Re: NPR Show on e-voting, Raba report

From: Alan Dechert <alan_at_openvotingconsortium_dot_org>
Date: Mon Feb 16 2004 - 10:06:37 CST

David,

> The report is at servesecurityreport.org
>
Eventually, I read the report. In Appendix C you describe an alternative to
Remote Unattended Internet Voting. It bears some resemblance to the Remote
Attended Internet Voting (RAIV) idea I have described (FYI, there is a brief
mention of it in our obsolete NSF proposal
http://www.openvotingconsortium.org/ad/NSF_Voting_Proposal_Draft_7.pdf )

You kiosk system is something like my RAIV system without the Internet.
Respectfully, I don't like your system. It may solve some problems but it
has too many other problems.

Like,

- you say the ballot data will come on DVDs. How many DVDs is that? For
100,000 ballot styles? Are you including audio files for these 100,000
ballot styles? If not audio, then you need some text-to-speech which is
highly problematic. I think you have a serious problem finding the data for
a particular ballot style. Are you thinking there will be a juke box of
DVDs at these kiosks? If you limit to military personnel overseas, that may
be feasible but it may not scale well for a more general system of remote
voting.

- I agree, of course, on the ballot printing idea (which gets mailed). But
if you don't transmit the electronic ballot image (EBI), we have to wait for
the mail in order to know the election result if a contest is very close.
These mail-in ballots are at risk. What if it is known that only a few
ballots left to be counted could throw the election one way or the other?
The remaining ballots become a target for destruction.

We need a system where the election results can be known quickly. This will
minimize the temptation to monkey with ballots as they trickle in toward the
end of the count. We want the tabulation done and published before anyone
knows how close/not close the contests are. If we have both paper ballot
and the EBI, these can be checked during the canvass period so that the
count can be verified. The kiosk approach you have could be improved if you
could transmit the EBIs on election night so they may be included in the
tabulation right away.

The RAIV system I have described could have some of the vulnerabilities you
describe, but the incentive is less since cheating could not go undetected
as the EBIs get checked against the paper.

Assuming we get money for R&D, I would be willing to investigate the
non-Internet kiosk approach as well as RAIV. I not ready to say that I know
that RAIV will work, but I see lots of problems with your alternative.
Soon, I hope, we'll get money to do some S-C-I-E-N-C-E covering such ideas.

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 Sun Feb 29 23:17:02 2004

This archive was generated by hypermail 2.1.8 : Sun Feb 29 2004 - 23:17:03 CST