Open Source community needs to take over electronic voting software

[zulchzulu]

http://www.opensource.org/

There are many in the Open Source community (including myself) who I think feel a need to spearhead an effort to quickly make the software used for BBV machines and other electronic voting software controlled by an independent Open Source coalition of developers to make it completely neutral and non-partisan. We need to get rid of anything that uses Microsoft software and Diebold, ES&S and other electronic voting interfaces.

Using Linux interfaces (which use a small footprint) and secure Java DB that is secure, robust and capable of being managed securely is one idea. Old computers and PDAs could run Linux. It would not be utterly expensive if we look at the options.

This could be the next big internet happening....saving democracy. So many programmers that might be outsourced now could help pitch in and test the software and make sure it is 21st Century secure.

Any ideas are welcome.

[mhr]

1. Already Underway

See here:

http://open-vote.org/

http://www.openvotingconsortium.org/

Software Demonstration article here:

http://www.verifiedvoting.org/article.php?id=1766

I am sure there are others.

[zulchzulu]

4. Thanks...I knew there was some stuff happening...but it's way in the bg

We need to have this at the forefront and get a solution quickly.

[thinksmart]

7. This could be great

[intheozone]

2. We need the TruVote system

developed and put in place nationwide. That is the system invented by Athan Gibbs, who I understand died earlier this year in a mysterious auto accident. The system sounds great, is voter verifiable over internet and, from what I read, is easy to use and inexpensive to produce. If our democracy is going to survive, we have got to get this or a similar system in place.

[lapfog_1]

3. Why complicate things so much?

Any software can be used if the BBV machine doesn't COUNT the votes,
but instead prepares a paper ballot. No hanging chads or mispunches
or circles or checks not in the proper places.

Then tabulate the votes either by hand or by machine, but if by
machine, use two machines by different manufacturers.
After the vote for each polling place is completed and counted,
post the results on the web and send to the Secretary of State
to be certified. If there is a hand count, or a recount, reps
from all parties should be invited to witness the count.

[zulchzulu]

5. Yes...that's the point...open source voting software that is verifiable

It also has to be backed up on several open source servers to compare. And yes, a paper trial.

[St. Jarvitude]

10. Open source is the ultimate electronic transparent voting option

The entire voting process is auditable by the American people, not just the votes themselves.

If every step of the way is verified by a discerning public (via 100% open source), there is no need for a paper trail.

Free software is the only way to ensure a free people.

[lapfog_1]

17. I'm sorry - I fail to understand the logic of this stmt

Are you saying that BECAUSE it is public domain software on the
BBV machines, it can't possibly be hacked? And if hacked, it can
be discovered without a paper trail? I think you are simply wrong.

What I advocated in my post was a computer assisted ballot creation.
You can use anyones software, proprietary or public (I personally think
it should be on the web, downloadable to your PC). One uses this
software and a computer to navigate the choices one would like to vote
for (ranked voting anyone?), etc. After the selections are made,
whether at home or at the polling place, a PAPER ballot containing
your specific choices is created. NO recording or counting is
required at this step. No encoding is done either... the selections
are in plain text, possibly in two languages (English and your native
language). You the voter reviews the ballot for correctness and turns
it in.

At this point one COULD use machines to optically scan and count
the votes... and if one does, I would recommend that it be done
at the polling location, and done twice by two different machines
using software derived from two completely different sources. The
results are then compared (they should be identical) and the result
posted on a web page for everyone to view (as well as sent to the
proper authorities for inclusion in regional, state and national
totals). However, I would be just as happy to have the vote count
take a bit longer and have multiple sets of HUMANS count the vote
at each polling location (people from all interested parties plus
international observers). After counting by whatever method, the
ballots are collected, DUPLICATED (using standard XEROX technology)
and both copies locked up and stored for some number of months or
years. Sort of like checks at banks.

Only human readable paper (or other media) trail is acceptable for
something like this. If it requires a machine to count it, it
can be corrupted. I don't care whose software you use.

There are plenty of reasons to use public domain software for many
applications. This just isn't one of them. We are only talking
about 100 to 120 million or so ballots. Maybe only a few thousand
per polling place. This doesn't (and never did) require computers,
though I see where computers can be used to PREPARE the ballots and
assist people in making their selections.

[St. Jarvitude]

18. It's not only a matter of publically displaying the code.

One of the biggest misconceptions about free software is that it is public domain. Public domain implies merely public publication of the source code, whereas free software implies actual involvement by the end user in the writing of code or patches.

If the software is truly free, then it is developed by a community of volunteer programmers. Just posting the code up on a website does nothing - if there is an obvious security hole, it is indeed counterproductive because it is akin to giving a car thief the keys to your Aston Martin.

However, when the code is both open to scrutiny by the people and open to changes by the people, it works in the opposite direction. Need an example? Look no further than http://apache.org/httpd">Apache's httpd - far and away the most popular web server software on the internet (this one, anyway :)) and the very httpd DU trusts for its website. When was the last time you heard about a "Code Red"-esque worm wreaking havoc on Apache servers? You probably haven't, and that's because security flaws in the code are either spotted before they're committed to the stable branch or are fixed immediately + submitted to the Apache developers by an independent programmer.

I hope you didn't come away reading my other post believing that I think making the software free (as in speech, not as in beer) will solve all of our election problems. Obviously, if we're going to go the electronic vote tabulation route (which is, sadly, the way we appear to be headed), other steps have to be taken: hardware and software firewalls on all tabulation machines, dual and separate vote tabulation (a great idea, by the way), and intense scrutiny of the code by people who are not associated with scum like Diebold. All I am saying is that using free software for vote tabulation is a step in the right direction.

[rman]

19. corporate voting systems are *business secret*

we *are not allowed* to audit and verify Diebold's voting systems.

*open source - by definition - is open to public scrutiny*

open source is public property - a nightmare for corporatism

clear?

[Syrinx]

6. open source, peer reviewed software would be the way to go

But I like good old "put your X here" paper ballots. So, it might take a few days to count the vote, but that's okay with me.

[GetTheRightVote]

9. I am a programmer and I prefer paper too

On paper you can not change my vote, on a computer there is always the chance of it when bad men & women live in my neighbor.

[Turn CO Blue]

8. You should read this thread at BBV, it mentions Linux

SourcePlus, Diebold, ChoicePoint, the usual suspects, plus the people posting on the thread are researching the connections of lots of companies and shady individuals. There is also a link in the story that mentions a whistleblower at a county suing Diebold for using illegal software.

Anyway, it is creepy and sinister reading (but you need to get past the first 20 posts before it gets really good.)

<http://www.blackboxvoting.org/cgi-bin/dcforum/dcboard.cgi?az=show_thread&om=56&forum=DCForumID4122&viewmode=all>

I would start a new topic about this thread, because it is INTENSE, and I was spellbound reading it half the night (I even sent it to Olbermann) -- but I am not able to start a thread yet.

If these companies/people are all tied together -- IT WILL BE THE SMOKING GUN -- it would be bigger than the theory of everything.

[zulchzulu]

11. What a murky story to try to explain to the general public

We need to have Wired or somebody do a story on this. It does have Frontline written all over it. The word needs to get out immediately. All this has international implcations as well with voting electronically.

That thread is spooky as hell on BBV.

[Turn CO Blue]

15. Yeah, I think it's creepy and the thread keeps growing!

I always thought that cabals were something in sci-fi novels, huh?

Anyway, I did send the link to Keith Olbermann, but it will take his research staff days to make any sense of it at all (if they even read it or all those links).

To your point, the implications about the code should be studied by computer-gurus like yourself and Wired, etc. but the story itself needs somebody major to break it (with all due respect to Bev, BBV and the good-bloggers.)

Can you start a new thread on this topic with the link (since I can't yet)??

[AntiCoup2K4]

12. I'll agree that Linux systems are more secure, but that's not the point.

Without a tangible, printed record of the votes themselves, any database is still subject to manipulation - even with a reduction of access to it (pardon the pun)

[zulchzulu]

13. An obvious part of this is a paper trail

Having the results uploaded to various redundant servers would also allow for being able to check discrepencies in ballots.

Yep, it's complicated...but there would be a paper receipt as well...like any ATM.

[AntiCoup2K4]

14. And that's one of my big problems with Diebold....

They make ATM's as well, so there is no way that they can claim any technological hardship in printing instant results from a vote.

Could you imagine if people couldn't get instant results from an ATM as to whether a cash transaction went as planned? It's too bad more of them don't look at voter fraud in that exact way, because when the Chimp crashes the economy AGAIN, they'll be losing money too.

[sandnsea]

16. mail in, paper ballots

The ballots are the ballots, they are used in the recount. Optiscan with open source for any tabulating programs. Registrations with carbon copy receipts, signed by the person doing the registering. All organizations doing registration drivers keep names and signatures on file. Lather, rinse, repeat. I'm going to say this over and over and over....