration, but there were
complaints about the quality of the sound
attached to the images. Conversely, users
were generally much happier if the sound
had been `cleaned' but the images were
unmodified. This unexpected result shows
that user needs are not always what we now
consider them to be!
ur most recent major programme
began around two years ago our
aim is to collect materials based around a
specific theme, for instance, one person. We
then attempt to digitise a comprehensive
(or at least representative) selection of these
programmes to provide a detailed and var-
ied view of the subject in various formats
and styles. INA customers tend to be very
demanding, each requiring material that is
not only high quality but also original, i.e.
has not been used already by a competi-
tor. We hope that, by focusing on specif-
ic themes, we can provide material that is
suitable for the purposes of a large number
of our customers and will increase the
choice they have.
s a participant of the information
society, I have many tools to support
my information needs. I can access mil-
lions of examples of written material in the
modern `global library'; however, there is
comparatively little free access to audiovis-
ual material. Digitisation will provide this
access in the future and make film, televi-
sion and radio programmes as ubiquitously
available as the written word. My aim is to
encourage work and research on archival
materials and to catalogue the future of this
sector. INA is involved in many forward-
looking projects, from tracking user issues
and the ways with which user interfaces are
interacted, to the most important issue for
`It is useless to wait for the
`perfect' storage format
we will never have it!'




Main entrance to Maison de Radio France, Paris