Path: utzoo!yunexus!geac!daveb
From: daveb@geac.UUCP (David Collier-Brown)
Newsgroups: comp.software-eng
Subject: Re: Formal methods - does anyone use them?
Message-ID: <3091@geac.UUCP>
Date: 28 Jul 88 12:46:16 GMT
Article-I.D.: geac.3091
References: <781@cernvax.UUCP>
Distribution: comp.software-eng
Organization: GEAC Computers, Toronto, CANADA
Lines: 26

From article <781@cernvax.UUCP>, by hjm@cernvax.UUCP (hjm):
> The subject of formal methods was brought up recently (VDM, Z et al.), and I
> was wondering to what extent people use these methods on 'real' projects.  Do
> people just use them for the design phase and then just code it normally, or do
> they refine the mathematical description until it turns into executable code?

  Well, for certain well-specified things, my implementations are
taken directly from formal specifications, or from composing the
specifications of several parts to make a new whole.  This can be a
real time-saver, especially If you collect formal (algebraic,
mostly) specs.


  I only do about 10% this way, mostly because I'm better at using
specs than writing them...
  Currently I'm reading Wladislaw M Turski & T.S.E. Maibaum's "The
Specification of Computer Programs" (Addison-Wesley, 1987), in hopes
of improving the odds.

--dave (I also use WEB) c-b
  
-- 
 David Collier-Brown.  {mnetor yunexus utgpu}!geac!daveb
 Geac Computers Ltd.,  |  Computer science loses its
 350 Steelcase Road,   |  memory, if not its mind,
 Markham, Ontario.     |  every six months.