Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uwm.edu!linac!pacific.mps.ohio-state.edu!zaphod.mps.ohio-state.edu!wuarchive!uunet!visix!news
From: amanda@visix.com (Amanda Walker)
Newsgroups: comp.object
Subject: Re: A Hard Problem for Static Type Systems
Message-ID: <1991May3.212005.29453@visix.com>
Date: 3 May 91 21:20:05 GMT
References: <554@eiffel.UUCP> <1991Apr26.203642.17387@leland.Stanford.EDU>
	<556@eiffel.UUCP> <52166@nigel.ee.udel.edu>
	<1991May1.143831.2065@maths.nott.ac.uk>
	<1991May3.184332.28319@visix.com>
Sender: news@visix.com
Organization: Visix Software Inc., Reston, VA
Lines: 25

In article <1991May3.184332.28319@visix.com> adam@visix.com writes:

   However, I think you overemphasize the importance of completely
   specified problems.  In my view, underspecified problems are not
   just quick prototypes.  System browsers are underspecified.
   AI programs are underspecified.

Indeed.  In fact, one of the advantages I have found in OO design and
implementation is that it makes it possible to attack problems over
incompletely specified domains.  In particular, dynamic type systems
allow existing code to work over domains which were not originally
specified, as long as those domains have the necessary
characteristics.  A trivial example of this idea is that a heapsort
should operate on any set of objects for which I can provide
comparison and exchange operations.

One all too common example of this situation is using a precompiled
module with a new data type (i.e., code reuse at the binary level),
which we can expect to happen more and more often as OO concepts
become more "mainstream."
--
Amanda Walker						      amanda@visix.com
Visix Software Inc.					...!uunet!visix!amanda
-- 
"It's not the tragedies that kill us, it's the messes."	   --Dorothy Parker