A Modest Proposal - Design And Produce Our Own Voting Machines

[mhr]

Hi All,

Like most here, I have anguished over the election and the possibility of election fraud. This is made more real, personally, since I observed the ES&S touchscreen machines err in Dallas, TX.

So, instead of using much of our valuable energy fighting for existing election machine changes, I suggest that we short circuit the problem by leading an effort to design and produce an Open Source voting system. This is easier than it sounds and can leverage the vast technical resources at DU and elsewhere. One has only to look at all the successful software developed for LINUX.

Using an Open Source model we can build an election system that is open and transparent to all scrutiny. The end-game would be the gift or sale of the technology to all counties, cities, and sates at no more than cost. If managed properly, we could undermine the existing corrupt e-voting companies and enable fairer voting by cash strapped governments. Eventually voters would demand their votes be counted on trustworthy equipment.

Since I have not the resources of a Soros, I can volunteer only time and skill. To that end, I now volunteer to coordinate this effort and look forward to hearing from all similarly motivated individuals. I also suggest that Skinner and company consider setting up some forum space which will enable the initial and continuing dialog on this effort.

Working together we can lick this problem and lick the Repubs at their own game.

Sincerely,

MHR

[mhr]

1. Kick For Discussion - Let's Stop Bitching And Do Something!

mhr

[scottxyz]

2. I agree

But I want to take it further.

Design some sort of system which would somehow accurately summarize many people's opinions on a wide range of issues.

This system could be used more frequently than once every 4 years on a Tuesday. It could allow people to express a preference for a candidate - or for a policy, or a budget item.

A system that transparently, verifiably reflected the will of many people on a set of issues - a system whose legitimacy were "self-evident" - this would really be something.

The internet is currently filled with millions of opinions all over the place. Could there be a way of consolidating those opinions into a set of policy or budget proposals? (Let's just do away with the candidates and do a 100% direct system.) Could there be a way of then testing those policy or budget proposals in various "SimWorld" type setups, under different scenarios? If a system allowed us to propose, vote on and even test various policies - would the black-box single-candidate once-every-four-year system just wither away?

[mhr]

3. Interesting Idea - I Would Only Suggest We Proceed One Step At A Time

An Open Source voting system would thoroughly challenge the Republican owned e-voting companies and will draw their ire.

Changing minds and opinions is a slow process. Done carefully though, minds can be persuaded eventually.

[scottxyz]

5. Open-source is an implementation issue

But before an implementation is developed, we need a specification or an architecture. Non-programmers (people who understand, say, auditing or security) can work on this.

A specification or architecture needs to be developed. We need to specify a system which takes the opinions of people who may be adversaries and allows them to get together and come up with a consensus - about candidates or policies or budgets. The "legitimacy" or accuracy of the consensus should be self-evident to all participants.

Blackbox addition over a modem is probably one of the worst candidates for such a system. We need to look at things like Approval Voting or Condorcet Voting - and probably specify a configurable system which could support a variety of these more-accurate, more-detailed voting protocols.

I think www.accuratedemocracy.com is a step in this direction. With all the global information systems out there, it should be possible to create a system which supersedes America's creaky Electoral College.

There's lots of opinions out there (too many!) - is anyone looking at ways to hammer out a consensus among them?

[Quick n Low]

7. Open Voting Consortium is already there.

Open Voting Consortium ( http://www.openvotingconsortium.org ) is currently working on an Open Source based system that will run on inexpensive computing equipment and provide a voter verified paper ballot and verification for impaired voters.

Check out their site and if you feel you can help, sign up to be a contributing member and help make voting transparent.

[pinkpops]

4. s'more options

The basic security need can be met by printing serialized blank cards. The precinct voting officials need to hand these out one to a customer. The task of the "voting machine" then becomes one of printing on the card. The voter verifies what has been printed. The software then has no need to be secure in any way, it just has to be capable of displaying choices and printing the voter's selections. This is just PC stuff, no bells and whistles. The task of reading the ballots electronically I will leave to the reader. OCR readers have read bank checks for many years.

[HootieMcBoob]

6. how 'bout a clear box with a lock and a slit on the top

to put the ballot in?

[scottxyz]

8. Love it!

Let me just go one step further:

Use CARBON PAPER for the ballot and have two (or three) transparent ballot boxes.

Let independent auditors total up each box and compare at the end of the day.

Who'da thunk???

[republicansareevil]

12. Nah.

I'd rather my county spend a fortune on error-prone electronic systems that can't be verified and don't have a publicly open counting system. :crazy:

[blm]

9. I think it's an excellent idea. Make the system transparent and auditable

and then call a news conference demonstrating the transparency and contrasting it with BBV.

[jmowreader]

10. Record locking! It's got to have record locking!

Hmm...are there any open-source database managers with record locking on them?

Here's what I'm thinking:

* The system doesn't use Windows because that's too hacker-prone. Linux is cheap and it works.
* Each ballot is a separate record.
* After the voter makes his/her choices, the DBMS sets the record to read-only.
* The whole thing is a client-server system with touchscreen terminals.
* The server contains two CD-RW drives; once an hour or so it burns the ballots cast in the last hour or so to two CD-Rs--this is for security purposes; if one disc gets eaten, the other should still be okay.
* At the end of the voting period, it dumps a final CD-R pair.
* The hard drives--there will be a mirrored pair--will be removable from the system so they can be impounded until the vote is certified. The preferred method will be on sleds.
* If you make this system capable of doing other kinds of opinion polls, cities can rent it out and make some money with it.

[mhr]

11. Kick

eom

[karlrschneider]

13. I'd rather just go to a pencil & paper system. Not that I dislike tech

solutions for some or many "problems" (how many of those are actually 'real' anyway?)...and of course they work well in many applications, but I'm a bit - no, a lot concerned about the obvious avenues for mischief.
For God's sake, look at what 'hackers' are able to do with Viruses, Worms, Trojan Horses and other compromising bits of, uh...bits in cyberworld! The very FACT that there are half a dozen successful companies existing for the express purpose of detecting and inoculating computer users against these threats/attacks ought to be enough of a warning not to trust them to our most basic principles!

[genius]

14. David Dill and the Open Voting Consortium are doing just that

They've created voting computers with a voter verifiable papertrail, using donated old P.C.'s so that it can be done for next to nothing and so that the state can own the proprieorship rights to the software.

[shraby]

15. Paper and pen, it's faster, cheaper and

if the power goes down, it doesn't..also hack proof and verifiable.