Re: BBV: Auditing scanned ballots

From: Alan Dechert <dechert_at_gmail_dot_com>
Date: Mon Jun 19 2006 - 13:34:37 CDT

I support most any effort to develop more public oversight on the whole process. So, in that sense, I have no objection to anyone promoting this idea -- I might even support it.

However, it's not the right approach for a comprehensive solution. I think Irwin Mann defined open voting systems very well in his 1993 paper. Note where he talks about "institutionalized protocols."
  a.. there are institutionalized protocols for public monitoring of all components and the electoral process, sufficient to find any hypothetical discrepancy from the intended design, if it should happen to exist.
The extent to which the Votoscope approach might lead to an institutionalized protocol is positive, but Votoscope doesn't define an institutionalized protocol.

The main thing I don't like about the Votoscope approach is that it assumes we can't define and implement the right institutionalized protocols. Since we can't do that, let's just throw in some monitoring here and there. In other words, it's half-assed.

I think we can define and implement the necessary open voting institutionalized protocols. If we have to re-write the election laws top to bottom, so be it. With the right Secretary of State in place in CA, we could build such an exemplary system (would include a transparent certification process) here. Once we have a great open voting system established in CA, it could be replicated in other states pretty easily. Our open voting bill (need reincarnation of AB 2097 to happen soon) and Debra Bowen's election would be big steps toward making open voting happen.

One problem with Votoscope is that it chose an unlikely place to find anything really bad. The problems that would become apparent would have to do with voter intent issues. And setting up a parallel vote counting system isn't necessary to uncover or discuss those. Donna Frye lost her write-in campaign for mayor because thousands of voters that wrote-in her name did not also fill in the bubble next to the write-in line. What would Votoscope tell us about that? Nothing we didn't already know. Donna Frye knew it in advance too. Her literature explained that voters would have to fill in the bubble in order for the write-in to be counted.

Voter intent is a serious problem with hand-marked optically-scanned ballots. The percentage of ballots with such problems is normally small enough that it doesn't change election outcomes. But in instances where the outcome depends on interpretation of voter intent, it makes this system look bad. One of the main motivations for the Electronic Ballot Printer architecture (EBP) is to eliminate voter intent issues. If a Votoscope project underscored this problem and caused EBPs to replace these, that would be nice. Then Votoscope would be obsolete along with the opt-scan system.

Bev exaggerates the potential value of Votoscope and oversimplifies the complexity in making it work effectively. There are very few places that are using such scanners. The vast majority of ballot scanners in use do not produce a ballot picture: they process the marks as the ballots are scanned. So, potential coverage is poor. As for the complexity, I don't know anything about how Votoscope works, so I don't actually know how well it handles all the issues. Maybe Jim March knows something and can add details. For one thing, you have to have the database of contests and candidates in order to tabulate votes. i.e., you need a list of precincts linking them with the contests voted on in each precinct, along with candidate data, party affiliation, etc. Under current election law, can we get this in a usable format? Does Votoscope expect this data in a particular format? Or does Votoscope expect this to be entered by hand? If by hand, this is an enormous duplication of labor and is highly error prone. You also need to know ballot rotation rules. All of this has to be entered into Votoscope perfectly or the results will be wrong. If you're expecting a bunch of volunteers to do this, good luck. Is Bev going to fund this work? How? In order to be anything beyond a mildly useful educational project, it would have to find sustainable funding. Otherwise, it would be done once and forgotten.

Better to make public oversight built-in to the official system.

Alan D.

  ----- Original Message -----
  To: ; ;
  Sent: Sunday, June 18, 2006 9:44 PM
  Subject: [OVC-discuss] BBV: Auditing scanned ballots

  Apologies if this is a duplicate, but it merits discussion.

  BBV is proposing using scanners that take digital pictures of
  real ballots as they are being scanned, and making those
  pictures available to the public ASAP on election day on
  CD ROMs or DVDs

  This would allow the public one way to conduct a full check
  of an election. Some of the software for scanning the
  ballots exists (VotoScope), so it seems feasible when
  the infrastructure is in place to collect the CD ROMS/DVDs,
  and to do their own counting.

  Jim Soper


  OVC-discuss mailing list

OVC-discuss mailing list

= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
Received on Fri Jun 30 23:17:09 2006

This archive was generated by hypermail 2.1.8 : Fri Jun 30 2006 - 23:17:12 CDT