Information Technology Reference
In-Depth Information
Chapter 23
Sharing the Effort
23.1 Chain of Preservation
The concepts introduced here are aimed at providing an infrastructure which can be
shared across the widest range of information. In that way we can spread the cost
and effort of preservation and reduce the risk that the infrastructure upon which we
would depend might itself disappear by broadening its potential support.
In
Chap. 16
those concepts which are discipline independent, and those where
the domain dependence occurs, was described in some detail. This chapter builds
on that discussion and considers a number of practical implementation issues.
We further focus on things to do with Representation Information in that can be
shared between data archives, whereas the Preservation Description Information
(PDI) tends by its nature to be unique to data instances and archives.
Some of the components which we identify, illustrated in Fig.
28.1
could con-
ceivably be large, unique, resources, for example the Registry and the RepInfo Gap
Manager (although issues of scalability would naturally arise). Other components
make no real sense as single resources - for example the Archival Store (i.e. it would
be difficult to advocate keeping all data in a large central data store). Nevertheless it
would make more sense to allow multiple instances of any component, in order to
provide adequate resilience and scalability. For example it would be perfectly rea-
sonable to have multiple independent Registries, some of which may interoperate or
federate.
23.2 Mechanisms for Sharing the Burden of Preservation
The rather generic scenario is as follows:
•
A user uses an application to perform some task - this can range from displaying
a document to performing some scientific analysis to playing a piece of music.
