I’m writing up a full point-by-point review of the ITIF eVoting report. [Update 9/20/07: It’s written. Here is the point-by-point review]
For now, here is a quick summary of my impressions.
I agree with the basic premise of the report that the debate about electronic voting needs to be broader and include other verification technologies than voter-verified paper audit trails. I am in basic agreement with the policy recommendations of the paper but I feel that these recommendations need some caveats. I discuss the recommendations below.
I disagree with much of the setup of the report. The susceptibility to fraud of electronic voting machines is downplayed too much as is the ability of voter-verified paper audit trails to mitigate that. The tone of the report when talking about organizations promoting voter verified audit trails or promoting distrust of eVoting is absolutely poisonous and Mr. Castro should be ashamed. I suspect that much of the poor reception this paper is getting is due to that.
I fear that this report will be used as a motivation to continue to use eVoting without any sort of verifiability (paper audit trails or otherwise) rather than used to encourage the use of better verification technologies than voter verified audit trails. Some of the reactions on the web suggest that this paper is received that way. For example, Tech dirt’s review focuses on the paper trail bashing rather than the potential of better verification technologies and, in fact, completely misses that the report is proposing the need to refrain from legislating away better verification technologies:
A think tank has released a report bashing the idea of requiring paper trails for e-voting systems. The logic behind this uses some sleight of hand and some misdirection to make such a statement actually try to sound sensible. The key argument the group makes is that a paper trail would not increase security while increasing cost. That’s actually true — but that’s not the point. People aren’t asking for a paper trail to increase security. They’re asking for a paper trail to make the machines auditable so the machine’s ability to count accurately can be checked. In response to this, the think tank notes that the paper trail might not be perfect, so it’s a waste. They point out that printers jam and the hand counts of paper trails may not be accurate either. That’s nice, but again it’s missing the point. Without those things, there’s simply no way of knowing whether or not the computer count was accurate or whether the votes were tampered with.
The recommendations
The report makes three recommendations:
- Congress and the states should allow the use of fully electronic ballots, not restrict electronic voting systems to those that create paper ballots.
I do not fully agree. For such a recommendation to be acceptable it must be coupled with the system having an acceptable verifiable audit trail. It is my fear that this report will be used to justify continued use of electronic voting systems without any sort of verifiability. - Congress and the states should require that future voting machines have verifiable audit trails, not require machines with verifiable paper audit trails.
I agree. I am concerned that this recommendation does not limit the continued use of non verifiable systems that are currently in use. I am also concerned about the details of what is considered an acceptable verifiable audit trail. - Congress should provide funding for the US Election Assistance Commission to issue grants for developing secure cryptographic voting protocols and for pilot testing new voting technology.
I agree with the principle of this recommendation. Ideally funding is for open academic research of voting technology. I am unsure if the EAC is the correct vehicle for providing this funding.
All About Voting 
Pingback: The BRAD BLOG : 'Daily Voting News' For September 21, 2007
You are a software developer. You will always think that some use of machines in elections is OK. Computers are good in every situation where the users have a vested interest in honest tabulations. I couldn’t live without mine. But they have no place in elections.
Elections are simple. There is no heavy number crunching, only elementary school level addition. Computers only add more opportunities for dishonest people to steal elections. The energy , time and money spent attempting to make sure the machine counts are reliable is many multiples of the amount necessary to conduct self-auditing hand counts.
Nancy Tobi of New Hampshire has a handbook of how the hand counts are conducted, just search her name at OpEd news.
Don’t get me wrong, I’m a computer nut. But hand counts of paper ballots have been conducted since the beginning of democracy in order to have a solid record as well as to prevent the theft of votes. True, some jurisdictions have power structures that can’t be beaten, but most precincts in the US have enough people of multiple parties to insure the vote is tabulated honestly, right before their eyes.
We could eventually protect computerized elections, perhaps, but the structure would look like Ft. Knox and cost about as much.
I am in agreement with you about the three recommendations of the report.
Thanks, Ping.
Off-topic: Ka-Ping Yee has introduced a useful tool in evaluating election methods called ‘Yee pictures’. You can read about them here:
http://www.rangevoting.org/IEVS/Pictures.html
Good work.
Chris Bundy:
>You are a software developer. You will always
>think that some use of machines in elections
>is OK.
As we both know, many of the opponents to the use of electronic voting machines in elections are computer scientists and software developers. Read through comments in Slashdot articles about the topic if you are not convinced.
>[Computers] have no place in elections.
>Computers only add more opportunities for
>dishonest people to steal elections. The energy,
>time and money spent attempting to make sure
>the machine counts are reliable is many
>multiples of the amount necessary to conduct
>self-auditing hand counts.
I am largely in agreement. I don’t feel that computers should be involved in elections unless there is a clear overwhelming benefit and fraud prevention is built in by design.
I think that end-to-end verifiable voting techniques offer such a chance and may tip things such that the purported benefits of voting machines are available without concern about fraud. Honestly, I am not convinced that voting machines are worthwhile even if they were used in an end-to-end verifiable system as there are still significant cost and reliability issues in my view. However, if the fraud issues are sufficiently mitigated I will not strongly oppose voting machines and let jurisdictions make their own choices about them.
>But hand counts of paper ballots have been
>conducted since the beginning of democracy
>in order to have a solid record as well as to
>prevent the theft of votes.
Unfortunately, the US has a long history of election fraud much of which occurred with paper ballots. Hand counted paper ballots simply is not a good enough solution.
Thanks!
My website has the original article proposing these election-method diagrams. I neglected to give them any name. I think it was Warren Smith who first started referring to them as “Yee pictures.” It’s flattering but not that informative, and I’m certainly open to suggestions.
Pingback: Daniel Castro’s response to my ITIF review « All About Voting