[ale] Re: Vote Today - Message from Marie

[Jeff Hubbs]

Greg -

You allude to some significant issues here and I'll try to address them.

On Tue, 2002-11-05 at 15:30, Greg wrote:
> I just voted and specifically looked for those items which came up in
> previous posts.
> 
> First, the second instruction says to "Touch the vote you want to change to
> undo it and then vote again." or some such thing to that effect.  Seems ok
> to me, but intuitiveness is in the eye of the beholder.

The notion of each vote being a "bistate latch with an initial state of
FALSE" wasn't terribly obvious but was explained in the little cards
they were handing out.

> 
> Second, the clerk at the door was not keen on the lack of paper ballots.  I
> asked about recounts & was told that they would just take the totals out of
> the machines again ... "just like the first vote"  ... "yup".  "So a recount
> is somewhat impossible, huh ?"  "yup"  After admitting that he was just a
> simple clerk, I was jokingly advised to take it up w/ the State Secretary of
> State (hehe, redundant, huh?).

Basically, the notion of a "recount" has gone away.  That, to me, is one
place where this system is broken and, IMHO, may actually run afoul of
the law if only in spirit.  An analogy to older voting systems might be
a situation where a punched paper ballot is counted, a recount is
requested, and instead of actually going through the ballots again, the
tally figures are simply re-consulted.  You and I know that that would
be a joke of a recount.  But, with the new system, the point in the
process where human intention becomes a tabulated figure has changed;
it's moved to the opaque realm of the computer.  

> 
> At this time a Field Coordinator (according to his ID - looked like a MS
> drone to me) told me in reply to several of my concerns that 1) they had
> batteries in the machines in case of power failure.  I did not ask as to
> whether they were AAA or D batteries.  

At my polling place, I think I saw a bag with what looked to be some C
batteries on a table.  

> Hopefully they were not the little
> CMOS batteries, though I have replaced only one of this type in my entire
> life.  2) the vote is recorded in three places, one including flash memory.
> OK, so any mistake is now multiplied times 3.  I get it.  Redundancy = =
> accuracy  3) In reply to my asking if I could see the code, I was told that
> the code was audited by Kennesaw State University to comply with Federal and
> State specs.  

I am singularly unimpressed by such a statement.  I'd buy a peer review
with published names and qualifications.

I am hoping that the KSU folks that reviewed the code were
> code knowledgeable in some if not many respects and instead were not the
> Political Science Dept. folks.  Mr. Field Coordinator did not answer the
> question, but what can you expect in dealing with anything touching politics
> ?? (ok, I'll say it "or M$").

I would appreciate an answer to the question of "Who is accountable for
all this?"  Also, the code isn't audited unless ALL OF IT is audited. 
That means any OS code as well.  And, if you think that it's
unreasonable to pick through hundreds of thousands or millions of lines
of source code to verify the integrity of the process, I agree with you
- which only means that that much code doesn't need to be used in a
voting machine.  

> 
> At this time I left before asking any further questions, since my doctor
> told me my blood pressure is starting to get problematic (& I a wee lad of
> 39).  I left feeling as if this was an expensive piece of crap system w/ no
> redundant and independent check on it.  I would rather have the cards and a
> chad problem if it came to that.

I say, believe your own two eyes and know what you can't know.  I did
not know that the card I was handed prior to my sticking it into the
voting machine was in a state that was any different from what it was
when I removed it from the machine; I don't know what information, if
any, was encoded into the card when it was handed to me nor do I know
anything about what was encoded on the card when I handed it in.  It
would have been somewhat reassuring if I could have stuck the card into
a different machine and seen my own votes displayed back to me.  

> 
> As to another point in a post; so why not ask to see the code ?  Sounds like
> a worthwhile project to me.  I mean, did not taxpayer money fund this ?  I
> wouldn't buy a house or car without looking it over and software should not
> be any different.

But you, more than most people, realize that consumers are given little
choice when it comes to software as far as "caveat emptor."  Houses and
cars are individual entities with differing characteristics; that holds
true even for identical spec houses in a cookie-cutter development or
the exact same make and model of car.  With software, even the tiniest
manufacturing variabilities are absent.  And, if there is something
wrong with the software - perhaps a feature that you really rely upon
but is buggy somehow - exactly to whom would you say, "I'm not going to
buy this until you get it fixed!"  I would not accept a new car with an
inoperable turn signal; I'd expect the dealer to fix it on the spot.  I
have no such recourse with bought software; the software is not offered
to me by an accountable party but rather comes from uninvolved and
unknowing cash register operators at the likes of CompUSA.  

> 
> If there is a tight race with plenty of discord, I am sure that all of the
> concerns & issues expressed on this forum will move to center stage in the
> media.  

All anyone has to do is say, "Prove XXX is the winner!" and the house of
cards will fall - at least to the extent that people can be made to
care.  

I'm going to do a new message on this subject shortly; stand by.

- Jeff


---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be 
sent to listmaster at ale dot org.