[EM] Why We Shouldn't Count Votes with Machines

[Kathy Dopp]

 Date: Sat, 23 Aug 2008 01:02:44 -0400
> From: Dave Ketchum <davek at clarityconnect.com>
> Subject: Re: [EM] Why We Shouldn't Count Votes with Machines

> Federal certification?  The many horror stories tell us either:
>     Equipment is failing that has never been "certified" or
>     The certifiers are signing off without bothering to look
> seriously for the many defects in the offered systems,

The second scenario is true, and there are loopholes in the standards
which allow systems to be certified despite not meeting the standards.

>
> Thus the certification process needs overhauling.

Yes, but certifying voting systems is a fundamentally flawed concept
anyway, because if the software is changed at all, then it is not
certified any longer and many states require that only certified
software is used. This makes it legally impossible to do security and
bug fixes because it can take a year (or perhaps more, but a really
long time) to get a new voting system software federally certified.
Smart State Election Officials are beginning to see that federal
certification is not a good idea, but many states would have to get
the legislatures to change state statutes to no longer require federal
certification of their voting machines.

The state with one of the best, most economical voting system is
Oklahoma who programmed their own paper ballot voting system rather
than buying one from a vendor so OK uses standard optical scanners to
count their paper ballots.  I would think that this means that OK
could possibly have an open source voting system.  I heard that OK
decided to forgo taking Help America Vote Act funds for a new voting
system.

Cheers,

Kathy