The primary goal of the Alexandria Project is to construct a distributed
digital library (DL) for spatially-indexed materials
.
The Alexandria Digital Library (ADL) will comprise a set of nodes
distributed over the Internet,
with each node supporting library components
that include collections, catalogs, interfaces, and ingest facilities
.
Key aspects of the ADL development strategy include 1) a design focused on digitally supportable extensions to traditional library functionality and consistent with requirements of the library community; 2) a distributed library that is accessible by many classes of users from the Internet; 3) user access to both the explicit and implicit information available in DL collections; 4) an evolutionary, reusable, and incremental approach to design and implementation; 5) particular focus on developing the user interface (UI) and catalog components of the DL architecture; 6) initial focus on building collections of spatially-indexed materials. In particular, we believe that basing ADL on the Internet and following an evolutionary and incremental path of development is the most appropriate given the current rate of technical development.
A critical goal for the Alexandria Project is the design and implementation of UI's that support general DL functionality and a catalog component that supports the full range of user queries entered into the UI. An important subgoal relating to the UI is the provision of library services to users with a variety of backgrounds and needs. A school child, for example, may wish to use ADL to find a map that shows nearby rivers and trails for a camping trip; a scientist may wish to find datasets of elevation and rainfall for a study area in order to develop a vegetation model. Different users need different ways to formulate search queries and have different expectations of, and requirements for, search results.
An initial, and now completed, increment in the development of ADL involved
the design and construction
of a stand-alone ``rapid prototype'' (RP) system [5]
.
The second increment involves providing an
augmented version of the functionality
of the RP over the World-Wide-Web (WWW).
We term this the ``WWW prototype'' (WP)
.
The third increment will focus on developing
a catalog component that is based on a general model of metadata.
In the current paper, we focus on the resolution
of issues that arise both in
expanding the functionality of the RP
and in extending it to the WWW.
The paper is structured as follows.
We first describe the basic four component architecture
of ADL, as adapted for the WP version.
We next describe our resolution of key issues that arise
for these various components in a WWW environment.
Finally, we describe the application of two key technologies
that have implications for each of the components,
namely wavelet transformations and parallel processing
.