Date: Thu, 31 Mar 94 15:55:57 PST
From: RISKS Forum <risks@csl.sri.com>
Subject: RISKS DIGEST 15.72

RISKS-LIST: RISKS-FORUM Digest  Thursday 31 March 1994  Volume 15 : Issue 72

----------------------------------------------------------------------

Date: Thu, 31 Mar 1994 15:34:26 -0800
From: Phil Agre <pagre@ucsd.edu>
Subject: Rental cars and financial derivatives

The 3/30/94 New York Times includes two articles that illustrate the
vexatious trade-offs inherent in emerging computer-based systems.

  Matthew L. Wald, Two technologies join to assist lost drivers, New
  York Times, 30 March 1994, page A13.

This article is about a computer system that Nynex is developing, and that
Avis will be testing, in which rental cars are kept in close contact with the
rental agency through wireless communication.  The technology is sold as a way
of protecting drivers such as the tourists who were attacked in Florida; the
cars will be equipped with "panic buttons" and the like.  The article also
says that drivers will be able to call in for directions on wireless phones,
with the phone operators having access to digitally encoded GPS information
plotted against detailed maps, enough to be able to say "take the next left"
remotely.

So far so good.  But, at least the way the article describes it, the system
will also allow the company to track all drivers for all purposes, regardless
of whether they are in danger or need directions.  A natural suspicion is that
this generalized tracking capability is a major part of Avis's actual motives
for promoting the systems.  Motives aside, the privacy concerns may be serious
in any case.  How might these concerns be weighed against the advantages?  How
might the system be designed to obtain the advantages without the
disadvantages?  The article contains no hint that such questions are being
asked, and this is unfortunate.

  Barnaby J. Feder, Sophisticated software set for exotic financial
  trades, New York Times, 30 March 1993, pages C1, C5.

This article concerns "a marriage made in techno-geek heaven" between computer
people and high finance, specifically software for analyzing and administering
complex financial transactions based on so-called "derivatives" (see Risks
15.66).  One virtue of these systems is that they reduce the possibilities for
error, which are pretty serious when these kinds of transactions are done by
hand.  At the same time, such systems allow derivatives to be traded in much
larger volumes, and in much more complex ways.

Much popular imagery associates derivatives with speculation, for example
high-stakes gambling in commodity futures, but the real issue is almost the
opposite.  The usual purpose of these transactions is to engineer little
islands of stability and predictability within the swirling chaos of global
financial markets.  Indeed the metaphor of "engineering" is frequently used --
the software discussed in the article is referred to as "financial CAD
(computer-aided design)".  The potential trouble comes when massive financial
edifices are engineered badly.  When a steel-and-concrete building falls down,
the earth is there to catch it and a limited number of people get killed.  But
that's not how financial engineering works -- one collapsing structure has the
capacity to take others down with it (again, see Risks 15.66).  Obviously it's
in their interest to be careful, but let's hope they know what they're doing.

Phil Agre, UCSD

------------------------------

End of RISKS-FORUM Digest 15.72
************************