Best Practices For Audio Preservation
There are several implications for audio preservation system design and analysis that can be
derived from these examples. In general, to use two clichés: if you put garbage in you will get
garbage out, and the system is only as strong as its weakest component. More speciﬁcally,
quality control for each element of the system is necessary to assure that each part contributes
optimally to the whole. And, an analysis of what system parts have the greatest impact on the
output of the system can be valuable for the allocation of resources. In other words, if funds
are limited, where should they be concentrated for maximum beneﬁt to the ﬁnal product?
Are they so limited that preservation-worthy products are not possible? Analysis of questions
like these may affect system design, the allocation of resources, and the decision on whether
to use vendors for all or part of preservation activity.
The ﬂip side of this principle is that the system has holistic properties not manifested by any
of its parts and the whole may be greater than the sum of its parts. Something distinctive
emerges when parts interact. The emergent product may have properties or qualities that do
not exist within, or are unique compared to, any one part or partial combination of parts.
What emerges from our audio preservation system is a preservation package containing
prioritized content residing in a robust storage system, representing the strongest prospect
for long-term survival of the content. This stored package, including its on-going value as the
best carrier for prioritized content, is not inherent in, and cannot be produced from, any one
system element or any partial combination of elements; every individual part of our audio
preservation system is needed to produce this speciﬁc result.
Division of Labor
In a complex system, specialized units perform specialized functions. Audio preservation
systems in the digital age necessarily require greater specialization, and more collaboration
among specialists, than their earlier analog counterparts and ours is no exception. An effective
preservation system in the analog domain—back when open reel tape served as the target
preservation format—might involve only an archivist, an audio engineer, and a few others.
The current IU audio preservation system requires the contributions of archivists, librarians,
audio engineers, digital librarians, IT specialists, computer programmers, intellectual property
experts, content specialists, and others who work collaboratively across disciplines and
administrative units at IU. This is a relatively more complex audio preservation system—in
the sense that it relies on a greater number of specialists—than existed at IU in the 1980’s
and 90’s. It is also true, however, that the current IU audio preservation system is designed
to provide greater functionality and efﬁciencies than the older analog system. The level of
functionality that the OAIS model demands, for example, was never realized during the days
of analog preservation. Consequently, we are reaping additional beneﬁts from the type of
system that is necessary in the digital domain.
Elements of a system must be regulated so that system goals are realized. This involves
feedback and the detection and correction of deviations. Regulation is achieved in any
system, including those built for audio preservation, through quality control and quality
assurance. We have learned that these must be part of nearly every system element. Our
approach to QC/QA is still evolving, but we have attempted to build these functions into
much of our system.
117 We think that, as discussed in earlier sections, in terms of audio ﬁdelity, expert staff, analog playback equipment
and expertise, and analog-to-digital conversion probably have the greatest impact on the ﬁnal product. Of course,
all steps along the preservation chain are important and many can impact ﬁdelity. In addition, other stages such as
appropriate storage and metadata are obviously critical to the long-term viability of the content.