Re: Critical analysis of VoteHere

From: Alan Dechert <alan_at_openvotingconsortium_dot_org>
Date: Sat Dec 20 2003 - 12:39:39 CST

David,

> I think it is important to be able to compare and contrast OVC to other
> systems proposed and in use out there. It is important in order to
> articulate the strengths (and weaknesses of OVC). Also, other people will
> ask for for those comparisons, or will make them, and we have to be able
to
> answer correctly and fairly.
>
I agree totally.

However, we also need to focus on the core goals. The demo continues to
slip and this is costing us. I've had reporters call and ask, "where's the
demo."

We should not spend a lot of time analyzing other projects. We've been over
this quite a bit already. Professor Brady and I spent a lot of time doing
that in 2001 and, frankly, the picture is not dramatically different. The
first presentation I heard on VoteHere technology was in the office of Kevin
Shelley in 2001 (when he was in the Assembly). One of Shelley's staff
members was describing to Henry Brady and me what the VoteHere people were
telling him.

The only thing that's really different in 2003 (ouch, almost 2004) is that
people seem to be serious about trying to add printers to DREs. Saltman and
others have pointed out that this is not as simple as it seems. For
example, Saltman wrote:

     Due to the fact that the printout is created by the computer
     and is not a document-ballot, such a printout is a sop to the
     layperson ignorant of the inner workings of computers.

http://www.fairvote.org/administration/papertrail.rtf see #7

Since the authentic vote is on paper with OVC system, Saltman refers to it
as a document-ballot See #8 -- Saltman didn't have all the details right due
to the fact that the idea has evolved slightly (e.g. he has "mark-sense"
ballot reader where it is now barcode scanner -- although this could change
again for the production system).

Last Spring Dill asked me why Saltman wouldn't sign his resolution. I
explained to Dill why the voter verified paper trail was not a solution. It
is an essential feature of the solution, but, by itself, it is not a
solution. And could actually make a crummy DRE into something even worse.

It's not even clear that you still can call the unit a DRE if you add a
printer. It really depends on how the election is administered with such a
goofy thing. If you say that the "receipts" will only be used in a recount,
then the way to go if you're a crook, is to figure out what conditions
trigger a recount and make sure that this never happens. Say a recount is
done only where the margin is less than x (say .5 percent), then you make
sure that your candidate wins by more than that. Make sure the printouts
always match what the voter intended then record the votes however you want.
There are lots of other ways to rig a DRE with a printer. For example, if
the machines knows that the voter can't read (e.g., because they're using
headphones), then give the voter the right oral feedback but print the
receipt and record the vote however you want. It all depends on how the
election is to be administered with these. But, generally speaking, I do
not consider a DRE with printer to be a valid voting machine.

The OVC system is NOT a DRE system. Votes are recorded electronically, but
those electronic ballot images are conditional and only become valid ballot
images when they have been matched with corresponding paper ballots.

Alan D.

==================================================================
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
==================================================================
Received on Wed Dec 31 23:17:15 2003

This archive was generated by hypermail 2.1.8 : Wed Dec 31 2003 - 23:17:19 CST