Voting machine source code must be available for review

Congratulations to the Governor and the people of Wisconsin for protecting the voting process with Assembly Bill 627. Although there have been other methods used to manipulate the voting process, especially in the '04 election (voter intimidation, voter suppression, i.e. not enough voting machines in heavily Democratic precincts), at least this law will plug the hole of manipulating voting machine results.

I think that this is a very good idea.

This is significant. The movement to have honest voting in the coutry is getting larger and larger. There is substantial evidence of fraud. By the Republicans. Against the intent of the constitution. Pretty telling.

While distributing the source to the application can make it possible for others to verify the functionality of the software, it does not prevent fraud in any meaningful way. The first problem is that there is no means by which the source available can be verified to be the code actually operating the machine. This just places "open source" as an amusing buzzword.

The second problem is that development of the software is often the largest cost item in producing such a machine. By distributing the source you enable the "cheap Chinese" knock-off where they are freed from all R&D expense.

So tell us again why this is so important?

Please correct your headline, as it is in error. As the text of the article makes clear, this will NOT require that the voting machine code be open source. It requires only that it be available for audit. Open source refers to software which you are allowed, not only to examine, but to modify and redistribute freely. This is a huge distinction.

That said, this is an obvious step to take. It's not sufficient to stop fraud by any means, but any state that doesn't adopt this requirement is positively begging for fraud to occur.

As to Crowleys comments - there IS a way to verify that the code you see is the code running the system, if you have the code. You simply compile that code, and compare the binary that results to the binary actually in the device. This is trivial and obvious.

Secondly, if this bill did require that the software be open source, your second argument would have merit. However, as the article text makes clear, it does not require anything of the kind. Meaning that any 'knock-off' that used that code to shortcut their development process would be liable for massive statutory damages for copyright infringement, as well as subject to immediate injunctions preventing them from continuing to produce or sell their product in nearly every country in the world.

Regarding the headline and the definition of open-source software, I understand people who belong to the "open-source community" can feel somewhat proprietary, if you'll pardon the pun, about the definition of the term, but there's a common, colloquial usage that refers to publicly available code, not necessarily the legality of modifications and redistribution for use.

It's similar to the use of "open" in "open records laws," which refers to availability, not modifiability. (In this case, similar in content as well as word usage.)

On another note, I agree you can never be completely certain that someone, somewhere isn't committing fraud. That applies to paper ballots, which could be forged, shredded or burned if the right (wrong) people are corrupt, as well as to electronic machines. The more factors are involved and open to review, the more difficult that is.

We'll be following up on this story later, especially as municipalities go to work figuring out what will be used in the next major elections.

Congratulations, Wisconsin, on being already the second state to take this necessary step to restoring citizen confidence in US election results. Only 48 states to go.

Voting machine design can cheaply be such that source code is capable of being copyed directly from an active machine to some standard device provided by the curious voter, such as to a USB card. That way the voting individual can be assured of receiving the "real thing" and can print it out or examine it in detail on a home computer, or take it to an expert.

Note to the poster Paul: Have you heard of copyright law, and software licensing? That will protect software considered proprietary by the voting machine vendor from being reused contrary to license terms. Open source GPL software, for example, is protected by copyright and licensed so that distribution requires sharing of improvements.

But, what is wrong with using identical open source code for every voting machine regardless of brand? The programming logic is not complicated. Why shouldn't it be a comodity, like salt? Only the interfacing need be different, and then only IF the machine's interfaces are proprietary (not an unencumbered standard). Won't that lead to less expensive voting machines?

IBM-compatible computers work that way, you know, and run Linux securely on open source software (plus that Redmond o/s and a few others). A digital voting mache is just a dedicated-use computer.

Ben

Contrary to the suggestion provided by bbaston, enabling the voting machines to provide a copy of the source code used in their programming to the interested voter isn't adequate, because there's no proof that that source matches the running binary. Having a low-level hardware implementation that copies the binaries that the device booted against -- such that the binary can be compared against source otherwise distributed -- would be, on the other hand, an excellent approach, so long as one's opponent is incapable of tampering with said hardware.

It may not make electronic voting machine fraud impossible, but certainly should make it more difficult -- and help to ensure that systems inadequately designed are publicly exposed as such.

That said, making the source available for inspection is only one componont: the use of a verifiable paper record (which can be retained for use in a recount) is even more critical: Covering one's trail is considerably more difficult when the record consists of more than bits and bytes.

Congrats Wisconsin!

It will be neccesary to run checksum tests on the source code presented by the vendor and that installed on the voting machines and computers to insure that it is the same unmodified code.

David Allen
www.blackboxvoting.com

Open Source means different things to different people, but let us assume, for the sake of argument, that it means that anyone, anywhere, can see the source code anytime. Assume furthermore that the code is written in 'C' and audited and validated by a mix of people drawn from accredited academic institutions and elected government. A dedicated compiler would generate the executable code, and this again would be open source, and subject to the same constraints.

The voting machine could incorporate hardware designed to generate an MD5(or similar) checksum of the executable; this would ensure the machine was running the correct code.

Mr. Stitt,

With all due respect, you are absolutely wrong. Open can certainly have a range of use in colloqial language, but 'Open Source' has a very specific meaning in the software world and the original headline was indeed simply incorrect. See http://www.google.com/search?oi=defmore&defl=en&q=define:Open+source - every single authority agrees you that open source software necessarily implies legal rights to modification and reuse, not simply that the code has been published and can be viewed.

The corrected headline as it appears now avoids the problem, and you're to be commended for making the correction.

Thanks.

This is a very positive step, not just for the state, but for the whole country. Hopefully other states will follow Wisconsin's lead. However the voting machines should provide an easy method whereby parties can verify that the software actually installed is identical to the source code they are provided. Perhaps after each election, machines should be purged and software reinstalled on the day prior to the next election, after each party is satisfied that their media copy of the code matches that of the installation media copy.

Also, voting machines should only be networked together with cable, not wireless, and should not be WAN connected - no internet connection should exist at all. Any WAN connection to convey vote tabulation should be via a separate computer that is not networked to the voting machines.

Important first steps. Though, as others have already noted, this is not enough by itself to stop fraud, it is the first step toward that achievement. It is also an important message to the likes of Diebold that we won't accept their tinkering with our democracy.

The term I prefer to use is "publicly inspectable software" rather than "open source software." Security and auditability make it necessary that the in-precinct voting systems include a paper trail and have their software be publicly inspectable, but those are not sufficient. A critical component is that they central tallying system have adequate security and audit features. Section 5.84 paragraph (3) indicates that "the municipal clerk shall provide ... the coding for software ... to tally the votes cast." Thus the central tallying system must also be made available.

Open questions are the manner in which the software is made available. For example, is it a hardcopy printout, is it machine readable (could be a PDF "printout"), is it in a compilable form, is it made available on the Internet, when is it made available (before the election or after it), is there a potential for a denial of service attack effectively denying "any person" from getting access to the "coding for the software" (e.g., you have to read it in the county headquarters, make no notes or copies, and only 1 person can look at a time for 1 hour at a time, and the waiting line is filled continuously with Diebold personnel), can "any person" retain a copy of the "coding of the software" after the municipal clerk provides it, how much of the software must be disclosed (what if the system is run on top of proprietary software, like Microsoft Windows or Microsoft Access), is the coding for the firmware included in what must be provided, and does this apply to optical scan voting systems as well as DRE (Direct Recording Electronic) voting machines (I think the wording implies it should apply to both)?

Congratulations to the State of Wisconsin. Now let's see how well this legislation works in practice with respect to the kinds of issues I identified.

Arthur Keller
Founder and Board Secretary
Open Voting Consortium
http://www.openvotingconsortium.org
We are promoting the adoption of open source voting systems

The actual law as signed by the governor is here, I believe:
http://www.legis.state.wi.us/2005/data/acts/05Act92.pdf

Aside from the fact (mentioned above) that the enacted bill does not open the source code, it *does* enforce that the software placed in escrow is the same used to count the votes:
"The verification procedure shall include a determination that the software components correspond to the instructions actually used by the system to count votes."

How they'll do that is undetermined, but at least there's an acknowledgment of the potential problem.

hmm... looks to me someone will probably ID an exploit and compromise the integrity of the process... its like giving a terrorist blue prints of the sears tower.

Now if we can just get democrats to stop slashing van tires on election day we'll be all set.

What about this?

Government opens a bank and gives every person an account.
The accounts are filled with various "currency" Local election dollars, state election dollars, national election dollars, etc. Of course the currency is non transferable.
Every person is given a VISA check/credit card for their account.
To vote you go to a regular ATM, put in your card, your PIN, and Vote.

Advantages: The systems are already in place, tested, and reasonably secure.

During elections you would "buy" a vote for which ever candidate you were voting for. The candidate would have an account and your vote would transfer to their account. The candidate with the most "voting dollars" in their account wins.

Fraud? You could check your "voting statement" or "online voting account" and verify no one is stealing your votes.

You could vote in the line at the supermarket.
If you had questions you could call the 1-800-Visa-Vote number and the customer service rep could review your account.

Thoughts?!?

You could even have two or three Voter Banks to ensure competition?

I would like to point out that Doyle is a Democrat and the state legislatures are controlled by Republicans. So this means both sides came together and got something good done. Shocking isn't it.

Alas, the correction points out once again that the people don't understand how security works. Wouldn't any sane person want as many people voluntarily looking for bugs as possible to make sure the software was bulletproof? Hiding the code only makes us all wonder what is really going on, and doesn't make the system any more secure.

An overall perspecive on e-voting:
Just Say No Just Hasn’t Worked — Trying to Control the Uncontrollable

http://www.opednews.com/articles/opedne_bret_ste_060106_why__93just_say_no_92_ju.htm

My apologies - the correct link is:
Reader Contribution: Rigged Elections Not Taken Seriously Enough
http://www.buzzflash.com/contributors/06/01/con06011.html

For the record, I am a Texas Republican. I have been a precinct election judge for twelve years using Op-scan paper ballots. Last year I was one of many people involved in a local recount of an election that ended up in district court.

The judge noted in his decision that while there were still questions open regarding the initial election night reporting, no one had questioned the recount, or the security of the ballots until the recount.

I have been thankful ever since that the the recount I helped in had physical ballots to be counted by bipartisan groups. I have, along with friends from other parties, spent a lot of time researching electronic voting systems and ballot security. None of us has any faith in the systems being offered today.

Thank you Wisconson, and thank you for the correction on the law as passed. It sounds very close to the North Carolina model, not quite open source, but inspection upon request. I personally view open source or any variation on that theme as a tool to head off electronic voting. We have yet to find an electronic vendor without security problems.

I favor the Op-scan ballot in tandem with an open, non-corrupt election administration. After reviewing several thousand ballots, we a had small handful which the bipartisan counting tables could not agree on the intent, nearly all of those due to sheer voter incompetence. In my opinion, electronic systems are several iterations away from reliability.

I am far from a luddite technophobe. A Libertarian partner and I operate a non-partisan local issues Blog. This comment demonstrates my belief in computer tech, but it is that very familiarity that makes me doubt the superiority of electrons to, for instance, purple-dyed fingers and paper ballots.

You and North Carolina, and most recently Connecticut, where the Sec. of State suspended further purchases of DREs pending investigation of security vulnarability, all wil be cited in our on-going campaign to bar any system with less than total producibiliy

sir i need source code of online votting.plz send to me by my mail address

Quote from poster "Richard Welker":
[quote]
This is significant. The movement to have honest voting in the coutry is getting larger and larger. There is substantial evidence of fraud. By the Republicans. Against the intent of the constitution. Pretty telling.
[/quote]

Richard, you have it pretty much backwards. In the last presidential elections, there were indeed quite a few discrepancies and indications of voter fraud - but it appeared to be done by supporters of Democratic candidates.

The specific known instances were:
* Voting multiple times (some hundreds of Marquette students admitted to voting more than once for John Kerry, at different polling locations...one even appeared on camera on the news admitting this)
* Republican affiliates who ran "get the voters to the polls" operations found the tires on their vans slashed. The culprits were found to be the son of a Democratic official and his friends.

Don't try telling us that the Republicans committed election fraud, when all signs point to that in fact being done by the opposition. The facts show that Bush won by an even larger margin than the first time, despite the double-voting and vote-prevention tactics of Democrats.

So back up your allegations, or don't post them - your untruths cannot be substantiated.

The voting systems should be open just like the current system. We can all sit around and discuss how the various physical ballot machines count ballots. I can go to my local election board and ask them to produce a ballot machine, I can unscrew the top and peer into the machinery. I can take pictures of the physical ballot machine and share them freely with anyone I meet. We already have "publically inspectable voting machines".

One we move to electronic voting the process should be no different. I should be able to ask my local election board for a copy of the source code used in voting. I should be allowed to use the legal system to force them to produce a log file that contains a standard set of log messages which I could use to verify that all votes were counted.

Anything less than complete transparency in out voting systems will lead to a corruption of the voting system.

Just ask Venezuela, in the last presidential "election" a large majority of the opposition boycotted the election because the electronic voting system wasn't to be trusted. Start hiding election machinery behind proprietary electronic voting, and we'll lose the Democracy we fougth so hard to establish.

Using the voting machine one can vote two times with different names.is it possible.is it a drawback of the voting machine.

please send me the code of making g-mail in google i need it please...........send it to me by mail at falconionstar@yahoo.com

Electronic voting machine corruption


Election 2006 - who/what organization is working to fix the problem in other States?

Sharon and Rob

Sir, I need source code of online votting. Please send to me by my mail address.

Sir,

I need a source code for interface with the machine. Please do need full to me.

Hmmm. Can I have a copy of this voting machine. I mean the source code of this project, please.

I looked into this subject with the new developments I've been watching develop since in fact 04' and with the "Money Bomb" on the 5th I decided to take a look-see and find out what the state of fraudulent voting was in WI in the hope it turned around so maybe there was hope for Ron Paul. It's good to hear but I think an independent team should investigate the source code, they could use it to make sure there was no recount AND the source code wasn't needed. But also, my comment is What about the big picture? What can we do to help the movement in other states as well? There's no point if other states come to be the cause of another rigged election. A move to reopen this year old day or two discussion of EXTREME fundamental importance!!

Thank you!

Sir, I need source code of online voting. Please send to me by my mail address.

can anyone plz msg me the source code in c for an electronic voting machine.

Obama smokes cigarettes. If he doesn't care about his own health, how much will he care about ours?

The mission of the your site.

For Real,
He Doesn't Even Care About his own health do you think he's going to care about ours?

sir,
i need source code for my mini project

i want a aassembly language program for electronic voting machine

Please sir,send me an electronic voting system source code in C and JAVA to my email.I need it for my school project.