Xref: utzoo comp.object:1596 comp.software-eng:4049
Path: utzoo!censor!geac!nixtdc!doug
From: doug@nixtdc.uucp (Doug Moen)
Newsgroups: comp.object,comp.software-eng
Subject: Specification Tools and Code Testing
Message-ID: <1990Aug13.140347.9441@nixtdc.uucp>
Date: 13 Aug 90 14:03:47 GMT
References: <4800097@m.cs.uiuc.edu> <243@cadlab.sublink.ORG> <5456@stpstn.UUCP> <1990Aug12.134735.22528@cbnewsm.att.com>
Reply-To: doug@nixtdc.UUCP (Doug Moen)
Followup-To: comp.software-eng
Distribution: usa
Organization: Nixdorf Computer Engineering Company
Lines: 36

cox@stpstn.UUCP (Brad Cox) writes:
)My IEEE Software paper (Nov 1990) goes into the distinction between 
)specification and implementation in some detail...
)
)Where inheritance really belongs, and where it offers its greatest
)potential, is in the (as-yet-nonexistent) specification tools; i.e.
)specification language "compilers" that compile an input notation, the
)specification, into but *tests*; executable code, or "gauges", that determine
)(by testing) whether a given implementation is within tolerance to its 
)specification.
)
)In case this isn't obvious, the meaning I'm using for "specification"
)includes not only *static* specifications (i.e. what method names/types
)are listed in some interface file; i.e. class Stack has methods push and pop),
)but *dynamic* specifications (i.e.  if I push 1,2,3 on an instance of class
)Stack, pop should return 3,2,1).

gregk@cbnewsm.att.com (gregory.p.kochanski) writes:
>Won't the process of writing the specifications be as complex as writing
>the program, if you want to specify things in such detail that all
>operations of a class can be tested?

My group is currently in the test phase of developing a large document
image processing system.  This is a commercial project, and we are spending
a great deal of effort in testing to avoid releasing a buggy product.
What we are discovering is that testing a big system is *hard*, and
that having a sound testing methodology is crucial.

A specification language, together with a compiler that automatically
generates test code, as Brad describes, would be tremendously useful
for us.  I don't think it matters that writing the specifications is
as labour intensive as writing the original code; we are already spending
a lot more time in testing and documentation than in coding already.

Does anyone have references for testing methodologies
or automated testing tools for large systems?