LaSSIE: a Knowledge-based Software Information System

Abstract

Invisibility is an inherent and signiicant problem in the task of developing large software systems. There are no direct solutions to this problem; however, there are several categories of systems|relational code analyzers, reuse librarians , and project management databases{ that can be seen as addressing aspects of the invisibility problem. We argue that these systems do not adequately deal with certain important aspects of the problem, namely, semantic proliferation, multiple views, and the need for intelligent indexing. We have built a system called LaSSIE, which uses knowledge representation and reasoning mechanisms to directly address each of these issues. LaSSIE provides a knowledge base explicitly representing a large software system, and an interactive interface to give programmers direct access to the software through several semantically-based views. LaSSIE is limited in a number of ways, but it illustrates how representation and reasoning mechanisms can be used to alleviate invisibility and complexity.

DOI: 10.1145/103167.103172

Extracted Key Phrases

1 Figure or Table

Statistics

02040'93'96'99'02'05'08'11'14'17
Citations per Year

461 Citations

Semantic Scholar estimates that this publication has 461 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@inproceedings{Devanbu1990LaSSIEAK, title={LaSSIE: a Knowledge-based Software Information System}, author={Premkumar T. Devanbu and Ronald J. Brachman and Peter G. Selfridge and Bruce W. Ballard}, booktitle={ICSE}, year={1990} }