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 4 Issue 70

Wednesday, 1 April 1987


o Rocket Shot Down By Faulty ``Star Wars'' Weapon
Phil R. Karn
o ATMs, phones, health hazards, and other sundry subjects
o Computer Risks in Theatre
Warwick Bolam
o PC fumes
Dick King
o A real eye-catching headline
David Chase
o Risks of being fuzzy-minded
Ted Lee
o ATM discussions
o Re: ATM experience ... it actually gets worse
Allen Brown
o Info on RISKS (comp.risks)

Rocket Shot Down By Faulty ``Star Wars'' Weapon (From the AP wire)

Phil R. Karn <>
Wed, 1 Apr 87 19:34:50 est
AM-RocketFailure-StarWars     04-01  0400
AM-Star Wars,400
Rocket Shot Down By Faulty ``Star Wars'' Weapon
By Lou Flirpa
Associated Press Writer

    WASHINGTON (AP) _ Reliable Pentagon sources have reported that last
Thursday's explosion of a $78 million Atlas-Centaur rocket carrying the
$83 million military ``FltSatCOM'' communications satellite was in fact caused
by a ``minor malfunction'' in a highly secret experimental Strategic Defense
Initiative beam weapon, commonly known as ``Star Wars''.
    ``We're not sure yet what happened'' said one highly placed source,
who spoke on condition that he not be identified. ``But we think the
autonomous boost-phase battle station we launched on Delta last year mistook
the Atlas for a Soviet ICBM and shot it down. Naturally we all feel pretty bad
about this. Gosh, we're real sorry. Really.''
    Speculation had been mounting after the launch failure that the Atlas
had been hit by lightning.  According to sources, however, ``a charged
particle beam weapon is essentially an artificial lightning machine.''
    Since the launch took place in a rainstorm, it was easy to jump to
the conclusion that lightning struck the vehicle, the sources said,
especially since no one actually saw the explosion because of the cloud cover.
    While the exact cause of the ``malfunction'' has not yet been
determined, there is early speculation that the on-board ``clock'' of the battle
station was incorrectly set five hours ahead of ``universal'' time instead
of five hours behind, leading it to ``believe'' it was over the Soviet Union
when it was really over Florida.
    ``It looks like some of our scientists got confused over which way the
earth turns. I guess they found out the hard way,'' said another source.
    SDI director Lt. General James A. Abrahamson was reported to have
``mixed feelings'' when told of the accident.
    AP-NR-04-01-87 1313EST

ATMs, phones, health hazards, and other sundry subjects

Peter G. Neumann <Neumann@CSL.SRI.COM>
Wed 1 Apr 87 22:29:27-PST
In the epicycles of RISKS, I think we are ebbing.  12 recent messages to
RISKS were slight variants on earlier ones, and I have decided (of course,
very arbitarily) to blow the whistle.  Sorry to those of you who composed
careful messages that are not included in this issue.

I conducted a few informal polls, and feel (at this point in RISKS) that I
have been too permissive lately, and have even lost a few readers who cannot
devote the time to screening (literally).  Thus (for a while, at least), I
will try to include only the more incisive contributions.  (You may notice
that I try to put the more exciting things FIRST -- unless they are very
long, in which case I tend to put them LAST.)  On the other hand, fear not
for withdrawal symptoms -- some new disaster always tends to happen, and we
are off again in another direction...

By the way, there was this response to my earlier note on this metasubject: 


  I'd like it to enter the culture that whenever someone runs into an
  incredibly obscure bug, they feel a sense of responsibility to share it with
  the community, to save others the same hassle and danger.  RISKS could
  become the customary channel for this.

Following are a few messages that I let slip by.  

Computer Risks in Theatre (Re: RISKS-4.68)

Warwick Bolam <munnari!goanna.oz!wjb@seismo.CSS.GOV>
Mon, 30 Mar 87 10:45:03 EST
Recently, a stagehand was severely injured in a Melbourne theatre.  He was
on a stage-ladder.  These are large, free-standing ladders that are wheeled
from place to place on the stage to facilitate access to the grid area above
the stage.  The ladders are massive, very stable and hydraulically operated.
The accident occurred when someone activated the computerised stage moving
system.  This system allows sections of the stage to be raised, lowered and
moved about.  The ladder was at the front of the stage, the parts of the
stage that were intended to be moved were at the rear.  A mistake was made and
one of the sections that the ladder was standing on was moved.  The ladder
toppled and the stagehand suffered a fractured skull and a broken pelvis.
It was fortunate that no one else was hurt.  Standing orders are not to move
the stage when there are people on it, but this is commonly ignored.

        Warwick Bolam wjb@goanna.oz

PC fumes

Dick King <king@kestrel.ARPA>
Mon, 30 Mar 87 13:48:41 pst
    From: vortex!lauren@rand-unix.ARPA (Lauren Weinstein)
    Subject: Fumes from PC's
    The most likely cause of a problem is OZONE..

Induction motors don't generate ozone, and those are the type used in
computer fans and [probably] disks.  A more likely source of ozone is the
CRT high voltage.

There may be other sources of fumes in a PC, such as undried solvent
-- does anyone know anything about this?

A real eye-catching headline

David Chase <>
Sat, 28 Mar 87 02:25:20 CST
IEEE Spectrum, April 1987:

      "Inherently safe nuclear reactors"
                                            [Add to the oxymoron list.  PGN]

Risks of being fuzzy-minded

Mon, 30 Mar 87 17:43 EST
All right, already.  My pilot ("Overconfidence in Airplane Computers")
was more right than I:  the thrust of the plane IS measured in the same
kind of units as its weight, and to say that one is half of the other is
a meaningful statement (the plane takes off with half the acceleration
it would have if it were dropped off a cliff).  My only defense is that
as a defrocked physicist I'm so used to people getting mass and weight
confused that I automatically assumed it had happened one more time.
The letters can stop.

Sat, 28 Mar 87 08:40:00 PST
Deposits on ATM:

Various banks have various systems.  As an example, at CITIbank
a deposit was made to a specific account.  Your account was updated
with a MEMO update, i.e. it would show up on your balance.  However
it did not become AVAILABLE funds until it was verified by a teller.
On the envelope was Customer ID number, the envelope number and
the Entered dollar amount, the branch # and the Machine #.

There was also a selection for OTHER PAYMENTS.  This allowed you to
dump any deposit into the ATM.

What are you assured then when you deposit to an ATM ?

1) You have a banking RECORD (not a reciept at Citibank).  If you
   have this record, there is a VERY high percentage that you
   deposited something at that ATM.

2) Some banks have ways of crediting your deposit RIGHT NOW.
   This could be done by a balance in another account (i.e. a long
   term C.D. or a line of credit.)  That way they can get you if
   you lied.

ATM Splitting a Card in half

   I've worked with about 75% of the types of machines on the market
and NONE of them split a card in half upon swallow.  However, some
NETWORKS have a policy of  slicing a card to avoid security

Trusting an ATM.

Interesting you should bring this up, I'm just bruising up a paper
describing a REAL situation where your card and PIN are in the clear.
This involves a customer using a bank that is part of a network.
All the information was available to folks in DP, if they put in some
efforts to get it.

Re: ATM experience ... it actually gets worse [Chapman 1987 03 26]

Allen Brown <>
Tue, 31 Mar 87 15:21:54 est
   [Included for the reference.  Perhaps it will stave off further repetition.]

Brent Chapman makes reference to magnetically encoded deposit slips, and
the interesting differences between human and machine interpretation
of the same piece of paper.

In one story, a customer surreptitiously laid out courtesy slips
on the bank counters which had been magnetically encoded with his account
number. It ended in the customer's withdrawal of $100K of others' money and
his subsequent disappearance. Such actions have, apparently, taken place
in several banks.

In another case, a cheque had been magnetically encoded with a valid bank
branch code (and a bogus account number) that was different from the name of
the bank on the cheque paper. The perpetrator had originally deposited a
large sum of money in the bank indicated on the cheque paper.  Then he
opened bank accounts in a number of other banks using these cheques. Owing
to machine-sorting each cheque bounced back and forth between two banks,
with an associated transit time of two days per rebound. The machine at one
end could not validate the account and hence dumped it into a pool for
manual sorting, where the human response was to assume a simple routing
error (because the bank name on the cheque was certainly not theirs), at
which point it was sent to the named bank. At the named bank the cheque was
machine-sorted for final clearance, and since it was coded for another bank
(the first one), it was automatically directed (back) there. The hoax was
only discovered because the well-travelled cheque became too frayed by
machine handling to be further automatically processed. Having had a number
of such cheques accepted for deposit, the depositor had made withdrawals and
had disappeared with $1M by the time of discovery.

These stories, and a number of others are recounted in a ``delightful'' little
book called Computer Capers (Mentor, 1978 - no ISBN) by Thomas Whiteside.
Most of the material appeared originally in The New Yorker. Whiteside has
a good bibliography for titles published between 1966 - 1977, but
the book is clearly now a bit dated. White-collar crimes have undoubtedly
advanced beyond the ``stone tools and knives'' stage of ten years ago, but
you can be sure that we won't hear about them from the banks, etc.

Allen Brown

Please report problems with the web pages to the maintainer