jon@june.cs.washington.edu (Jon Jacky) (11/09/88)
I keep seeing mention of a technique for estimating effort, cost, or errors in software projects based on something called "function point analysis". I don't know much about it, but it is apparently based on assessing the functionality described in the specification, which seems obviously more useful than measures based on the size or other measurers of the code (which you don't really know until you're almost done with the project). Although I've seen it mentioned in trade papers like COMPUTERWORLD, and a recent posting in SOFT-ENG DIGEST says it is the basis for a software cost estimating product, I have never seen a paper in a journal that describes how the metric is calculated, and presents data showing how its predictions match with actual experience. COMPUTERWORLD says it was introduced in 1979 by A.J. Albrecht, and IBM researcher, but I find no articles by that author in THE IBM SYSTEMS JOURNAL since 1979 (actually there is one, but it has nothing to do with function points or any other metrics). Can anyone enlighten me? - Jonathan Jacky, University of Washington
uichanco@unc.cs.unc.edu (Edilberto Uichanco) (11/09/88)
The following articles describe and analyze Function Points: "Software Function, Source Lines of Code, and Development Effort Prediction: A Software Science Validation" by Allan J. Albrecht and John E. Gaffney, Jr. IEEE Transactions on Software Engineering, Vol. SE-9, NO. 6 November 1983 "Function Point Analysis: Difficulties and Improvements" by Charles R. Symons IEEE Transactions on Software Engineering, Vol. 14, NO. 1 January 1988 The first article is co-written by Albrecht, the developer of the Function Point estimation method. I hope these help. -Bong Uichanco (uichanco@cs.unc.edu)