A characterization various classes of users and an understanding of their requirements is of fundamental importance for the design, development, and testing of the proposed system. Initially, a significant proportion of research resources will be allocated in ascertaining such requirements and ensuring that the design and development of the digital library is based upon a firm understanding of user needs. We will adopt a three-fold strategy in order to determine user requirements: first, we will analyze existing studies of user requirements with respect to spatial data in the context of library systems. One such study (GRIN) was, in fact, carried on at the UCSB Library, and much of what was learned from the construction of the GRIN external design will be used in the Alexandria testbed development. Second, a rapid prototype system will be built using currently available software. This system, and in particular its interface, will serve as an important vehicle in further investigating and evaluating user requirements. Finally, we view the main function of the testbed system that we will design, develop, and test over the greater part of the project, as providing a vehicle for determining whether our characterization of user requirements is met by the system, and indeed whether it is an appropriate characterization.
Components of the user requirements upon which the design and development of our experimental system will be based include the various levels of users and their relative data volume needs; the classes of objects of interest to the various classes of users; the sets of operations that users wish to apply to objects; the nature of the interfaces that are appropriate; and the levels of system performance that must be met. In the early stages of system testing, few patrons will be knowledgeable about how to find spatial information. We expect spatial data user experience to run the gamut from general (users with little or no computational skills), to highly sophisticated (users with substantial knowledge of a given field and who, for example, may will wish to perform complicated scientific modeling on accessed data sets.) Hence, at a minimum, the system must accommodate a range of users, from beginner to sophisticate, by using a multilevel graphical interface; it must be easy to use and not require a manual; and it must be able to search on all data types. The system will provide a collection of interface options including: Z39.50 text search for persons who prefer the command-line approach; non-spatial graphic-interface searches for others, e.g., persons dealing with non-political areas; and full graphical spatial searching, utilizing polygon overlay on background maps, for users dealing with complex multipoint geographic areas. During the design, development and assessment of the main testbed system, we plan to continue focusing a significant proportion of research activities on the issues of user characterizations and requirements. Types of assessment questions will be: Does the prototype give you the information you need? Does it get you the information easily and quickly? Does it give you information you do not need? Are the error messages clear?