Democratic Underground Latest Greatest Lobby Journals Search Options Help Login
Google

Elections done right: Verifiable voting systems.

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
This topic is archived.
Home » Discuss » Archives » General Discussion (01/01/06 through 01/22/2007) Donate to DU
 
meldroc Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Nov-06-06 04:46 PM
Original message
Elections done right: Verifiable voting systems.
We've all went round and round and discussed the voting systems in current use made by Diebold, ES&S, Sequoia and others. All of them have serious, severe flaws, and our elections are at a very high level of risk of fraud.

But I don't want to talk about the existing flawed systems. I want to talk about doing it right.

In other threads, I've brought up Ron Rivest's ThreeBallot System. It's pretty damn clever in that it (for the most part) allows voters to take home a copy of one of three ballots they cast, and be able to prove they voted, but not be able to prove who they voted for, thus preserving the secret ballot. They can also compare their ballot copy with the copy that's scanned and online on the website, to verify that that ballot's in the system correctly, and because all ballots are publically available, everyone can check the count, make sure the correct number of ballots have been cast, and check the list of voters to make sure that their dead grandmother didn't vote.

That said, ThreeBallot does have some flaws - the paper does now show a few attacks that can be made by fraudsters to cook the election. Alos, you're filling out three ballots instead of one, and you HAVE to fill out many more bubbles than the conventional OneBallot system, so usability becomes a big issue. It's still a big improvement over the current clusterfuck.

I just found another transparent, verifiable voting system - Punchscan. It's kind of hard to describe, but the website shows how it works - you use a magic marker to fill in your choice in a two-layered ballot through holes in the top layer, then you separate the layers, so the information about which choices are made is separated. You scan one of the layers of the ballot, destroy the other layer then the tabulators use a cryptographic algorithm to reconstruct which choices you made (designed to be auditable and observable along each step), and you get to take home one of the two layers of the ballot (you pick which one) and again, you can verify your ballot was cast correctly by comparing your layer with the scan available on the web site, but you can't prove who you voted for, preserving the secret ballot.

There you go, two different ways to make elections better, more secure and transparent and fraud-proof. How would you do it?
Printer Friendly | Permalink |  | Top
porphyrian Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Nov-06-06 04:48 PM
Response to Original message
1. Instant Runoff Voting
Printer Friendly | Permalink |  | Top
 
meldroc Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Nov-06-06 05:04 PM
Response to Reply #1
2. Instant runoff's not bad, but...
I was looking more into ways to fraud-proof elections. I guess I'm coming at this from a computer science viewpoint, having followed the cryptography community for a while, and saw a couple election systems designed for maximum transparency, with ways to verify that you voted, and ways to verify the count, while preserving the secret ballot.

I can see an instant runoff election being held on Punchscan ballots. That would be cool. Instant runoff's harder to do on ThreeBallot - the math there makes my brain hurt, and the ballot would be a huge pain to fill out. (ThreeBallot has the virtue of not requiring cryptography, but it has usability problems. Punchscan does use cryptography for some steps, meaning you might need some trustworthy eggheads to help with auditing and verification. Punchscan's code is open source, unlike Diebold's :smoke: )


Printer Friendly | Permalink |  | Top
 
porphyrian Donating Member (1000+ posts) Send PM | Profile | Ignore Mon Nov-06-06 06:17 PM
Response to Reply #2
3. Right, used in combination. - n/t
Printer Friendly | Permalink |  | Top
 
DU AdBot (1000+ posts) Click to send private message to this author Click to view 
this author's profile Click to add 
this author to your buddy list Click to add 
this author to your Ignore list Fri Apr 26th 2024, 05:12 AM
Response to Original message
Advertisements [?]
 Top

Home » Discuss » Archives » General Discussion (01/01/06 through 01/22/2007) Donate to DU

Powered by DCForum+ Version 1.1 Copyright 1997-2002 DCScripts.com
Software has been extensively modified by the DU administrators


Important Notices: By participating on this discussion board, visitors agree to abide by the rules outlined on our Rules page. Messages posted on the Democratic Underground Discussion Forums are the opinions of the individuals who post them, and do not necessarily represent the opinions of Democratic Underground, LLC.

Home  |  Discussion Forums  |  Journals |  Store  |  Donate

About DU  |  Contact Us  |  Privacy Policy

Got a message for Democratic Underground? Click here to send us a message.

© 2001 - 2011 Democratic Underground, LLC