ELICITATION, REPRESENTATION AND MANAGEMENT OF SOFTWARE DESIGN KNOWLEDGE release_qv3df5qfvvfylmlg6ojtprcq3a

by Behrouz H. Far, Mohsen Afsharchi

Published in Proceedings of the Canadian Engineering Education Association (CEEA) by Queen's University Library.

2011  

Abstract

In this research we focus on understanding the nature of the knowledge used during the various phases of the software development process. We have found that there are two types of knowledge involved in software development: (1) descriptive knowledge represented by conversion and coding rules, e.g., a rule for splitting a class into two; and (2) prescriptive knowledge required for deployment of global or local strategies at a micro design level; e.g., knowledge required to answer the question "why should a class should be split into two?" Most of the already existing knowledge management solutions address descriptive knowledge. Elicitation and management of the prescriptive knowledge is difficult in the sense that it is probabilistic, personalized, distributed and context specific. Also we have found that prescriptive knowledge tends to be used in decision making processes involving multiple stakeholders with different perspectives (e.g., designer, tester, software architect and project manager). We also report on a prototype system called ISS-OKM to extract and reuse both the descriptive and prescriptive knowledge.
In application/xml+jats format

Archived Files and Locations

application/pdf  3.9 MB
file_pbsidnnwunfqvcisg2a5iixgwi
cv.znu.ac.ir (web)
web.archive.org (webarchive)
application/pdf  4.3 MB
file_5phochpeyfbadejgyoc7xav6jm
ojs.library.queensu.ca (web)
web.archive.org (webarchive)
Read Archived PDF
Preserved and Accessible
Type  article-journal
Stage   published
Date   2011-08-15
Proceedings Metadata
Open Access Publication
Not in DOAJ
In ISSN ROAD
Not in Keepers Registry
ISSN-L:  2371-5243
Work Entity
access all versions, variants, and formats of this works (eg, pre-prints)
Catalog Record
Revision: 4721cffe-8e5f-4846-9ee7-8f5323f292f5
API URL: JSON