[LINK] [UK] Call for e-voting to be scrapped amid security fears

Rick Welykochy rick at praxis.com.au
Wed Jun 27 09:18:52 AEST 2007


Janet Hawtin wrote:

> Can you test it on voting re surveys in shopping centres or for big
> brother or jjj
> at football events? surely there are some friday night shopping places
> which generate a flow of people which would help in testing. what
> about using it as an ancillary method for 10 election cycles until
> youre sure its an honest reporter of people's choices.

I see two main aspects to testing this sort of thing:

(a) The workstation itself, for recording votes. This station
     only needs to operate at the speed of a human, i.e. display
     candidates, securely record the vote, issue audit trail.
     Stress testing is simple: just hire some monkeys to pound away
     at human speed. It will work or fail. The stn does not need
     to run under any particularly heavy load, just a normal
     work stn load.

(b) The vote aggregation and tally server. This requires severe
     stress testing. One way to accomplish this is to write some software
     that simulates any number of concurrent e-vote work stns (see
     (a) above) and point them at the tally server. This is NOT
     rocket science. If you are interested in a simple example of
     this kind of stress, see the apache bench program
     <http://httpd.apache.org/docs/2.0/programs/ab.html>

The original posting indicated that the WORK STATIONS themselves
were crashing. This is unforgiveable and should result in contract
cancellation and damages.

cheers
rickw


-- 
_________________________________
Rick Welykochy || Praxis Services

Any belief that can't stand up to objective scrutiny is hardly worth having.
      -- LJ McIntyre



More information about the Link mailing list