International Workshop on Database Programming Languages. 1993, -
The major motivation for database programming language (DBPL) research is to facilitate the construction and maintenance of large dataintensive applications. To fully bene t from DBPLs, supporting methodologies and tools are needed. This paper reports requirements for such methodologies and tools that were experienced when constructing a multiauthored, multi-levelled Thesaurus Application (TA) in a higher order DBPL. Although built in a speci c language (Napier88), the major principles discovered apply to other DBPLs.
The TA comprises several looselyintegrated components constructed by di erent programmers. The components were themselves implemented using general purpose sub-components, including libraries. We experienced that a realistic application could be constructed quickly in a DBPL.
Rapid construction was facilitated by the use of libraries, code reuse and an incremental construction methodology supported by the persistent store. Language features such as a polymorphic type system and structural type equivalence were important. Nevertheless, some problems were encountered with code reuse, with integrating independently constructed components, with the lack of concurrency and with build management such as installation and recompilation. Our experiences lead us to suggest several improvements, including models, methodologies and supporting tools for persistent application construction and maintenance.