Re: N-version software compared to others

From:         dick@silicon.csci.csusb.edu (Dr. Richard Botting)
Organization: CS Dept., Calif. State Univ., San Bernardino
Date:         14 Jul 96 22:43:28 
References:   1 2 3 4 5
Next article
View raw article
  or MIME structure

Perhaps a good software development method for non-mass marketed
software is one that forces the developer to ask nasty questions about the
requirements, dig out the inconsistencies, nail the ambiguities,
plug the holes with no known behavor.  Also spot unrealistic
assumptions and infeasible wishes... before spending a $1e6?

N-version ( and 1-version) programming do not help with this.
Upfront methods with a mathematical basis (JSP/JSD/VDM/SCR/RSML/...)
do seem to do this however.

Now, in the more rigorous methods (JSP/...) I've often seen
almost identical solutions by parallel development teams.  So N-version
does not win you very much.

--
dick botting     http://www.csci.csusb.edu/dick/signature.html
Disclaimer:      CSUSB may or may not agree with this message.
Copyright(1996): Copy freely but say where it came from.
	I have nothing to sell, and I'm giving it away.