Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site ames.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!think!harvard!seismo!lll-crg!dual!ames!eugene
From: eugene@ames.UUCP (Eugene Miya)
Newsgroups: net.cse
Subject: Re: Criteria
Message-ID: <1164@ames.UUCP>
Date: Thu, 26-Sep-85 02:05:04 EDT
Article-I.D.: ames.1164
Posted: Thu Sep 26 02:05:04 1985
Date-Received: Sun, 29-Sep-85 08:00:50 EDT
References: <716@wdl1.UUCP>
Organization: NASA-Ames Research Center, Mtn. View, CA
Lines: 31

> 
>       When you design large structures, such as bridges and buildings, the
> design is normally validated by structural analysis, with generous safety
> margins to allow for flaws in the materials and other problems.  But you
> don't normally build a scale model of the entire building and subject it
> to static testing; the theory is well-enough developed that static behavior
> is predictable.  So structural engineers must learn how to design, on paper, 
> buildings that will definitely stand up.
> 
> 					John Nagle

Actually, I worry alot about this [note this is not directly CS].
Two DOD cases in point: the Sargent York and M2 Bradley fighting vehicle.
Neither of which has undergone "full-testing."  York is well documented.
Bradley -- FMC refuses to hit one with a shell fully loaded.  Have we lost
confidence in our design and engineering methods?

My subtle worry is about using computers to economize (CAD) on a product
which otherwise might be over-engineered into "safe" design.  Such designs
might be alright under normal conditions, but in the twilight zone of
the boundary condition: perhaps, the building falls, the plane drops out of
the sky, the bridge .....  This has happened in many cases, and I am glad
the fa.risks board has started.  I don't think software is any different
except perhaps we might be a bit more removed from the consequences
[See discussion on the ARPA soft-eng board for more].

From the Rock of Ages Home for Retired Hackers:
--eugene miya
  NASA Ames Research Center
  {hplabs,ihnp4,dual,hao,decwrl,allegra}!ames!aurora!eugene
  emiya@ames-vmsb