[E-voting] Submission

John Bernard Lambe icte-jlambe at johnlambe.com
Fri Mar 19 18:31:30 GMT 2004


We have to remember that whoever is writing the Department's submission
will presumably read ours first (and are probably doing so now), and we
presumably won't have the opportunity to see theirs before submitting
ours.

This gives them the opportunity to try to rebut our arguments in their
submission and their 'rebuttals' might need only to be convincing
enough to persuade non-technical people.

We could make the arguments more difficult to rebut (or give the
appearance of rebutting) by providing more detail (in effect, giving
the case against possible rebuttals).
For example, specifying why random sampling (for transfer of surplus)
is not a problem and how you could compare VVAT ballots to the
electronic record very accurately.

By the way, does anyone know anything about who is writing a submission
for the Department?



More suggestions:

We could refer to existing DRE machines which support a VVAT:
There is one manufactured by ES&S (http://www.essvote.com/) (I don't
know whether it's on their web site).

Refer to the growth in voting systems which support a paper audit trail
in the US:
In her paper, "A Better Ballot Box"
(http://www.notablesoftware.com/1002evot.pdf), Dr. Mercuri cites a
survey of the use of voting machines by type in the US.
The total proportion of US counties using exclusively either paper
ballot, optical scan or punched card systems rose from 60.3% in 1980 to

71.9% in 2000 (this figure does not count counties which use multiple
systems).

Refer to US states which are planning to switch to optical scan
systems.


Section 3.6:
"Given enough time, a seal may be broken and repaired. If the attacker
has suitable resources, a counterfeit seal may even be manufactured."

That might sound like we are making claims that we have specific
information about this seal. I think that the wording should make it
clear that we are talking about seals in general, but that this still
applies to the system in question.
Maybe "Given enough time, any seal may be broken and repaired...".

We could remind the reader at this point, that the count centre PC does
not (as far as we can tell from the information released) have any
similar protection from tampering the hardware or software and that
tampering with the count centre PC is logistically easier since only
one machine has to be tampered with to throw an election in one
constituency.


P.10:
"there will be no evidence which might distinguish a vexatious action
from a valid one."
I suggest something like:
"there will be no evidence which might distinguish a vexatious action -
whether in good faith or malicous - from a valid one."



=====
John Lambe
---------------------------------------------------------------------
Phone (mobile): +353 86 2895286
Phone:          +353 1 4905842  
Address:        64 Brighton Road, Rathgar, Dublin 6, Ireland    
Email:          jlambe at johnlambe.com




More information about the E-voting mailing list