Re: A brief introduction and some questions. Follow up.

From: Arthur Keller <arthur_at_kellers_dot_org>
Date: Mon May 03 2004 - 16:19:52 CDT

At 1:35 PM -0700 5/3/04, Edmund R. Kennedy wrote:
> Thanks for your thoughts. It must be nice to
>always know what you're talking about. ;-) I've come
>to realize that using new commodity machines is a good
>approach. Trying to make a ragtag bunch of oddball
>machine of various, makes, models, vintages and quirks
>work reliably and predictably sounds actually more
>expensive than buying new identical machines. Also,
>each machine needs to be capable of burning CD's which
>is not something of which many old PC's would be
>capable. However, commodity machines generally have
>this capability as a standard item along with
>sufficient ram to efficiently run a Live CD and built
>in sound capabilities for the blind. Since I wrote
>the letter you've commented on, I've researched the
>issue of touch screen monitors more and found that the
>costs are not that much different than a regular
>monitor. Also most of these monitors seem to be ship
>as being capable of using several different touch
>technologies that are either software or hardware
>selectable. These include both your recommended
>capacitive and restive technologies. Hardware,
>without a bulk discount, should cost in the
>neighborhood of $500-600US/unit. This is assuming
>that these machines run on line power and are not
>laptops. This is just a rough first approximation.
>It however beats the heck out Diebolds $3,000 each

I have been migrating to that conclusion myself, and particularly in
the beginning. It'll be a lot easier to pass certification if we do.

> A more general question here involves the issue
>of ballot images. David or Alan, I think, mentioned
>that the ballots need to be written in random order to
>the CD (the same CD that would run EVMx) so as to
>assure privacy. Also, this is something that was
>suggested as being done after the closing of the
>polls. In the meantime, I would imagine images would
>have to be written to the hard drive of the voting
>machine. Is my understanding correct? It can be
>notoriously difficult to erase hard drives in such a
>way that the information can't be recovered. I had
>suggested that there needs to be an automatic erasure
>routine built into the EVM program as part of the
>process of closing down the machines. I used to have
>a Symatech program that wrote 0's and then 1's
>repeatedly to any place I specified on the disk to as
>to fully randomize the bits. I recall it being a
>rather time consuming process. Does anybody have a
>better idea?

Elsewhere, we have talked about using flash memory cards to maintain
ballot information without date/time info, but digitally signed. The
ballots could be placed randomly in slots on the flash memory card,
or even placed in numerical order by random ballot ID for that voting

> Finally, about the issue of sealing machines. I
>understood that each of these machines were supposed
>to be put in some sort of enclosure that would keep
>people from fiddling with them or worse. (In the
>unlikely event that no one had thought of this, these
>enclosures would surely need supplements cooling.)
>These enclosures seem to be something that could
>readily be sealed simply by something as simple as a
>plastic loop with a unique number. This would be
>affixed after the CD has been removed and the erasing
>completed, by the election judges or representatives
>of the competing groups in a manner similar to sealing
>a ballot box.

I think we've not considered those issues enough.

Best regards,

Arthur M. Keller, Ph.D., 3881 Corina Way, Palo Alto, CA  94303-4507
tel +1(650)424-0202, fax +1(650)424-0424
= The content of this message, with the exception of any external 
= quotations under fair use, are released to the Public Domain    
Received on Mon May 31 23:17:05 2004

This archive was generated by hypermail 2.1.8 : Mon May 31 2004 - 23:18:15 CDT