Are you sure that wasn't the US census? I suggest
you do some research on Herman Hollerith who started IBM.

Anthony Kerstens P.Eng.

 

2 problems with this. OnLine elections would be a horrible mistake. MS just got hacked. Don't try to tell me that with millions of dollars invested in the outcome of an election, nobody would try to win by hacking the central database. Any computerized election process MUST be completely local, and be disconnected from the
internet.

Problem #2. In the US, we have a extreme interest in the anonymity of the voter. This dates back somewhat into history, but the idea
was that if you can identify how someone votes, then you can come after anyone who voted for your opponent later. Also, there are organizations (especially unions) that would *LOVE* to be able to hunt down anyone who did not vote *their* way. Whatever system is used, the identity of the voter must remain disconnected from the vote after it is turned in.

I did like the idea propsed earlier of being able to take my vote and verify it before turning it in. This would allow me to vote, certify my ballot, and then turn it in. This would eliminate anyone trying to interpret (read change) what I really meant later. Perhaps this
could be enhanced using PGP technology so that I vote, run my ballot through a verification machine, make sure all my choices are properly
read, and then the 'verifier 9000' machine would print the key characters at the bottom of the ballot. That way, even if someone tampered with the ballot, the intent could be determined using the code at the bottom of the ballot.


- --Joe Jansen

 

> ... A touch screen allows
> candidate selection and validates that the voter does not double
> select and queries any non-selection ("...are you sure you don't want
> either candidate for dog catcher?")...

... "No, I'm not sure I don't want either one."

... "Yes, I'm sure I don't want either one."

--
Ken Irving

 

Here's an argument against all of this beautiful technology.

Unless I directly create the permanent, physical record of my vote, then I don't trust it. Once I make it, you can add them up with any technology that you'd like to use.

We've had far too many scams with computerized systems.

For example, I recall a scam where someone was making "aftermarket" versions of firmware for gas pumps. The firmware would short the customer
a few percent on every transaction. However, the firmware would issue the proper amount of fuel if you pumped an even number of gallons. Therefore
the bogus firmware was in operation in a lot of pumps because the government agency that audited the pumps always pumped an even number of gallons.

Another case in point - credit card fraud. Credit card fraud happens every day through a variety of means. The system is not designed to be invulnerable. Credit cards are secure enough that the losses of your bank are less than the amount required to move to the next level of security. I don't think that you can ever make such a judgement with voting.

The moral of the story - no matter how good your validation test is, I can probably figure out a way to get around it.

There are plenty of ways to mess with a paper ballot system, but there are more ways to mess with a computer system and the ways can be very
subtle. If we're not able to eliminate all of the not-so-subtle possibilities for paper ballots, how will we eliminate the subtle
possibilities of computer systems?

This is not going to be so simple to figure out. Good luck.

Darold Woodward PE
SEL Inc.
[email protected]

 

I can't resist adding my 2 cents (my opinion, for those outside the US/Canada) Yes, punch cards are in widespread use because they are cheap,
most were purchased years ago, they do work quite well, and they are very simple (K.I.S.S.).
What is needed is: (1) a foolproof voter authentication to reduce fraud (reportedly
rampant in some areas), and (2) (near-)realtime validation of ballots while the voter is at the polls; i.e. a machine that can check each ballot
card before it is cast so the voters can leave the polls knowing that what they executed is truely what they intended (correct holes punched (the Buchannon/Gore debate), and holes
punched correctly (the slew of chad problems)). In essence, a mini card reader, cheap enough to provide one for each voting precinct/polling
place, and private enough so each voters confidentiality and the secrecy of each
ballot is maintained. Voters could check before they left the polls for under-/over-votes, and if
an undervote is what they intended, they could leave it that way.

OK you smart inventors, get working and get those patents filed.

Alex Pavloff wrote:
> Votomatic machines are in use because 1) they're paid for and 2) they pretty
> much work.
--
********************************************
Phillip G. Costantinou, P.E.
Instrumentation and Controls Engineer
I.C. Thomasson Associates
2950 Kraft Drive
Suite 500
P.O. Box 40527
Nashville, TN 37204-0527
(615) 346-3400 Switchboard
(615) 346-3431 Direct to desk
(615) 346-3550 FAX
[email protected]
http://www.icthomasson.com

 

I am not insinuating that computers are not involved now. I am insinuating that if we go to networked internet connected voting as has been suggested that I have an issue, I am less concerned with stand alone counters but have an issue with Internet voting.

I do web sites on the side and it is unbeleavable how many attempted worm and hacks I get in a week. If you knew how many attempts were being made to your home when the phone is connected you would be real nervous..........

Dave Ferguson

Blandin Paper Company
UPM-Kymmene
DAVCO Automation

 

Interesting and scary points about Internet although somehow expected . What software tools do you use and how do you::
-detect the worms/hacks
-stop them?
Perhaps naive questions but before doing anything serious over the Internet I'd like to learn more. I was just considering a Labview internet
application due to NI's new (and free) Labview Player that allows VI's to be run in a webbrowser . That would cut software distribution costs dramatically for me down to $0 , unless I get the hidden problems you're mentioning and the support will cost more than the software.

Matt Tudor,MSEE
http://personal.lig.bellsouth.net/~mariusrf

 

> Votomatic machines are in use because 1) they're paid for and 2) they
> pretty much work. I think quite a few of us have heard THAT refrain
> before, huh? Its become obvious now that indeed, they're not good
> enough.

I think that they are good enough but they are, apparently, not easy enough to use. The basic problem with punch cards seems to be that it is
too easy to use improperly leaving the results ambiguous (particularly if you are unwilling to assign responsibility to the operator that they must use it properly).

Regardless of the specific system, is it really possible to come up with a system that can count votes to an accuracy of 0.000001% ? This is the
accuracy required for a national popular election (1 in 100,000,000) in the US. Or even 0.0000167%? Which is what Florida alone would require (1 in 6,000,000). You are approaching the bit error rates caused by cosmic rays here.

The only reason that there is such a fuss is that the vote was a statistical dead heat and the resolution of a dead heat in the US constitution doesn't involve a direct public vote. Some people seem to have a problem with that.

Regards,
Ralph Mackiewicz
SISCO, Inc.

 

I recently heard one of the talking heads on cable news coverage mention that punch cards were first used by the U.S. census in the late 1800's, and I think that IBM played a role. You can be the judge as to whether the commentators can be
trusted.

David L.

 

The problem isn't that it can't be done -- here in Sonoma County, CA, I didn't use a Votomatic, but a printed ballot that I took a provided marker
to mark in boxes. The ballot was not a generic ballot with columns labeled 1 to 100, but a ballot that actually listed the candidates for president, my congressmen, and all the issues. Someone can make a system that reads the ballot more accurately than the Votomatic (man, even the NAME is straight out of the 60s...!)

The problem is one of cost. Many counties in America (I heard like a 1/3rd) have Votomatic machines that are paid for. Who is going to pay for brand new voting equipment?

One can apply the same logic for industrial systems, after all. Joe Blow's Industrial Company has X machine that has worked for 30 years... sure there are machines that are 1/8th the size, twice as fast, and can talk to all the
latest SCADA packages out there, but Joe Blow doesn't have the money to spend on it, even though it'd be obviously better.

 

I agree with Phillip

The problem is not the Votomatic, it's the various errors the voter has introduced.

If a simplified "card reader" were provided at each polling station each user could slide his card into the reader and verify his vote within
seconds.

This could be a simple device that mimics the current Votomatic, but with a light source below the card. A light would shine through each punched hole, lighting up a small indicator next to a candidates name. The voter could turn each page as he does in the Votomatic and quickly verify his selections.

Sounds simple and inexpensive.

Care to start a patent application Phillip ??

Mark Hill
[email protected]

 

> I do web sites on the side and it is unbeleavable how many attempted
> worm and hacks I get in a week. If you knew how many attempts were
> being made to your home when the phone is connected you would be real
> nervous..........

I recently installed the free personal firewall, ZoneAlarm, on my home cable modem after a colleague told me that he noticed numerous break-in attempts after he did. I am amazed. I had no idea at the number of attacks my home computer was getting.

I highly recommend that everyone go to: http://grc.com/x/ne.dll?bh0bkyd2
and check out the tools and links they have there. Just because someone hasn't broken in yet, doesn't mean that there isn't anybody trying.

Regards,
Ralph Mackiewicz
SISCO, Inc.

 

The biggest undiscussed problem with this election isn't the technical end. The real issue is the identification of unauthorized voters. There are some simple technical solutions to this issue, as well, but for one major problem. The databases that must be used to verify who is and who is not a legal voter are not being permitted to be used.

Here in Washington State, several races, including the US Senate race, were decided on less than 2000 votes...and one local legislature race was decided on less than 200 votes. But there is no "controlling legal authority"
requiring that the databases for felons, for illegal aliens, for death certificates, for birth certificates be used to determine eligibility of
voters.

If we had access to those databases, it would be possible to even make Internet voting work.

Walt Boyes

---------------------------------------------
Walt Boyes -- MarketingPractice Consultants
[email protected]
21118 SE 278th Place - Maple Valley, WA 98038
253-709-5046 cell 425-432-8262 home office
fax:801-749-7142

 

>> Votomatic machines are in use because 1) they're paid for and 2) they
>> pretty much work. I think quite a few of us have heard THAT refrain
>> before, huh? Its become obvious now that indeed, they're not good
>> enough.

Harris County, Texas, (Houston area) announced a couple of weeks ago that punch cards will be discontinued for all elections beginning November 2001. Although paid for, the machines are old and failing at an increasing rate. Election
officials have opted for less expensive electronic systems rather than replacing the manual system. I think this highlights something missing from this rather interesting thread. There are real budget pressures on governments to reduce expenses. Few official will fight for money if it is to be spent on votomatic machines that will do no more than collect dust for all but a couple of dozen days over its useful life. Why not, in the minds of officials, make use of the rather large investment already made to put computers in schools and librarys around the country.

>Regardless of the specific system, is it really possible to come up with a
>system that can count votes to an accuracy of 0.000001% ? This is the
>accuracy required for a national popular election (1 in 100,000,000) in
>the US. Or even 0.0000167%? Which is what Florida alone would require (1
>in 6,000,000). You are approaching the bit error rates caused by cosmic
>rays here.

As an engineer, if I run an test in which the result is less than the margin of error for the technology used to measure the results, then I declare the test inconclusive and I re-run it. Our election laws don't allow for this, but maybe
they should. It certainly is rational logic, and follow-up elections are the norm in many other democracies acround the world.

David L.

 

You've nailed it, Ralph.

Regardless of the voter input method, there will always be one or more of the following:
a) a margin of error (this comes into play in tight elections such as this year)
b) opportunities for voter fraud
c) confusion
d) significant consumption of time

I just wonder if people really feel that the voting procedures are flawed or if some of the debate stems from party loyalty.

I think the real problem is that in a race this close, there is the very real chance that a candidate can win by a single vote. Unfortunately, none of the methods I've heard discussed so far can give that kind of accuracy
consistantly. That's when I think all that can be done is to implement and adhere to laws specifically aimed at handling such situations.

Philip L. Glass

 

Ralph & List,

Thank you so much for this link! I just ran all the tests on this site, and I am relieved to know that my computer is a "black hole", that is, it looks to a hacker as though my computer doesn't exist. This is probably because my connection (since October 1) is through a VSAT terminal. There must be some pretty fancy firewall equipment on the other side of my link! However, I have friends who HAVE had hack attacks, especially with cable modems. A lot of IT professionals (myself included) have one or more computers on 24/7 with a full time internet connection; it could be embarrassing or even disastrous to be hacked, and perhaps easier than generally thought, especially during the middle of the night.

Regards,

Willy Smith
Numatics

 

IIRC, it was the 1880 census, and I am old enough to remember programming in Fortran on Hollerith cards...

Walt Boyes <talk about hanging chads....>

---------------------------------------------
Walt Boyes -- MarketingPractice Consultants
[email protected]
21118 SE 278th Place - Maple Valley, WA 98038
253-709-5046 cell 425-432-8262 home office
fax:801-749-7142

 

And oh by the way something we all forget......keep it CHEAP as small town
governments cannot afford them which is probably the primary reason this system is still in place.

Dave Ferguson

Blandin Paper Company
UPM-Kymmene
DAVCO Automation

 

Walt,

Me too! Unfortunately, mispunched cards were not subject to interpretation in any of the classes I took that required programming with card decks. The programs didn't even get rerun, especially if they crashed the computer on the first run. But the chads (we called them punchies) made great confetti!

Regards,

Lou Heavner

 

Does this software tell you if these attacks "bots" that scour the web for data, or a serious hacker?

Anthony Kerstens P.Eng.