[EM] Re: Election-methods digest, Vol 1 #152 - 6 msgs

[Dave Ketchum]

On Wed, 9 Jul 2003 23:09:21 -0700 (PDT) Alex Small wrote:

> John B. Hodges said:
> 
>>Greetings- Everyone on this list should read the following link:
>>
>>http://www.truthout.org/docs_03/voting.shtml
>>
>>Which makes me seriously question any voting system that requires
>>computers to count or calculate at any stage of the process.
>>

Over the past 50 years we have learned to build computers that dependably 
do EXACTLY what they are told to do - they get used in ATMs, and in many 
ways where errors mean that people can be hurt worse, and even die.  So I 
see nothing wrong with computers being part of voting systems - in fact we 
argue in this list in favor of election methods that need computers - 
think of trying to elect a governor using ranked voting via compuers and:
      Condorcet - counting the ballots is simple, but intolerably tedius. 
  Forwarding precinct results and summing them tedius.
      IRV - worse.  IRV is not bad with only a few candidates and paper 
ballots that can be sorted and summed at a single precinct.  Come to NY 
where we talk of 8 or 10 candidates for governor and millions of ballots.

Now I read of black boxes being sold with conditions attached that 
customers are FORBIDDEN (and the courts seem to agree that the CONDITIONS 
are air tight) to check to see what the contents of these boxes are told 
to do and permit being done.  Unless representatives of poll watchers are 
allowed to check on what is inside these black boxes, the poll watchers 
should refuse to accept anything THESE boxes do as being known to be 
serving their proper purpose.

> 
> Having volunteered at a local precinct on two occasions, I have a few
> observations germane to the above link:
> 
> In Santa Barbara County the voters fill out paper ballots, and the ballots
> are read by optical scanners.  A computer keeps a tally of the results in
> that precinct.  The paper ballots are retained in a locked compartment.


If the computer stacks them neatly and they are kept that way, then if Joe 
was the 6th voter I could find his ballot as 6th in the stack.  In NY, if 
paper ballots go in a ballot box, the contents are supposed to be shuffled 
to prevent such looking.

> 
> At the end of the day the results are sent by modem to election
> headquarters.  At the same time, the hard copies are placed in boxes which
> are sealed, and the seal is signed by all of the volunteers in the
> precinct.  The boxes are driven to a central location, at which point they
> are taken into custody by election officials.  Police guard the central
> location.  I don't recall whether it is mandatory that the volunteer
> driving the car with the boxes be accompanied by another volunteer, but I
> always made sure to have another volunteer with me in the car as I drove
> from the precinct to the election headquarters.
> 
> 
> This system is robust against tampering.  Sure, the digital data can be
> manipulated by skilled people who know all of the backdoors.  The average
> citizen might not know enough (I certainly don't) to ascertain whether the
> digital data is adequately protected.  However, there are still the paper
> ballots, which are the voters' original documents, and they can be used in
> the event of fraud allegations.
> 
> Sure, paper ballots can be manipulated too (see Chicago), but with
> reasonable security measures the ballots can be protected against
> cleverness, leaving only corruption as a concern.  The surest protection
> against corruption is transparency, so as long as the system is
> transparent the citizens can have confidence that the original ballots are
> secure.  Here's how I'd guard them (I have no idea if Santa Barbara County
> uses these measures):
> 
> Keep them in a locked location, keep a large staff of guards, and never
> allow a single individual to guard the ballots by himself.  Keep thorough
> records of who accesses the ballots and allow the general public to
> observe any time an official accesses the ballots for any reason.  While
> the ballots themselves must be in a locked room, all entrances to that
> room should be visible to the public 24 hours per day.  Surveillance tapes
> of the doorways and the interior of the room should be available to the
> public, and any citizen who wishes to should be free to visit the facility
> at any time of day or night and observe the guards.
> 
> 
> So, I share Mr. Hodges skepticism of any all-digital voting technology,
> but I have confidence in any system that retains hard copies of the
> voters' original ballots and has very stringent but transparent security
> measures.  I wouldn't mind touch-screen voting if the touch-screen machine
> produced a hard copy which the voter had to inspect and submit before
> his/her vote is counted.  Optical scanners and paper ballots are good. 
> Punch-cards have proven to be dubious.
> 
> Anyway, that's my opinion based on experience "in the field."
> 
> 
> 
> Alex
> 
> P.S.  Fraudulent voting by unqualified individuals (a perennial
> conservative concern), intimidation of voters (a perennial liberal
> concern), and "vote early and vote often" (a perennial Chicago concern)
> are also important issues, but these issues pertain to voter registration
> and precinct administration rather than vote tabulation.  I'll refrain
> from commenting on this subject in this thread.

-- 
davek at clarityconnect.com  http://www.clarityconnect.com/webpages3/davek
  Dave Ketchum   108 Halstead Ave, Owego, NY  13827-1708   607-687-5026
            Do to no one what you would not want done to you.
                  If you want peace, work for justice.