|
Home Writing (Nick) |
|
|
|
|
PurposeThis document describes a voting machine that complies with HAVA requirements while still producing a system of verification that is transparent to the typical citizen. Assumptions & ContextThe voting systems currently available, while HAVA compliant, have no method for verifying that the vote cast has been recorded faithfully. This has been established elsewhere and hence won't be revisited in this document. Opportunity
The problems with the current voting systems won't go away, therefore
there is an explosive growth opportunity for a company with a product
that can be quickly brought to market to solve the problem soon after
the 2004 general election. Inevitable public dissent following this
election will create a huge market for a viable replacement to the
existing systems. The market is not limited to the U.S. There are several keys to HAVA compliance:
In addition, to be competetive, the systems must be direct-recording and allow for instant tabulation. The SolutionThe bulk of the user interface will be functionally identical to the several competing systems on the market now. They have solved the HAVA compliance issues neatly, but left verification by the wayside in the process. We'll build on the part that they did right:
The current systems all have a final verification display that looks something like this:
President of the U.S. -> John Kerry
Indiana Senator -> Richard Lugar
Indiana Secretary of State -> Todd Rokita
Wayne County Clerk -> Sue Anne Lower
The verifiable voting system adds one step to this system. To the
right of the screen is a window behind which an adding-machine tape
is spooled. The last step shifts the display one column to the right
so that the name of the candidate is printed on the tape next to the
display of the office. The voter then confirms that the tape displays
the correct votes. If the voter does not confirm, "VOID" is printed
on the bottom of the tape and the voter is returned to the screen
menu for correction. If the Voter confirms, the tape prints
"confirmed" under the names and the paper is sliced off and falls
into a bin. All of this is visible the voter.Each precinct has one machine for visually impaired voters with a Braille impact embosser behind the tape. This machine still prints the names on top and normally functions without the embosser. To allow a visually impaired voter to vote, the poll worker simply removes the cover over the tape, allowing tactile access to the tape. This automatically turns on the embosser, thus allowing the voter to listen to the names/offices in order over the headphones, and confirm that the tape has the same names in the correct order on it using the braille. None of this interferes with the direct recording and tabulation of votes, so the slips of paper simply become an easily used audit trail for recounts if the system is ever called into question or if there is a mechanical/electrical failure on the voting machine before the votes are recorded. Because the names are in office order, the use of a "mask" to facilitate manual recounts is possible. (This was one of the advantages of punch-cards.) To allow meaningful testing, the voting machines should be produced using standard hardware and operating systems, thus producing a larger pool of computer experts qualified to do testing on behalf of local election boards. The software itself, although it might be copyright protected, should be published in source form, thus allowing further inspection. The binary version of the code should be compiled under the supervision of a non-partisan group, and an md5 "signature" generated and published to help local inspectors confirm that the version of the program being run matches the version that was certified at compile-time. Copyright, Updates, Comments, Questions.
Copyright © 2004 Nick Fankhauser. The reason I require that the entire work be reproduced is that I will probably update it periodically. I believe that all of the statements are true but I am not perfect and may need to make corrections or perhaps even change my opinion. I also hope to be able to report some progress toward a solution eventually. If you are reading a reproduction, please go to the following URL to make sure you have the most current copy: http://www.fankhausers.com/articles/ I invite comments, questions and corrections. I'm willing to post opposing opinions if they are reasonably well written. My EMail address is nick@fankhausers.com |
|
| Friday, 04-Jun-2004 13:13:28 EDT [Return to the top of this page] |
|