Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!zaphod.mps.ohio-state.edu!sdd.hp.com!spool.mu.edu!agate!lima.berkeley.edu!bks
From: bks@lima.berkeley.edu (Bradley K. Sherman)
Newsgroups: comp.software-eng
Subject: Specification Languages
Keywords: Missouri
Message-ID: <1991Jun14.171653.26325@agate.berkeley.edu>
Date: 14 Jun 91 17:16:53 GMT
Sender: usenet@agate.berkeley.edu (USENET Administrator)
Organization: University of California at Berkeley
Lines: 21
Originator: bks@lima.berkeley.edu


   I've always found the concept of a "specification language" to be
a bit dubious in the context of the programming projects that I have
been involved in.   I can remember being told by a superior that I
should not have used the word "algorithm" during a meeting that
we had with upper-management types from one of our clients.  What they
would have made of a document written in a synthetic language is
amusing to contemplate.

   Could proponents of the use of specification languages (other
than natural languages) please give doubters like me a little taste
of the art?   I won't ask for a specification for, say, producing and
tracking purchase orders, or for a complicated but well defined function like
qsort().  How about specifying the C function abs() rendered in English
like this:
	abs() returns the absolute value of its integer operand.

-------------------------------------
	Brad Sherman (bks@alfa.berkeley.edu)

History does not repeat but it does rhyme --Mark Twain