Re: Is hand count transparent?

From: Ronald Crane <voting_at_lastland_dot_net>
Date: Wed Nov 14 2007 - 21:48:42 CST

Danny Swarzman wrote:
> There have been postings on this list suggesting that hand counting
> is intrinsically more transparent than machine count.
> I think this is completely false on both sides of the inequation.
> I have watched hand counting. I went with an expert election
> observer. He saw flaws in the process that I didn't catch. But it was
> really clear that neither of us could actually know if it was
> conducted fairly. Your team would need complete access to all the
> places were ballots are handled and counted. You would need to follow
> the flow of all the data. They would be there for weeks.
> To make such a process publicly viewable and secure is a real
> logistical challenge for the elections board. Not all of them are up
> to it.
> Where scanning and counting is done by computer, there are fewer
> points at which people intercede. It is easier to make those critical
> points publicly viewable.
> But, software isn't witchcraft. An ordinary person can figure it out.
> There are a lot of people who use computer languages such as Java and
> xml. There' a huge number of books on these subject.
> If you don't want to read source code, you can get the source code
> from the internet and follow directions and set up your own elections
> system. Print some ballots. scan some votes see who wins in your world....
There are many software engineers, but few who understand much about
security. How many software engineers know that the test: "Print some
ballots. scan some votes see who wins in your world" tells them almost
nothing about how the systems purporting to use the downloaded software
actually performed in a particular election? And how many of them know
some of the reasons why? And how many of them know that even reviewing
the purported source code still leaves open several attack vectors? And
how many of them know what some of those vectors are?

> Do you really think there is a larger class of people who can
> observer hand counting than there is who can verify open source
> software? They weren't in City Hall last Tuesday round midnight.
Yes, I do think that there are many more people who -- with only modest
training -- can effectively supervise hand counts, than there are people
who effectively can analyze a computational system's security profile.

Remember also that an observational failure during a precinct's hand
count might permit a limited amount of fraud in that precinct. A failure
during a computational system's security analysis, on the other hand,
might permit fraud in every precinct in which that system is used.

OVC-discuss mailing list
By sending email to the OVC-discuss list, you thereby agree to release the content of your posts to the Public Domain--with the exception of copyrighted material quoted according to fair use, including publicly archiving at
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
Received on Fri Nov 30 23:17:24 2007

This archive was generated by hypermail 2.1.8 : Fri Nov 30 2007 - 23:17:31 CST