MathematicalTables

DavidParnas? believes that software systems should be documented, and should be documented using well-defined notations. Not boxes and lines that could mean anything, nor neccesarily upside-down A's and back-to-front E's that mean who knows what (or at least if quantifiers and such are used it, certianly shouldn't the way they are typically used).

Parnas wants us developers to use mathematics to capture requirements and design systems. And he wants us to do it the way engineers in other technical communities do it, using "closed form" maths, essentially equations that let you calculate values you need. And emphatically not the theorem systems leading to proofs that most formal methods fans promote.

Parnas presents his mathematical tables as a way to do this. These tables can indeed use predicate calculus and similar hairy stuff, or they can use something much closer to the natural language of the system's users. Papers collected in Software Fundamentals: Collected papers by David L Parnas give examples of both kinds.

The code is really interesting, it doesn't look like anyone else's Java --SteveF ³host³³date³October 2, 2002³agent³Mozilla/4.0 (compatible; MSIE 5.5; Windows NT 5.0; .NET CLR 1.0.3705)³MathematicalTables