Re: code validation?

From: Ron Crane <voting_at_lastland_dot_net>
Date: Sun Feb 27 2005 - 21:23:49 CST

On Feb 26, 2005, at 9:09 PM, Ed Kennedy wrote:

> As for replacing the defective equipment like the Diebold Acuvote DRE,
> in San Diego County, California, (my home) it depends on how sharp
> the county officials were who negotiated the original deal. If I
> understand it correctly, in San Diego County there was something like
> a performance guarantee written into the purchase contract for the
> equipment. It seemed that the County officials had been tipped of
> that there was something of a 'bad smell' associated with Diebold.
> When the California Secretary of State decertified the Diebold
> Acuvote, Diebold had to eat the costs of a warehouse full of voting
> equipment and step up to the plate with optical ballot scanners at no
> additional cost.
>
> However, in Riverside County, California, they are probably going to
> have to pay the entire cost of replacing their election system.
> Considering the hubris of the Riverside officials, it might be
> appropriate for something like that to happen but who am I to judge?
> With the awful smog they have there I'll cut them a little slack.
> While I suppose a county could try to sue for product defect,
> misrepresentation or some other grounds to either break the contract
> or even recover funds, I wouldn't expect the plaintiff to have much
> success.

Overall, it seems unlikely that there will be much large-scale
replacement of recently-purchased voting systems, unless, as you
illustrate, replacements comes for free. But with the right kind of
backing, we could make it well, not free, but much cheaper than
buying and supporting existing systems.

> As for Secretaries of State and the issue of partisan adminstration of
> elections, this is a knotty problem. A Secretary of State is a
> position that is usually defined into existence in a state
> constituion as an elected (read partisan) official. Please be aware
> that as I consider the phrase, "Non partisan elected official as
> elected in a non partisan election," to be something like the tooth
> fairy, similarly I believe suggestions that the postion should be
> non-partisan to be a non-starter. I have considered the idea of a
> civil service postion of an election commissioner but as I am a civil
> servant I know that ultimately a civil servant has to answer to the
> people through their elected representatives. So, I'm open to
> suggestions.

But we can at least bar secretaries of state and possibly other
elections officials from high positions in campaigns (e.g. Katherine
Harris and Ken Blackwell), though the effect might well be more of a
feel-good pill than a cure. I need to consider this topic more deeply.

> I confess that I have the pervasive and nagging doubt that many
> elected officials deem elections as either something that they would
> not like to leave to chance or as just an inconvenient formality that
> they have to dance around.

I share your doubts. Far too many officials treat vital parts of the
process (e.g. provision of sufficient machines, use of the correct
ballot for each precinct, actually conducting hand recounts when the
law calls for them, truly random selection of precincts for such
counts, etc.) as optional inconveniences. But we're not going to change
that culture. Instead, we have to create a voting platform that works
securely despite it.

> In general, for the security/integrity issues, I like the idea of
> harnessing the aminosity and distrust that political parties have for
> each other in verifying the integrity of software and the election
> results.

Agreed! That's part and parcel of our public-comment and open-source
model.

> Specifically, Ron I dislike the idea of OVC going into the election
> system business. It leaves us wide open to the charge that we're not
> truely neutral and that we have a vested interest in promoting our own
> system. I'll admit that many of your concerns are valid and could be
> solved by having OVC or an affiliate make and market a system. None
> the less, I'm uncomfortable with the idea.

We'll be charged with that, and with other villainous intentions and
acts, no matter what we do. The very fact that we're writing software
implies that we have a vested interest in someone using it. That we
hold events primarily in the Bay Area will, by some, be used to imply
that we are not to be trusted. That our project is open-source will be
misunderstood by many, and vilified by others. All significant
public-benefit entities routinely are subjected to similar criticisms,
most of which are unfounded and many of which are even recklessly
defamatory. It comes with the territory.

But as for become a vendor, we're betting the entire project on someone
using our software to do so, or upon some existing vendor adopting it.
I give the second "Fat chance!" odds [1], so it's got to be the first.
Yet who, when the rubber meets the road, will enter the market with our
software? The market for elections systems is fixed and of limited
size. With recent scrutiny, the requirements and red tape are much
thicker than they once were. Many jurisdictions recently have purchased
systems, and are unlikely to replace them; many more will purchase them
before a platform based upon our software makes it to market. Basically
the profits are limited, which limits the financial incentive to enter
the market. What other incentives, then, are there? Only two come to
mind: the incentive to cheat, and the incentive to do good. A vendor
motivated by the former will be recalcitrant about conducting and
incorporating comments from public review [2], and probably will use
the hardware-based cheating techniques I've recently described to make
such review meaningless anyway. A vendor motivated by the latter
well, that's us or, at least, that's us with good financing.

What are some of the benefits of becoming a vendor? Aside from the
hardware security benefits I discussed earlier, we would get a platform
to market more quickly than a vendor (who would lack our detailed
knowledge). As a public-benefit foundation, we wouldn't need to profit
shareholders, so we could sell (and support) our platform more cheaply
than ordinary vendors; if we got particularly good financing, we might
even be able to give it away. This would give us a marketability
advantage, which could, in time, help drive the bad and doubtful
platforms off the market and out of the polls.

Becoming a vendor would be a big commitment. But I think it's a
necessary one, and a good one.

-Ron

[1] Why would an existing vendor adopt our software? Wouldn't it be an
implicit admission that its existing software is insufficient and
unfixable? And wouldn't it involve a big writedown of existing capital?
That'd be trouble with shareholders, even though it might be mostly
non-cash.

[2] Assuming, of course, that robust legislation is enacted to require
them to do so. But that's a doubtful assumption. Recently I reviewed HR
550, which requires, among other things, complete disclosure of voting
system software. I talked briefly with one of the bill's sponsors, and
he said, point blank, that he didn't think the GOP would even give it a
hearing. Much less, I think, will the politicos actually pass a bill
mandating David Mertz's recommended procedure (see, e.g. the 3/22
messages titled "code validation"). Not for a long time, anyway.

_______________________________________________
OVC discuss mailing lists
Send requests to subscribe or unsubscribe to arthur@openvotingconsortium.org
==================================================================
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
==================================================================
Received on Thu Mar 31 23:17:02 2005

This archive was generated by hypermail 2.1.8 : Thu Mar 31 2005 - 23:17:09 CST