background image
option was to use giant CD juke boxes,
which had the advantage that they could
be used for normal music CDs as well as
CDRs. Soon, however, we concluded that
in the long run the only feasible alternative
would be a digital mass storage device in
practice, a tape robot.The original `pro-
duction quality' sound would be stored on
the tape robot and the system would auto-
matically create a bit-reduced `listening
copy' of all sound files for quick browsing.
e also decided that the system
should be able to handle all of the
audio formats used in our company. All
programmes are archived in the format in
which they are created, but the system will
create conversions required by various users.
t this stage we had to decide what
our priorities were: what we wanted
our system to do.We wanted to integrate
the digital archive into our existing cata-
logue database, which had been recently
revised and we were fairly satisfied with it.
This was a critical decision, because we
found out that in practice this excluded
certain potential suppliers who wanted to
sell us their own database model as part of
their archive solution. Using the existing
database would guarantee that we could
keep information about the old analogue
collection in the same system as new digi-
tal material and effectively control the
digitisation process.
- Tape robot (digital mass storage) with
100 TB capacity
- Integrated with existing catalogue
databases (RARK, FONO)
- All sound files duplicated (archival
copy and browsing copy)
t was equally important to decide what
the archive system should not do.The
system was not planned for the capturing
done. At the YLE radio archives we have
today more than 50,000 digital sound car-
riers of various types (not including com-
mercial CDs) from the 1990s, and they
cause us more problems than our old ana-
logue tapes.
y the mid-1990s, these digital carriers
were gradually replaced by `tapeless
recording', recording onto computer hard
disc.This was introduced with the devel-
opment of computer-assisted radio (CAR),
where programme planning, production
and transmission were integrated.
However, CAR systems had very limited
archiving capacity.To preserve the pro-
grammes permanently, we had to copy the
sound files on CDs or DATs and retype
the content information to our database.
This was obviously not economical.
Something had to be done.
- 250,000 deteriorating analogue tapes
- 50,000 unstable digital carriers
- `Tapeless recording', born-digital
sound files
The solution
The main goals of our planned
archive system were:
- the possibility of archiving `born-digital'
material directly from production
systems (CAR systems), and using meta-
data created in the production process as
the basis of the catalogue database;
- the digitisation of existing sound carriers;
- to make the archives available online to
all producers in the company and
improve the service and effectiveness of
the archives;
- permanent storage (`forever').
hen we started studying new
archival solutions in the mid-
1990s, we considered various options. One
ith the computerised catalogue, we
could have managed for a long
time, if there had not been the problem
inherent in our archival medium: magnetic
tape. Professional-quality audiotape has a
maximum life expectancy of about 50
years, in practice often less. By the 1990s
we had an increasing number of old tapes
approaching the end of their lifetimes.
These tapes had to be copied onto a new
medium if they were to be saved.
rom an archival point of view, the eas-
iest solution would have been to copy
them onto a new analogue tape. Although
analogue copying always results in a loss of
sound quality, with professional equipment
the loss is so small that acceptable standards
could have been maintained for a few hun-
dred years more with analogue tape, just by
copying our archives every 50 years.
owever, digital recording technology
was introduced at this time. It was so
much cheaper, and in some respects better,
than the old technology that analogue tape
more or less disappeared within a period
of 5 to 10 years. Suddenly all radio pro-
duction was done on digital media such as
DAT tape, recordable compact discs and
minidisks. At YLE, we do not even have
analogue tape recorders in our studios any
more, therefore we have to copy all
archival tapes onto new media when they
are used in broadcasts.
n many ways, the 1990s were a night-
mare for the radio archivist. It was soon
discovered that the new digital carriers
were not reliable archival media. Although
they could, under ideal conditions, preserve
high-quality sound for a long time, their
preservation demanded such a high degree
of quality control that we found it impos-
sible to maintain. And although a digital
copy is in theory identical to the original,
this does not help much if your DAT or
CD player is unable to reproduce the orig-
inal. But there was little that could be