Report from Lancaster

From: Eron Lloyd <elloyd_at_lancaster_dot_lib_dot_pa_dot_us>
Date: Sun Apr 18 2004 - 21:13:00 CDT

Hi all,

Sorry for the delay in getting this out; it's been a truly exhausting week,
and I needed the last 24 hrs. to decompress (the last thing I need is to get
burned out at 24 yrs. old!). So, on to the Demo summary:

As I mentioned previously, I saturated the local media with the press release
and personally got in touch with many contacts I've gained over the last few
years. After the fantastic press in the Sunday paper I was given about a
month ago, another prominent reminder about the demo came out two weeks ago.
I tried as best as possible to get the word out.

In preparation for the event, I went to work producing a comprehensive
presentation giving an overview of the problem, based on information from
Black Box Voting, VerifiedVoting.org, and OVC. I printed out a ton of
information from VerifiedVoting.org, our FAQ, an OVC handout, and binders
with HAVA as well as The PA Voting Modernization Task Force report.

For the demo, in which I was pleased to be accompanied by a new OVC member,
Dr. Brian Adams, Associate Professor of Mathematics at Franklin and Marshall,
we unfortunately garnered only a small attendance. The day was just gorgeous,
so it wasn't a surprise. Off the top of my head, we saw about ten people, not
including Brian and myself. Of those, fortunately, were some important
people: two people from a local disabilities advocacy group I work with, the
chairs of the Democrats and Greens, plus a high-ranking GOP committeeperson,
and three other citizens. It made for a good discussion at any rate.

Of course something had to go wrong. I forgot my wedge-box from home, so I had
to swipe one from one of the library desks. Because of time-crunching, I
didn't really test it, wrongly assuming it would already be Code 128 enabled
(like most are). Since I didn't have a programming manual, which you need for
some wedge boxes, I couldn't properly read the paper ballots. I was able to
go into debug mode (which has some fail-over features if you already have the
EBIs on disk), and continued on. From now on I'm going to devote my time to
testing and preventing problems we could incur in with the system (and
documenting them).

I also wasn't able to display the ballot like I wanted to because the
projector I used couldn't use a high-enough resolution (something else to
watch for). I briefly discussed the process instead, which seemed to be good
enough.

A lot of really good questions came up. Most of the people had a hard time
with the whole "open" idea, which is more of a cultural barrier than anything
to do with OVC. If we can come up with good terminology and analogies to
explain this, it would be helpful. The main thing I kept hammering home was
the "transparency" of our system and the critical purpose of the paper
ballot.

One thing our FAQ doesn't adequately address and caught me in my last
interview was the "speed" factor of processing the paper ballots. I made sure
to emphasize that it isn't even an issue with our system, and that results
can be reach just as quickly as any other system. The idea of it being the
same as simply "closing out a cash register" seemed to work.

The interesting thing that I wasn't planning for came when the other fellow
mentioned in the first article, Chuck Gaston of the SAVIOC system, asked to
be allow to speak and demo his system as well. It was probably a great
benefit to have an alternative approach to contrast us against, because there
certainly are fundamental differences.

For those that don't know, the SAVIOC system is a FreeDOS based, ultra-simple
voting system akin to the "Typing Tutor" DOS apps I worked with in school. He
runs the entire thing off a floppy, FreeDOS and all, back to which the voted
ballots are stored. Anyone with experience in the unreliability of floppies
would immediately be suspect. Second, he believes for some reason that paper
ballots are actually "more insecure", so he instead uses a bizarre method of
distributing the precinct results (the actual ballots) to many people,
including citizens, that can then somehow "audit" their copy against the
official results. He says he can also produce a paper-trail if necessary, but
he really only means a paper receipt, which means nothing.

When asked how OVC differs, I broke it down to two obvious things. One, it
does NOT produce an official paper ballot, which I explained is the first
pillar of our system and one which several *thousand* experts have decided is
necessary to protect voter integrity. Two, it doesn't satisfy the second
fundamental criterion, which is a open-source system. Gaston says the program
can be downloaded and tested (in binary form), and he'd be willing to display
the code to third parties for inspection, but he insists on maintaining
control and IP-ownership, charging jurisdictions that want to license his
technology. A third interesting thing that came up in the discussion is if he
thinks a UNICODE limitation is a problem. He didn't see it that way at all.

We had a very interesting discussion, and I think I gave him lots of things to
think about. I did my best to be inclusive, but was also quick to point out
what I believed to be flaws in his approach and review the "OVC" solution.
Who know, we might see him come on board down the road.

While the turnout was poor, I am proud to say that I still think it was a
success. The two biggest benefits came in the form of another OVC member and
we made the front-page (albeit below the fold) of the Sunday paper today,
which is difficult to do even here in Lancaster County. I was probably a bit
too ambitious wanting to jump right to the Secretary of State's office, but I
would like to do this again. One opportunity I might have (and others might
think about doing this too) is to present to the local Linux Users Group. I
plan to use that opportunity to polish the presentation and put it through
the rigors of "Slashdot-esque" technical critique.

Time for some rest.

Eron

---
[This E-mail scanned for viruses by Declude Virus]
==================================================================
= The content of this message, with the exception of any external 
= quotations under fair use, are released to the Public Domain    
==================================================================
Received on Fri Apr 30 23:17:11 2004

This archive was generated by hypermail 2.1.8 : Fri Apr 30 2004 - 23:17:29 CDT