Re: [ga] Re: A Question for the Candidates

[Barbara Simons <simons@acm.org>]

Sigh....  I'm sure that others on this list understand what it means to have
a life in which one has other responsibilities in addition to ICANN related
dialog.  In addition to this thread, there have been multiple conference
calls relating to the election, and believe it or not, there are other
issues in this world that are important.  Indeed, some are even more
important than ICANN.

I have one correction to make to my previous posting.  When I was talking
about voting costs below, I mistakenly referred to email voting.  My remarks
were intended for computerized voting.  As I said in another posting, Santa
Clara County, which is the home of Silicon Valley, is about to spend $20
million to purchase computerized voting machines with NO voter verifiable
paper ballots.  This is a lot of money in normal times, and it's a huge
amount of money when state and local governments are being starved by a
combination of economic hard times and a federal government that is
redirecting funding from social needs to tax cuts and military expenditures.

I don't know how to respond to Jeff's comments, since they ignore the basic
issue, which is there is absolutely no way to know what is going on inside
of a computer.  Even if the voting companies allowed us to see their
software, which they don't, computer scientists are well aware of the
difficulties of finding bugs or trap doors in software - which is the
primary reason why there is so much buggy software floating around.

But even if there were a miracle and we were given the software by these
companies that are running our elections and even if we could verify that it
does precisely what it is supposed to do, we would have no way of knowing
that the software running in the machine is identical to the software that
has been verified. 

I don't care what kind of voting and tabulation algorithm one claims to use.
The bottom line is that there is no way of knowing what the machines are
doing.  That is why hundreds of computer scientists, including some of the
most prominent members of our profession, have signed David Dill's petition
on computerized voting.  If folks on this list are interested in signing
(David is also taking signatures from non-computer scientists, including
attorneys), you can find a link to the petition at David's website at
http://verify.stanford.edu/evote.html.

The only solution that we have come up with that provides the protection one
must have to prevent the incorrect reporting or outright theft of an
election (and is not too complicated for the voter or election official to
deal with) is a voter verified paper ballot.  The ballot should be printed
out behind a window, the voter given the opportunity to read and verify that
the ballot accurately represents the person's vote, and then the ballot
should be automatically deposited (so that the voter can observe this) into
a ballot box.

The paper ballots then become the definitive count.  Given the existence of
such ballots, it doesn't matter how buggy the computer code is nor if
someone attempts to steal the election.

Another benefit of the system I've just described is that it makes it very
difficult to sell votes, since there is no "certificate" that the voter can
show to prove that he or she voted as paid.  This is also true of
computerized voting machines with no ballots, but it is not true of machines
that allow the voter to handle the ballot.  In this later case, the first
voter can mark the ballot but not deposit it.  He or she then takes the
marked ballot out to the person who is purchasing votes and gets paid.  The
next voter is given the previously marked ballot, creates a new marked
ballot and deposits the old one.  The new marked ballot is then proof for
payment, etc.  (If you don't follow this example, think about how you would
steal an election if you were paying for votes).

This is truly my final posting on this thread.

Regards,
Barbara

On 3/12/03 4:17 AM, "Jeff Williams" <jwkckid1@ix.netcom.com> wrote:

> However I am puzzled that if you don't have time for such
> dialog, than why you sent this post, and why would you in your
> earlier post on this thread, acknowledge that such dialog/communication
> is important Barbara?
> 
> ( Further comments below Barbara's )
> 
> Barbara Simons wrote:
> 
>> As I said earlier, I do not have the time to engage in further dialog.  I
>> have only one comment to make, which is that computerized voting systems
>> without voter verified paper ballots represent the greatest threat to our
>> democracy that I've seen in my lifetime.  While is may be true that most
>> email voting systems are simple to implement (I would not agree with Jeff's
>> comment that they are moderate in cost vs return - in fact, they tend to be
>> quite pricey), I strongly disagree with the statement that they are quite
>> safe.
> 
> Vote.com has offered to provide the service for ICANN for free.
> We [INEGroup] have also offered our voting system to ICANN
> for cost of implementation.  That's really cheap!
> 
>> 
>> 
>> The bottom line is that computerized voting systems that do not have voter
>> verified ballots make no sense.
> 
> Very much agreed here.  Verification of ballots is not that difficult
> to achieve and has been done in several systems currently in use.
> 
>>  There is absolutely no way of knowing if
>> your vote is appropriately counted, and there is NO BACKUP.
> 
> Can you prove that?  I can prove the contrary...  Are you able to
> except that challenge?  If not, than I would have to say that you are
> simply opposed for other yet to be known reasons to evoting.  Yet
> it is here, growing, and shall be more widely used.
> 
>>  Think of what
>> it means to do a recount in this environment.
> 
> Not a problem with several systems presently being used.
> 
>> You ask the computer what its
>> counts are, and it responds "the same as what I told you last time, dummmy,"
>> unless it is grossly broken.
> 
> What!!??  ROFLMAO!  This is pure silliness here Barbara.  I also think
> you know it is, or I hope you do anyway.  Rolling tally's is a mathematical
> computation algorithm built into most good evoting systems.
> 
>> 
>> 
>> This means that elections can be subverted by buggy software, hackers, or
>> insiders, and no one will be able to prove that this has been done.
> 
> Buggy software, yes in some instances, Hackers, also yes this too is
> possible, but with good systems very improbable.  Insiders, sure, this too
> is possible.  But there has to be some oversight during said elections.
> 
>>  At
>> least with butterfly ballots (incidentally, the Democrat who approved them
>> was a Republican before the election and reverted to being a Republican
>> after the election) we all could see what was happening.
> 
> Oh we could?  What happened to the 18,000 ballots in the two northern
> counties in Fl.?  No one to this day knows.  And in Dade county, what
> about those ballots that were discarded before being reviewed or even
> originally available for counting?
> 
>>  While it's a nasty
>> business to observe an election being stolen without being able to stop it,
>> it's even worse if it is stolen and no one knows for sure.
> 
> I agree completely with you here.  And that is much less likely with
> good evoting systems.  ANd that is why in Florida, that several types
> of evoting is being implemented to replace the butterfly balloting system
> because it is SO much more reliable.
> 
>> 
>> 
>> ICANN is important, but the computerized voting issue is far more important,
>> as far as I'm concerned.
> 
> Both as far as electing ICANN BOD members are intertwined and must
> be dealt with unless ICANN chooses to become less relevant to Internet
> governance than it has degraded to already...  This will be the challenge
> to the new BoD members to step up to, or face becoming illegitimate.
> 
>> 
>> 
>> Jeff's comment that http://verify.stanford.edu/evote.html contains "scare
>> tactics and being panned fairly broadly on some aspect" is absurd.
> 
> Oh?  How so?  It was panned in several publications as you well know.
> 
>>  The
>> webpage was created by David Dill, a Stanford computer science professor who
>> has no financial interest in any e-voting system and who is concerned only
>> with trying to preserve democracy.  I urge anyone who is reading this email
>> to check out the website.
> 
> Yes you and I exchanged some ideas with David as I think you recall.
> He does however have another agenda, as you also know.  He of course
> is entitled to his opinion.  Other experts, including myself have a somewhat
> different opinion.  I agree that "DRE" and touch screen voting machines
> are a very weak and insecure evoting system.  Other evoting systems
> are much different and better as David well knows...
> 
>> 
>> 
>> I believe that computerized voting machines without appropriate backup
>> represent an enormous threat to democracy in any country in which they are
>> employed to elect governments.
> 
> Very strong blanket statement here Barbara.  To bad it is a bad position
> to take as it is broadly unsupported.

--
This message was passed to you via the ga@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe ga" in the body of the message).
Archives at http://www.dnso.org/archives.html