The Salisbury Project: Collecting an Existing Digital Resource

The Salisbury Project is a large online archive of images designed to
supplement existing scholarship published on the cathedral and town of
Salisbury. It consists of a highly structured, navigable hierarchy of over
500 images, associated descriptions, teaching materials, bibliographies,
and maps. Marked up in SGML, using the Encoded Archival Description DTD,
the archive has made every effort to employ the standards available and has
certainly reached a certain level of platform and display independence.
However, in order to present the materials to the user in the dynamic and
interactive fashion it requires, certain concessions had to be made. The
lack of appropriate standards and supporting software at the time of its
creation, made it necessary to rely on proprietary programs and formats for
its look and feel. This obviously presents problems for the collection and
preservation of such a complex object in the long term. The challenge for
SDS was developing an approach that made it possible to recreate the
Salisbury Archive using methods and tools that could be reused for
subsequent projects. Therefore in this, the first of our collecting
efforts, we decided to take a two-step approach: re-encoding the document
in the GDMS (General Descriptive Modeling Scheme) XML DTD and then using
the newly developed XSLT standard to emulate the dynamic presentation
software used by the original project. The combination of XSLT stylesheets
with transformation engine allows you to model complex dynamic behavior
but unfortunately it will not provide the contextual searching essential to
markup based projects. Fortunately another member of the IATH staff was
working on just this problem.