Voting Project Needs Python People

Alan Dechert adechert at
Mon Jul 21 23:27:15 CEST 2003

"Harry George" < at> wrote in message
news:xqxwuebfyym.fsf at

> I was curious why wasn't
> mentioned.  I had the impression that was an early (and continuing)
> portal for these issues.
Rebecca Mercuri and Peter Neumann are not part of this project.  I spent a
lot of time talking with Neumann 12/00 - 1/01 (met him in person 12/17/01 at
state Assembly Elections Committee hearings in Sacramento).   They have a
competing idea that would be much more expensive (dedicated DRE voting
machine with printer under glass so the voter can see it but can't touch
it).  Frankly, I don't see their idea going anywhere.

The serious competition we've had has been the Caltech-MIT project.  They
started with a grand idea but fizzled. At the top of the 12/14/00
Caltech/MIT press release about a "Uniform Voting System," the presidents of
MIT and Caltech were announcing "a collaborative project to develop an
easy-to-use, reliable, affordable and secure United States voting machine."
And they said, "America needs a uniform balloting procedure."

They certainly had the right idea, but I had already said the same thing and
had gone a step further to say that the machine should be PC based and open
source.  Part of the reason we had difficulty getting funding in 2001 was
because people told us (some important people, that is, holding some big
purse strings)  that they thought Caltech-MIT were already set to do what we

However, it turns out that while they started with a great idea, the team
assigned to carry on the project never embraced the idea of a uniform
system!  They set about to work with vendors.  I don't know about you, but
it seems obvious to me that vendors would never really get behind a uniform
system.  Afterall, the way the make money is with proprietary systems that
compete with a bunch of other proprietary systems.  They each have their

Caltech-MIT underestimated the problem -- especially the political aspect.
I don't see their project going anywhere.  I mean, people can say I've
gotten nowhere in 2.5 years but I never had any money.  They had lots of
money but bad clues.

We don't plan to work with existing election vendors.  I see a new crop of
election vendors we can cultivate from PC remarketers.  25 million PCs are
retired every year in the U.S. and remarketers are always looking for ways
to get a few extra dollars for what they get for practically nothing.  I
have already contacted some remarketers and they like our idea.

For example, Jay at Isis Technology would love to become an election vendor.

> Is the intent to do opensource all the way to commodity chips?  Else a
> proprietary BIOS could be the weak link.
No, we won't go that far.  But we will have software to validate the system
on bootup.  Bennet Yee of UCSD is one of the leading experts on this

The Trusted Computing Group is also working on technology that may be

Alan Dechert

More information about the Python-list mailing list