Search RISKS using swish-e 

Forum on Risks to the Public in Computers and Related Systems

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

Volume 2: Issue 24

Saturday, 8 Mar 1986

Contents

Computerized ballot stuffing

Andy Kegel

Progress report on computerized voting

Kurt Hyde

Wild Modems

Bjorn Benson

Misdirected modems

Phil Ngai

Power outages

Phil Ngai

Earthquake problems with Nuclear Reactors

Lindsay F. Marshall

Info on RISKS (comp.risks)

Computerized ballot stuffing

In our area (extreme suburban Chicago, aka "the boonies"), we use a
computer-counted paper-ballot voting mechanism.  I am fairly sure I recall
serial numbers on the ballots.  However, I recognize that human memory is
weak and subject to interpretation and assumptions.  There is an election
coming up this month, and I will be particularly careful to observe and
understand the relevant facets of the process.

Remember, in Chicago, the rule is "Vote Early, Vote Often."

This message does not represent the position of my employer, or
any individuals or organizations other than myself.

    -andy kegel

Progress report on computerized voting

A sincere thank you to all the people who have responded to my request
for assistance in computerized voting standards.

I called New Hampshire's Secretary of State and he will be meeting
with me and some other people regarding security standards. I will
be proposing something like the following:


  Computerized voting booths should print a paper ballot for each voter
  to view and check for accuracy.  The hardcopy ballot must be visible
  to the voter by appearing under a covered (transparent) window.  The
  dimensions of the window must allow for at least 10 votes to be viewed
  at one time.  The printer must then feed each ballot into a ballot box
  which is guarded from access outside access while the voting machine
  is in use.  The audible signal which confirms that the voter is completed
  may occur after the hardcopy of the ballot is no longer in view.

  In order to protect the anonymity of the voter casting each ballot,
  each ballot must be on a separate piece of paper when deposited in
  the ballot box.  It may be be cut after printing or be sheet-fed into
  the printer.

  This additional functionality allows for a recount.  The current
  machines do not have the capability of recounting the ballots.  They
  only have the capability to recalculate from subtotals.

  Because of recount capability, it will be possible to resolve election
  disputes at the place of the voting.  This means it will not be
  necessary to contact the FEC and National Bureau of Standards in
  order to perform an audit on the machine's computer programs.
  The procedure for the FEC and NBS to audit the machine's computer
  programs has not been established and is likely to be extremely
  complex as certainly procedures must be established to be certain
  that the computer programs haven't been tampered with in order to
  return them back to their proper state.

My students at Rivier College will still be investigating further into
the proper security controls.  One of them is considering a way to let
the voter see his/her ballot and abort that ballot.  The printer would
then print an appropriate message such as "CANCELLED" on the bottom.

Once again, let me thank all those who are participating.  Your assistance
is very valuable and appreciated.  Let us not let the United States
suffer from a similar disaster as the Phillipines.

                                      Kurt

Wild Modems

All this talk in RISKS about modems calling humans rather than computers
reminded me of an article I read about telecomputing in Europe: it seems
that laws in Europe require modems to have equipment attached to explain
what is going on in four languages, should the computer happen to dial
a wrong number.

                        Bjorn N Benson

Re: Misdirected modems

This is an often repeated wives tale by people who ought to know better.
With ordinary dialup modems of the 103/212 class, it is the *answering*
modem which initiates a tone. The originating modem (the one that dialed)
remains silent until it hears the carrier of the answering modem.

Thus, if a computer dialed a wrong number, the person receiving
the call would hear nothing, not a "funny whistle".

power outages

I am familiar with AMD's data center. It is relatively small by comparison
to some sites, having only four IBM 3081s and one 3090, but it does have
battery backup and a huge dual turbo charged diesel generator. The diesel
has a thousand gallon fuel tank, which will last it 24 hours. We have
arrangements to get refills within that 24 hour period, so our data center
could presumably survive an indefinite outage and you could continue to
order chips from us even during a blackout!

Earthquake problems with Nuclear Reactors.

This is not really computer related, but seems interesting all the same....

A recent article in The Guardian highlighted some investigations into the
safety of British nuclear reactors in the face of the kind of mild earthquakes
that we have here. In particular it mentioned the Calder Hall reactor which
is nearly 25 years old and is built quite near to the area of Britain that
has the most earth tremors. This installation has a reactor vessel weighing
2000 tons suspended 18ft above the ground which is now so radioactive that
it would be impossible to examine or modify. The investigation showed that
the original safety calculations "had been done on the back of an envelope"
and that the reactor bolts might shear with an earthquake of 0.5 (units?).
There was an earthquake of that intensity last year, but it is impossible to
find out if anything was damaged because of the intensity of the radiation
not forgetting the 5ft of concrete and steel surrounding the chamber.......

So if you hear that Newcastle vanished, you'll know why!

           [and we'll be back to carrying coals ...  PGN]

Search RISKS using swish-e 

Report problems with the web pages to the maintainer