The Election Reform Network got in touch recently to bring some items to my attention. One concerned the ease of accessing the information in some voting machines. I had seen news reports on this but hadn't read them thoroughly. This time I did.
A full professor at Princeton, Andrew Appel, purchased some voting machines on eBay. Let me backtrack and say why I made mention of full professor. This is not a rank that is handed out for longevity; you have to work for it, and not everyone gets it. Dr. Appel has a lengthy list of publications and presentations for organizations like IEEE (Institute of Electrical and Electronics Engineers) and ACM (Association for Computing Machinery). These are not fly by night groups -- it's the big leagues. Appel has been studying voting machinery for at least 3 years. In short, Dr. Appel is not a crackpot. Nor is he an egghead. He writes well and in understandable language.
His account of buying the voting machines and the initial look inside is recounted in "How I Bought Used Voting Machines on the Internet," (2/07) (http://www.cs.princeton.edu/%7Eappel/avc/)
I also really enjoyed reading his "Effective audit policy for voter-verified paper ballots in New Jersey," also available on his website, as a pdf file. This is an introduction to the feasibility of voter fraud and ways to thwart it through, you guessed it, effective audit policy. The article, dated 3/09/07, is detailed and very thorough but easy enough to understand that I could skim it while in the same room with small children watching American Idol. It's good stuff.
The good folks at the Election Reform Network say the machines Appel bought are the same basic model used in Montgomery County and this concerns them. After reading Appel's work it concerns me too.
Wednesday, March 14, 2007
Voting Integrity Resources
Subscribe to:
Post Comments (Atom)
Posts
5 comments:
Jane, I agree with you 99% of the time, but this is one area that just annoys me no end. I mean, given enough time, I'm pretty sure any computer savvy person, especially a full professor, could hack into any computer or program. The fact is, there are reams and reams of actual voting performance and lists of actual voters that can be used to cross-check discrepencies. There is data going back decades on who comes out to vote and how they do vote. Further,
this is no fool proof voting system
and paper ballots are surely one of the most fraud prone systems out there. There is no extra "paper trail" for paper ballots, and no one demands one, despite how easy it is to just change numbers and if you were discrete enough, a few votes here and few votes there and pretty soon it adds up to a margin of victory. If I were going to commit voter fraud, I'd want paper ballots just because it is so easy to manipulate results.
I don't think that studies agree with you, phillydem, but I will also confess that I am a bit too pressed for time today to look them up for myself. More if and when I do....
Sorry, ACM, the studies are done in the best circumstances. I agree with those who believe the studies are simply not realistically possible. I personally have counted votes on both Phila's old lever machines and the new full-face DREs. I know how easy it was to make a mistake or misread the numbers or transpose numbers or put them in the wrong block on the official return forms. I would not ever want to count ballots by hand. It's just too easy to make a mistake or manipulate votes.
Honestly, this is making a mountain out a molehill.
PD,
One concern I share with others is that the machines have not been tested rigorously by anyone not affiliated with or in contract with the manufacturer. I would prefer to hear test results by someone more neutral, such as the Princeton professor and his students. While he was able to get into the machine easily his students got in just as quickly without having a key. Princeton students may be smart but I don't think, on average, they are more highly skilled at jimmying locks that the rest of us.
Also, the pdf Appel article I pointed to outlined a plan for regular spot checking for voter fraud.
I, too, had not given this topic much consideration. However shortly before the election a regular reader and some time email pal who was one of the people who would be certifying election results wrote me. The training on the machines was terrible, security was bad, and he/she did not think it would be possible to accurately certify the results. This bothered me.
The paper trail recommended would be a back-up, to make sure the number of voters matched the votes, etc. In addition I think it would be well worth the money to spot check for fraud as Appel suggests.
To be sure there is fraud with paper ballots, too, but with people from both parties present while the votes are counted, there are at least some checks and balances. We need to do a better job of cleaning up the voting rolls and prosecuting those who vote repeatedly using the names of deceased or absent people.
If the fuss over the voting machines gets any of these reforms going I'll be happy. But, yes, I am concerned about the voting machines.
Well, you can check out the Cmte of 70's website for a pretty comprehensive review of the new voting machines (and the all the arguments the "paper trail" advocates make) we got here in the city in 2002. The review was done by the city and the committee, two
independent groups.
I can see where some training might not have been as good as it could have been, but then that's what
happens when new technology is introduced in a relatively short period of time. The fact is, it took Philadelphia 5 years to test, review and select new machines to
replace our old lever machines. That gave us time to get very good
training in place, but even with that some of the older poll workers
took more time than you or I would
to figure things out.
I think you should also consider if your source had an agenda, that is, is he/she an advocate of "paper trails" or not?
I was an election judge, too, for 4 years and also certified election results. The concerns about fraud are highly misplaced. Any election board official who's worked at least several elections quickly discovers the same voters come out year after year. You know what results to expect from your precinct; you know how many Ds and Rs come to your poll; and you know roughly, from polls, what the vote will be. I mean, you know how the voting's going just by listening to voters as you set the machine each time. Your own eyes and ears can tell you if something funky is going on pretty much right away.
Why do think the campaigns track turnout numbers at each precinct?
Because they already have a very good idea of what the vote will be.
If the votes don't turn out that way, they will ask questions. That
is check and balance, not some backup paper trail.
I just don't see this as a big issue and I guess I never will.
Post a Comment