Re: [tlug] Somewhat OT- open source software for US voting machines

[Edward Middleton <emiddleton@example.com>]

SL Baur wrote:
> On 9/30/08, Pietro Zuco <zmlist@example.com> wrote:
> 
>> Yes it is a great idea. I just wonder how much time it's going to take
>>  for M$ to pay money under the table to change that ideas...
> 
> No, it's a terrible idea.  The issue of Open -vs- Closed source on
> E-voting is a red herring.

Well it isn't,  It needs to be open source because you can't have a
transparent process without it being open source.  But while it is
necessary, it is far from sufficient.  What is the red herring of such
discussions is that speed is the central reason for considering an
e-voting system.

> Note that the message I was responding to referred to a
> "Nixon number", defining it as the number of people needed
> to change an election's results.
> 
> I wrote in Slashdot on the issue of electronic voting:
>> We are treating E voting like a minor IT procurement
>> project, when we should be treating it as Democracy's
>> Manhattan Project.
> 
> I presume the "Nixon" number refers to the 1960 election, stolen in
> Chicago by a handful of votes?
> 
> Right idea, wrong project. The Manhattan Project was a massively
> funded, mad dash for survival and let's face it, E-voting just is not
> that important.

It is not that important because if a vote is being decided by a
fraction of a percent both possible candidates are significantly
representative of the wishes of the voting public.

 Ideally it would be more like the mission to the moon,
> which was also massively funded, but each step of the way was
> carefully and meticulously planned and tested before being deployed.

This seens to be the big difference between what is happening in US
e-voting effort and the Australian one I linked to.

> As a matter of fact, it's really not a problem worth spending money on
> solving. There are some things that are done better by hand and
> counting election ballots is one of them.

In the US it is not worth solving because the number of people who don't
 vote is more significant then the number who's vote could be
miscounted with a manual vote.  The number of people not voting raises
more issues about whether the election is representing the wishes of the
public then the several percent error in a count.

> Thomas Edison's first invention was an automatic vote recorder for
> legislatures. It failed to generate any interest.
> http://www.conservapedia.com/Thomas_Edison [conservapedia.com]
>  
>> He obtained his first patent on his first "real" invention,
>> an automatic vote-recording machine. However, as with
>> many inventors first attempts, it was not well received
>> and turned out to be unmarketable. This was not because
>> it did not work; it worked well, it was because the market
>> was not receptive to the invention.
>
> The way I first read about this was more instructive, but I cannot
> find where the more detailed reference is. Edison was taken aside by
> one lawmaker in Washington who explained to him that if counting votes
> in Congress was too fast, they could well wind up voting for
> legislation that should not pass.

So Edison solved a problem that no-one else thought was a problem.
After the last US election there is no question that there is a problem.
 Whether e-voting is the solution is debatable.

> There is no need to rush the process. There is no need to declare
> elections over a month before votes are cast. There is no need to
> declare a winner before all voters have voted when votes are being
> cast. There is no NEED for E-voting. 12-24 hours to handcount paper
> ballots is sufficient and also enough to have the counting
> audited/supervised by independent parties.

There is a need for people to be certain that the declared election
result was "the will of the people".  Not being able to resolve the
election after a recount and having to resort to the courts is a failure
of the election process.

Edward