The Risks Digest

The RISKS Digest

Forum on Risks to the Public in Computers and Related Systems

ACM Committee on Computers and Public Policy, Peter G. Neumann, moderator

Volume 3 Issue 23

Tuesday, 22 July 1986


o Re: Comet and Electra
Jim Horning
o 100,000 Late Phone Bills
Mike McLaughlin
o Types of "Programming"
Henry Schaffer
o Info on RISKS (comp.risks)

Re: Comet and Electra

Jim Horning <horning@src.DEC.COM>
Mon, 21 Jul 86 14:19:04 pdt
The message from Mike Williams (RISKS-3.22) reminded me of two stories that
have been passed down through the oral tradition.  I have no direct evidence
concerning either.  Perhaps some readers of Risks have evidence that could
help set the record straight?

- A numerical analyst once explained to me why all modern airliner windows
have rounded corners: Anyone concerned with solving partial differential
equations knows that square corners lead to singularities.  He said that the
Comet crashes were traced to metal fatigue at the (square) corners of its
windows.  (He concluded that airplane designers should study Numerical
Analysis.)  Does anyone know whether computers were used in the design of
the Comet?

- I also heard that the structural defect in the Electra I wing design had
not been caught by the stress analysis program because of an undetected
overflow on a critical run.  Can anyone provide documentation for this?  (I
think this story was on the grapevine at the NATO Software Engineering
Conferences in 68-69.)

These pieces of our mythology are worth documenting or debunking.  There may
be valuable lessons to be learned from them, and we ought not to insist on
learning them the hard way.

Jim H.

100,000 Late Phone Bills

Mike McLaughlin <mikemcl@nrl-csr>
Mon, 21 Jul 86 16:03:50 edt
Excerpted from the Washington Post, Saturday, 9 July 1986, page D1.
[Omissions... (bridges) and [comments] as shown.]

    More Than 100,000 Getting Months-Late (telephone) Bills
        By Nell Henderson, Washington Post Staff Writer

More than 100,000 Chesapeake & Potomac Telephone Co. customers might think
they've had a summer vacation from telephone bills.
But yesterday the company said the vacation is over: The bills are on the
way after a two-month delay.
The customers... have not received bills for local or long-distance service 
or both - since a computer tape failure in mid-May.

The high-tech roots of the problem were "flaws" in computer tapes that were
programmed for preparing the bills, (a spokesman said).  "The problem erases
itself," he added.
The low-tech solution was to use people to put the billing information into the
system, using separate records of the calls, he said.
The result was that many of the customers did not receive phone bills for
several months....

(A) customer... was told to call... if he has any trouble paying the entire
bill at once. 
"We would be lenient on payment, and would be glad to speak to customers on 
an individual basis . . . We're sorry for any inconvenience,"... 
The problem also affected an unknown number of bills for long-distance service
provided by MCI Communications Service...

Types of "Programming"

Henry Schaffer <ecsvax!hes%mcnc.csnet@CSNET-RELAY.ARPA>
Fri, 18 Jul 86 23:37:38 edt
  "Programming" encompasses much more than the use of the traditional
languages (Basic, Ada, or whatever.)  Entering formulas in a spreadsheet
or specifying record and report structures in a database are also
programming - in higher-level, albeit specialized, languages.  Thus
JAN Lee *is* teaching his students to program, and in the most
appropriate and productive manner.  They can learn something quite
important and useful in this part of the class.  It is the other
faculty/administrative objectors (the ones who want to have 4 weeks
of traditional language put in) who are asking for something both
unproductive (most of the students will neither learn new concepts
nor something useful) and risky.

  There is an implicit understanding about a terminal course - that
you've been carried along far enough so that you can use what you've
been taught.  A student who finishes one semester of a CSC sequence
knows that he/she is not through learning,
and should not presume (one hopes) to take on responsibilty for a
critical application program.  However, a student who is taught
that programming is 4 weeks of a survey course in computing might
not be so timid!  (I assume that these students will not usually
take any more programming - if they generally did then there wouldn't
have been the pressure to push programming into JAN Lee's course.)

  Our university (NCSU) has recognized that the details of the
type of "programming" needed are dependent on the discipline, and
can variously include spreadsheets, statistical packages (I can
argue that one can "program" in SAS), etc., and also the more
traditional languages.

--henry schaffer  n c state univ   ...mcnc!ecsvax!hes  (uucp)

Please report problems with the web pages to the maintainer