Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
Why Electronic Voting Software should be Free Software.

By jeep in Internet
Mon Jul 09, 2001 at 11:43:06 AM EST
Tags: Freedom (all tags)
Freedom

In this article I explore why there are strong technical, legal and political reasons for Free Software technologies being the only ones considered for use in legally binding votes conducted with Internet or other electronic technologies.


The electronic voting market is exploding... numerous existing and start-up companies have identified the huge revenue potential that the private and public markets offer, resulting in a raft of products and services being offered.

There has been considerable discussion both on- and off-line regarding the merits of electronic voting as whole, in addition to controversy over the validity of different technologies. However there seems to be little public debate over whether proprietary software is the appropriate way to provide electronic voting in public elections nor whether it's use makes the best business sense in private implementations.

Before the electronic voting community and its onlookers make hurried assumptions over how this market should develop I am keen to put forward the arguments for a non-proprietary model: Free Software.

For those unfamiliar with the movement as a whole I highly recommend browsing the Free Software Foundation's site on the matter at http://www.gnu.org/philosophy/ . Within this article I shall be covering the merits of Free Software (sometimes known as Open Source - though they are not quite the same thing) only within the context of electronic voting.

Openness

A key benefit of releasing software under a Free Software license is openness. The (Internet-based) electronic voting community is only starting to take steps towards becoming more accessible and open through initiatives such as publication of The Bell. Being open is key to fostering trust and accountability - these are especially needed in the world of voting software.

The IVTA's recent commitment to open protocols is another positive step in the right direction, however (as I pointed out on the IVTA tech mailing list) it is no guarantee that the software that uses these protocols will be open or even that the software implements the protocols properly. But why should we want the software to be open at all? Two key reasons are security and observability.


Security

Many commercial electronic voting companies seem to rely on security through obscurity. They will not release detailed (or in some cases, any) technical information on how their voting systems work. From analysis that I have done, some of the guilty parties have good reason to hide their handiwork as their electronic voting systems are nothing more than trumped-up e-commerce systems that do not address any of the major security, privacy or reliability issues that we as a community are working to solve.

We all know that the only way to guarantee security is through peer review and careful audit by professionals. The IVTA encourages this with open protocols, and I support that whole-heartedly. However there are programs we could all nominate for not properly implementing freely available, publicly ratified, standards. Only Free Software, with the access to source code that it unequivocally upholds, enables anyone to verify (and repair) implementations of protocols. In other words "many eyes make bugs shallow".

Access to the source code also allows for the easy addition of new protocols whether for secure vote recording or voter authorisation (for example GNU.FREE's security model allows for the easy addition of new identification devices such as smart cards or retina scanners).

The FREE e-democracy project strongly feels that the slightest hint that privacy or security could be sullied by electronic voting systems may permanently damage the likelihood of such technologies being widely accepted in the public sector. Only full and permanent to commitment to a culture of openness will effectively counter such threats.


Observability

Free Software creates openness by allowing anyone to use the software, read the source code, modify the program and pass it on. People get involved and are naturally encouraged to learn how the programs work. This fosters a culture of observability where programmers expect their Free Software code to be read by a multitude of users with skills exceeding or way below their own. The result: programmers are careful to comment their source code and constantly evaluate the quality of their programming.

Furthermore anyone with the skills (or the time and willingness) can check the code themselves - helping them to trust the system. Why would you trust a similar proprietary system when the creators won't tell you how it works? What could be hiding in there? Do they know about problems or weaknesses that I don't? The very act of hiding the information creates distrust - especially when the provider is building the system for personal profit and thus not necessarily with our best interests in mind.

Anyone could audit a Free Software voting system before it went into use to guarantee the absence of trojan horses, hidden result manipulation functions or blatant weaknesses without having to sign Non-Disclosure Agreements and with the full ability to take their findings public and/or fix them if they were that way inclined.

Freedom

As Richard Stallman (founder of the Free Software Foundation) says "think of 'free speech', not 'free beer' ". Free Software licenses, especially the GNU General Public License, enforce basic freedoms and rights for users of the software. While there can be no denying the strong arguments for associating these freedoms with all software, I believe they are particularly apt in the context of electronic voting software. In my opinion the two key freedoms worth discussing with regards to this subject are freedom from dependencies and freedom from cost.


Freedom from Dependencies

Any county, state, country or organisation buying commercial electronic voting technologies is totally dependent on the strategy of the manufacturer. The company may develop the software in a direction different to one's own elections strategy or may simply refuse to provide the features you believe to be essential.

Take, for example, Iceland's struggle to get Microsoft to deliver Windows9x in their native Icelandic language. Alternatively consider the predicament users of Banyan Vines networking software were placed in when Banyan totally abandoned the networking market for e-commerce under the new name of epresence. Iceland risks being sidelined in the 'new economy' while Banyan users saw their massive technology investments become worthless. If the software had been Free Software they wouldn't have been in such dire situations.

No matter the direction, fortunes or internationalisation policy of your provider - if the software is Free Software it is totally modifiable. The user has the source code and so can develop the software however they want, thus investments in and commitments to technologies are protected. <u>Free Software users keep control of their technological destinies.</u>


Freedom from Cost

Thanks to the advertising of software associations and the scary licenses we get with software we have lost the culture of sharing that once used to be a big part of the computing world. Even academics, who are dependent on the discourse and sharing of ideas that used to typify academia, now prefer to patent before sharing - if they ever do share.

Free Software lets you share, the way friends and neighbours should, without any legal repercussions. This isn't just about sharing with our neighbours in the next cubicle or building, this is about fellowship with our global neighbours. I believe we have a moral duty to empower any country to follow the (often twisted) path towards representative democracy - electronic voting may well be the best way for many countries to do so. But will commercial voting companies share their systems with less well off organisations and countries? Aren't they dependent on the income from software and associated services sales for survival? Are we going to mindlessly follow the same worn path that the pharmaceutical giants, disdainful of limited third-world buying power, have trod?

We can share Free Software with whoever we want, wherever we want. And you can feel good knowing that license allows them to keep sharing that with whoever they want. It's a powerful thought.

One final note on the freedom from cost: Let me make clear that supporting Free Software does not mean opposition to commercial software and its developers. Software development is a good and decent way to make a living and despite what some argue, proponents of Free Software are not 'communists' against all forms of commercial activity. Furthermore paying for Free Software is perfectly acceptable, but Free Software provides a viable (and my preferred) alternative to pure commercial licenses in the same way that shareware and public domain licenses do. Free Software provides a key building block for creating a future that celebrates moral values as well as commercial success.

The FREE e-democracy project

So how does our project follow through on some of the promises that Free Software offers? The project has a number of aims that we keep at the top of our minds when developing the software and when running the project as a whole:

Software Development Aims

  • Provide a secure and private system
  • Create scalable and reliable software
  • Offer a non-commercial, non-partisan voting alternative
  • Use the GPL to create an open system that Internet users will trust
  • Release a system that can be used to support the growth of effective democracy anywhere in the world

Project Aims

  • Develop a leading electronic voting system
  • Advocate the free software paradigm
  • Evangelise the use of technology to strengthen democracy within a holistic understanding of the current malaise i.e. Internet voting alone isn't going to solve turnout problems

With these aims in mind we have developed software which has been through several iterations, currently standing at version 1.3. The software has been downloaded hundreds of times and a wide number of groups ranging from VICA to the Free Software Foundation are assessing it.

I invite anyone to contribute their time to a project which could have a long lasting impact on the world of electronic voting. Welcome to the community!

About the Author

Jason Kitcat is founder and co-ordinator or the FREE e-democracy project who, with the support of the Free Software Foundation, develop the GNU.FREE Internet Voting system which is released under the GNU GPL. He is co-founder and Head of Production at Swing Digital a digital consultancy.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Related Links
o http://www .gnu.org/philosophy/
o FREE e-democracy project
o Swing Digital
o Also by jeep


Display: Sort:
Why Electronic Voting Software should be Free Software. | 41 comments (37 topical, 4 editorial, 0 hidden)
Electronic Voting is a Bad Idea (4.25 / 4) (#3)
by Simon Kinahan on Mon Jul 09, 2001 at 09:02:08 AM EST

As was so graphically demonstrated in Florida last year, anything that makes the voting process harder, or the count less clear, decreases the extent to which the results can be said to represent the express will of the people. If we have trouble working out whether a hole has really been punched in a piece of paper, how are the possiblities of bugs, hackers, and activist software developers going to play out in the media ?

Personally, I favour a piece of paper, and a pencil. Once cross in one box per election, or its a no vote.

Simon

If you disagree, post, don't moderate
Vote-Counting Machines (4.33 / 3) (#5)
by SEWilco on Mon Jul 09, 2001 at 09:24:22 AM EST

I consider any vote-counting machine to be part of "electronic voting". That includes the popular optical recognition machines, where the voters put black marks on a page and the page is then run through the machine for immediate counting. I think the machines should be as open as possible, especially a machine which electronically reports to a central facility.

[ Parent ]
Quite (2.00 / 1) (#15)
by Simon Kinahan on Mon Jul 09, 2001 at 11:44:45 AM EST

And I think any vote counting machine is a bad idea for the reasons outlined above. Paper, pencils and manual counts are safer and more reliable.

Simon

If you disagree, post, don't moderate
[ Parent ]
Bullshit. (4.00 / 1) (#25)
by physicsgod on Mon Jul 09, 2001 at 05:58:01 PM EST

Just look at what happened during the manual recounts, people would amazingly find a hole right where their canditate was. Machines don't have any politics, so they report who actually won, as opposed to who the majority of vote counters think should have won. Just because Florida used a crufty system doesn't mean the entire concept is bad.

--- "Those not wearing body armor are hereby advised to keep their arguments on-topic" Schlock Mercenary
[ Parent ]
Ahem (none / 0) (#28)
by Simon Kinahan on Tue Jul 10, 2001 at 05:18:35 AM EST

Just look at what happened during the manual recounts, people would amazingly find a hole right where their canditate was.

Well, thats a remarkably serious accusation. Do you actually have any evidence to back it up ? There were certainly many accusations of fraud in Florida, mostly by Republicans trying to disrupt recounts, but I don't remember even one being subtantiated.

Punching a hole in a piece of paper is not a process with a binary result. There is therefore room for ambiguity. Because of the glorious and perfect reliability of machines (All Bow To The Machine !) noone bothered to set clear criteria for judging whether something was or was not a vote.

Just because Florida used a crufty system doesn't mean the entire concept is bad.

No, but it can be considered symptomatic. Complex, vague systems, as a general rule, have more failure modes than simple, clear-cut ones.

Simon

If you disagree, post, don't moderate
[ Parent ]

references. (5.00 / 1) (#40)
by physicsgod on Tue Jul 10, 2001 at 06:43:59 PM EST

I don't have any specific sites to point you to, but during the debacle I distinctly remember counting panels where two members would say "that's a vote for #2(or whatever the number was)" and the third member consistantly disagreeing.

Punching a hole in a piece of paper is much more binary than writing someone's name down (is that Monroe or Munroe?), putting a black mark on a piece of paper is better yet, and perhaps the best system would be a touch-screen saying "please touch the face of who you wish to vote for president. You have selected Linus Torvalds, is this correct?". You then have plenty of options of what to do with the data (which at this point is purely digital, and accuratly depicts the will of the voter). My own preference would be to send the data to a central server via a "private" network, plus backups on a local server, removable media, the machine's hard drive, and a machine generated punchcard (no pregnant chads), but I'm a real belt-and-suspender type.

The problem with hand counted ballots is that hand counting takes a long time, and is subject to personal biases. I know that every time I've been counting ballots for various unimportant elections and things weren't going my way I've been tempted to hide "wrong" ballots. And since the counts take time there are fewer of them, which means fewer chances to catch dishonesty.

No, but it can be considered symptomatic. Complex, vague systems, as a general rule, have more failure modes than simple, clear-cut ones.
This of course depends on how you define simple. Hand ballots, with the distributing, handing out, collecting, and counting are much more complicated in actual implementation than electronic voting, where you just need to set up the machines and distribute unvalidated smartcards, then collect the cards, check validation and sum registers. Design is much more difficult for the latter than the former, but as long as you thoroughly test the system beforehand you won't have problems.

--- "Those not wearing body armor are hereby advised to keep their arguments on-topic" Schlock Mercenary
[ Parent ]
Paper... not really that safe (5.00 / 1) (#10)
by jeep on Mon Jul 09, 2001 at 10:46:56 AM EST

Everyone assumes that the good old days of paper ballots were the ideal way of doing things but in fact they are susceptible to fraud, spoiling and so on. Also paper-based ballots are extremely costly.

That said paper-based ballots will always have a place in our world due to their simplicity and low barriers to entry. However there are a large number of corporate interests pushing hard for Internet Voting. That is why GNU.FREE was first created, to provide a Free Software alternative to the closed and proprietary offerings.


--
The FREE e-democracy Project
Promoting Free Software in Government

[ Parent ]
Safety (4.00 / 2) (#14)
by Simon Kinahan on Mon Jul 09, 2001 at 11:43:52 AM EST

Everyone assumes that the good old days of paper ballots were the ideal way of doing things but in fact they are susceptible to fraud, spoiling and so on.

But its safer. The same costs that apply to the people running the elections apply to fraudsters. Anyone wanting to rig an election has to transport vast quantities of paper, and insert it into a closely monitored manual count.

Similarly, because the counting is manual there is less ambiguity in whether or not a vote is valid, and in whether or not it has been tampered with. Its that ambiguity that ruined the legitimacy of the Florida count, even with fairly simple mechanical machines. Opting for a more complex system does not look like a good fix.

However there are a large number of corporate interests pushing hard for Internet Voting.

Just because "corporate interests" want to grab another space at the cash-trough by painting something as "inevitable" does not mean we need to accept their conclusions. Voting systems need to be safe, and reliable above all else, and the complexity of large software systems inevitably degrades those properties.

Simon

If you disagree, post, don't moderate
[ Parent ]

Not entirely bad (5.00 / 1) (#16)
by WinPimp2K on Mon Jul 09, 2001 at 12:12:08 PM EST

Personally, I favor the use of a computer that will then print a physical ballot that can be machine counted or manually counted with equal ease. For security reasons I do think that every vote must have an actual physical representation which is counted. After all, how do you manage a recount if there is notthing to actually count?

An electronic representation of the ballot would be a good aid to the voter in making their selection (variable font size, only one race onscreen at a time etc). And of course by only printing completed ballots the software could also do little things like printing the name of the selected candidate in boldface to make it painfully clear who the voter selected.

Of course, I also favor the instant runoff / ranked voting system used in Australia - with something like that, the names of the candidates could be progressively indented to indicate decreasing preference.

[ Parent ]

Ok, two thngs (4.33 / 3) (#6)
by psctsh on Mon Jul 09, 2001 at 09:29:12 AM EST

As you yourself said: "The electronic voting market is exploding... numerous existing and start-up companies have identified the huge revenue potential that the private and public markets offer, resulting in a raft of products and services being offered."

So what exactly is going to give these companies an incentive to create open-source software? While anyone can argue all day that it's possible to make money on free software, the reality is that most companies won't agree with you. Not only that, but they can go for heavier advertising and get better deals [than you], because they can expect to eventually make back the money they lost.

The other problem I have with this is a fundamental flaw in the open-source system (at least, in this particular case).

You claim that open-source software is more secure because everyone can read the source code and find/fix the security flaws in it. However, while this is truly representative of individuals' computers, where they're running the software themselves, this is *not* the case involving applications dealing with voting. As it is, since the organization that runs the voting system is the only party running the voting software; there's really no stopping them from writing the two extra lines of code required to log votes. They could also heavily modify the source, exposing more security flaws that will never be picked out by the public.

Because the final version of the software that gets executed is essentially up to the company running it, companies (in this case) get every benefit of both open source and closed source. They get already prepared software, and they get to keep the actual source hidden. Because not only are there ways around the GPL (dll's, spawning executables, etc), there's nothing keeping companies from distributing one version to the public for scrutiny, and using another in the final product.

I would expand upon the fact that nothing is actually requiring companies to update their software with the new changes/fixes given by FREE's team, but I really have to get to work.

You're right... (5.00 / 2) (#12)
by jeep on Mon Jul 09, 2001 at 11:01:49 AM EST

Potentially once the code is executable then it's closed code which could be abused etc etc. However we propose several ways to prevent abuse.

Firstly we would make sure that voter authentication (Electoral Roll servers) is run by an organisation independent from the body running the vote registration servers.

We would also ask that all code be audited by a trusted third party who could sign and message digest the final code and it's compiled form. Tests could be done before, during and after the ballot to ensure that the correct code was run.

Finally it isn't true that the code is restricted to the people running the ballot. The voters have code on their systems to perform the vote, so it is distributed and the possiblity of a cross-authentication net is being considered for future versions.



--
The FREE e-democracy Project
Promoting Free Software in Government

[ Parent ]
Client software... (4.00 / 1) (#20)
by MfA on Mon Jul 09, 2001 at 02:58:54 PM EST

I use my computer for far too many things (running windows most of the time...) to consider it secure, we all know that the vast majority of machines out there in people's home cant be considered remotely secure. The code being solid doesnt do diddly squat if the system isnt, and you cannot determin that remotely (please dont bring up blessed binaries, the key is in there... its just security by obscurity).

The question is, should you allow people to run your client software on potentially insecure systems? I dont think so... your vote is your own, but its not yours to willingly put at risk of being stolen.

A system usable for government elections would have to be a dedicated hardware device programmed in a secure environment with its own interface on which you express your vote IMO (something like the calculators some banks use to verify transactions for instance). With a little public key encryption thrown in there you can build something Id trust, but I know better than trust the security of my own box.

The only other alternative I see would be to give everyone a dead tree list of personal one time codes (one for each possible vote) they can enter to vote remotely ;)

[ Parent ]
Oops (none / 0) (#21)
by MfA on Mon Jul 09, 2001 at 03:11:50 PM EST

Actually the paper list would need to have both codes for the votes and replies they can expect if the vote went through sucessfully, being able to influence the results by surpressing votes for given parties is of course as bad as being able to hijack somoene's vote.

[ Parent ]
Client Security (none / 0) (#29)
by jeep on Tue Jul 10, 2001 at 06:38:01 AM EST

GNU.FREE's client software could be run on a home computer (or any device that supports Java 1.1 with AWT) but the servers should never be just run off a box at home.

At lot of the issues being raised are social/human engineering points which need to be resolved with the right kind of guidance and regulation as opposed to technology.

However there are huge problems with the security of client machines which aren't specifically for voting. Note that security is a process, not a state, and as such we will never have complete security - we need to manage risks to appropriate levels. Potential solutions include certified boot-disks, sandboxes and specific vote operating systems.

As for dead trees. Indeed the current authentication system relies on users being given a polling card with a one-time code. Potentially smartcards or biometrics could be used but there are huge privacy, cost and implementation issues implied.


--
The FREE e-democracy Project
Promoting Free Software in Government

[ Parent ]
Im not worried about authentication (5.00 / 1) (#35)
by MfA on Tue Jul 10, 2001 at 02:37:11 PM EST

Wether security is a process or a state is entirely besides the issue as far as Im concerned, a non dedicated system too insecure to even consider using in the first place. I considered a boot disk, its the only remote possibility. But even the bios cant be assumed to be secure, so its not much of one IMO. If fraud can be automated you have to be paranoid. We have seen what a couple of thousand votes can do, if remote voting is to be a success I shouldnt already today be able to conceive of a scheme with which I could easily have corrupted a shitload of votes if theres better alternatives (make patches for a series of the most popular bios's, use a combination of trojans and remote exploits ala outlook-express buffer overrun problems to compromise people's computers et voila... going to a boot disk is futile). Politicians wont be doing it, but some hacker will if given such an "easy" avenue to glory.

It would be a disaster waiting to happen, I infinitely prefer voting booths if remote voting has to be done without dedicated remote systems with their own de/encryption and interface. I wasnt even considering the problem of identification, thats a problem we already have today... Im more concerned with the new problems you are introducing.

[ Parent ]
Ok, I feel kinda underspoken (none / 0) (#26)
by psctsh on Mon Jul 09, 2001 at 06:13:17 PM EST

'cause my main point was going to be that nothing is forcing the pollsters (or whatever they're called) to update to the latest [bugfixed] code release (that's why I saved this point for last).

However, due to the fact that I *really* had to get to work, I didn't get to squeeze that one in.

When you say the voters have the code on their system, do you mean that as in "I'll be running it on my box at home" or "I'll be using it at the voting booth?" 'Cause I assumed that nothing would be given over to private use. I'm of the opinion [in this case] that the source should be open while the system remains closed. Do you have some kind of a check in place to keep malicious versions from fucking with the system?

[ Parent ]
You've convinced me... now convince my mom... (4.75 / 8) (#7)
by Anatta on Mon Jul 09, 2001 at 09:45:24 AM EST

It seems to me that everything you've said is logical, sensible, and overall is a good project plan for creating a fair, reliable system.

Now convince my mom the same thing. She knows how to use AOL, Windows, and IE, but she'd be hard-pressed to make a subdirectory and move files into it. She's certainly heard of Open Source, but she probably has no idea what it is. I'm sure she has no love for Microsoft, but when push comes to shove, my guess is she'd trust the Big Business types over the hackers whom the mainstream media has largely denigrated as a bunch of anarchists. Even though logically, the open system is much more secure, my guess is she'd still trust the Big Boys over the FSF people. As for Stallman, I would say he's unquestionably brilliant, but again, my guess is my mom would trust evil Bill Gates (irrationally or not) over RMS.

It seems to me that in order for this system to really take off, you've got to "sell" it to people that have *no idea* what you're talking about. Call them the sheep, the uninformed masses... I call them mom... but you have to convince at least some of them to call their congressmen and say "I want this system, because it produces fair and just elections".

I'm not exactly sure how you'd do that, but one way that comes to mind is to get IBM, Compaq, and other companies that like open source products on board, even simply in name only. My mom certainly knows who IBM and Compaq are, and my guess is she'd trust them with technology issues more so than RMS and the FSF alone.

Clearly you've hit on an important issue here; Microsoft (for example) would certainly have an "interest" in keeping the source code of such a system closed... but Microsoft would argue that the FSF folks are a bunch of anarchist hackers that spend their time looking through mom's credit card purchases online, and can't be trusted.

A voting system is going to transcend a number of lines that not even computers have transcended. I seem to recall that something like 55% of households in america have computers (and I would guess it's lower in most other countries)... that's 45% of households in america that for whatever reason don't even have computers, let alone know the real (non-biased) differences between FS and closed source.

You've responded to a large number of the technical issues of how to go about creating a fair, just electronic voting system. Now you have to jump through the PR hoops, which it seems to me are flaming, moving, and have nasty spikes below them... a much bigger challenge than the technology hoops.

I'll continue to do my best to educate the moms of this world on these issues, but in order for this to really work, it seems to me some MAJOR muscle needs to be brought in...

Best of luck.
My Music

It's our biggest challenge: Firing up the people (2.00 / 1) (#13)
by jeep on Mon Jul 09, 2001 at 11:09:11 AM EST

I agree totally. With the K5 community's grace this will be the first of a series of articles exploring these issues.

We have got a company to commercialise GNU.FREE, however we don't have many resources so we're just soldiering away trying to raise profile to make sure that decisions makers realise the consequences of not using Free Software.

Once the decision makers are on board then the challenge of getting our Mums' to use Internet Voting will be the next hill to climb.



--
The FREE e-democracy Project
Promoting Free Software in Government

[ Parent ]
Decision Makers != The People (none / 0) (#27)
by dzelenka on Mon Jul 09, 2001 at 06:28:02 PM EST

You agreed with the previous post, but you didn't understand his point. You plan to influence the decision makers first, then the people. The point of the previous post was that you need to have a groundswell of popular understanding and support BEFORE you can influence the decision makers. (Why DO we call them leaders if all they do is follow?)
"Are you talkin' to me?"
[ Parent ]
Possibly... (none / 0) (#33)
by jeep on Tue Jul 10, 2001 at 07:07:37 AM EST

Maybe you need public support first, but I'm not sure. This is a very technical and esoteric area and until the digerati and political classes are interested in, and discussing the issues, I think it's going to be near impossible to get the general public interested.

In fact I believe the public will need leadership in this area... otherwise getting public opinion focussed on Internet Voting will be like herding cats!



--
The FREE e-democracy Project
Promoting Free Software in Government

[ Parent ]
Lot's of voting software out there (4.00 / 1) (#8)
by khallow on Mon Jul 09, 2001 at 10:25:31 AM EST

There's a lot of open source projects on SourceForge dealing with voting. Usually, it is a little feature associated with the real project. Don't think that many of them could be extended to the degree that you want. However, I see one of the biggest problems being adoptibility.

FWIW, I believe that the open source movement may be somewhat ahead of commercial developers in the area of simple voting software and interfaces. Hmm, E-Democracy has an old site on SourceForge as well. Might be worth examining the voting systems that are at least in beta for ideas on interfaces.

Perhaps should be clearer... (5.00 / 1) (#11)
by jeep on Mon Jul 09, 2001 at 10:55:27 AM EST

Everything I discuss in the article is available today in our Internet Voting system, GNU.FREE. The sourceforge site isn't old, we still use it as a distribution point but the main info is on free-project.org.

I've researched the other systems on Sourceforge and elsewhere but they are all polling systems, not voting systems. We are, AFAIK, the only Free Software Internet Voting system. We're supported by FSF and FreeDevelopers.net



--
The FREE e-democracy Project
Promoting Free Software in Government

[ Parent ]
STV PR voting (4.00 / 1) (#17)
by winthrop on Mon Jul 09, 2001 at 02:08:40 PM EST

I was in charge of the voting for the last congress of the Mass. Green Party, and we would have died for a free (as in speech) electronic voting system. We use Single Transferrable Vote Proportional Representation, (aka the Cambridge system). For more information, see the Center For Voting and Democracy.

It is a much better system than plurality wins, but it is an absolute bitch to figure out who won. In one test run, with 6 candidates, 3 slots, and 15 votes, it took us 30 minutes to count the votes.

I tried going through your documentation, but I couldn't find the answers to these questions. Does your software support multiple-district races? Preference voting?

Does it determine the winner? Do you have different algorithms for determining winners? Can we submit algorithms for determining winners?Which is preferred: English, pseudo-code, or java?

Thanks!

The big one... (none / 0) (#34)
by jeep on Tue Jul 10, 2001 at 07:13:10 AM EST

I'm a big supporter of proportional representation. In fact I'm a member of the Electoral Reform Society, the oldest and most esteemed society to advocate PR and voting reforms.

As such I'm desperately keen to implement all the systems like STV in GNU.FREE however it's very hard to support all the different systems. We need to think about the different user-interface, protocol, counting and database issues arising from supporting many voting systems. So at the moment GNU.FREE merely supports simple counts of votes, not voting systems are explicitly supported. But it's at the top of our TODO list and any support is much appreciated, especially if its Java code ;-)



--
The FREE e-democracy Project
Promoting Free Software in Government

[ Parent ]
Cool (none / 0) (#38)
by winthrop on Tue Jul 10, 2001 at 05:33:08 PM EST

Thanks for all the work you've been doing.

I understand the difficulties and I'll be following your software.

You probably know this, but there's a gigantic market for voting systems out there for non-governmental elections (intra-party, cooperatives, clubs, etc.) and they'd probably be more willing to use as-yet-unproven software...

Anyway, good luck!

[ Parent ]

Collaborative writing (5.00 / 1) (#18)
by emmanuel.charpentier on Mon Jul 09, 2001 at 02:36:45 PM EST

Sorry for the rider, but it is exactly the kind of topic to which pertain VeniVidiVoti.

It is more that simply voting, it's a full political system implementing a form of direct democracy. The goal being to collaboratively write any sort of text, like a constitution, a petition, a novel or a newspaper.

It's built on a J2EE framework, using the great and free (GPL) JBoss. And it's actually about 80/90% features complete. What VVV lacks is a community to test drive it, anybody out there with server space???

I'll have to look at the FREE project, to see if there are any feature I can copy, but I'm afraid it doesn't have exactly the same orientation. For exemple it doesn't require the same stringent conditions on privacy, considering every participant can change his votes or delegations after casting them, the administrator or anyone with database access can see who did what... :-(

As far as voting software should be free I completely agree. In fact, I believe every software used or written by /my/ government should be GPLed

The GPL is a great hack on IP laws, the only way to save our ideas from corporations. In fact, I wonder if most corporations/multinationals could survive the destruction of the very (absurd) notion of Intellectual Property... :-) (great prospect)

VVV totally unlike GNU.FREE (none / 0) (#32)
by jeep on Tue Jul 10, 2001 at 07:02:09 AM EST

For exemple it doesn't require the same stringent conditions on privacy, considering every participant can change his votes or delegations after casting them, the administrator or anyone with database access can see who did what... :-(

I'm not sure if you're referring to GNU.FREE or VVV, but firstly GNU.FREE whole architecture is based around protecting voter privacy - at no time can you infer who votes what.

I belive I've previously discussed with someone at VVV the requirement for people to change their votes at a later date. The only way to achieve this feature is to link who has voted with what they have voted. Having such an explicit link is asking for privacy to be breached and we would not even consider such a design.

However VVV isn't a high-security application, it's a collaborative system, and comparing it with GNU.FREE is really a fruitless exercise ;-)



--
The FREE e-democracy Project
Promoting Free Software in Government

[ Parent ]
You are right (5.00 / 1) (#36)
by emmanuel.charpentier on Tue Jul 10, 2001 at 04:50:54 PM EST

you are totally right, VVV keeps count of who did what, and this is big trouble as far as privacy is concerned. Thus big responsability on the administrator, big responsability on the site security. Maybe steganography will partially be a solution.

I push VeniVidiVoti in this article because it is a political tool, alike FREE. They both are concerned with votes, and both are free software. But FREE is designed to securely take votes, while VVV is designed to manage propositions, votes *and* votes' delegations.

Sometimes in the future, I hope I will come up with a way to allow voters to change their mind, and to insure their privacy securely... sometimes... in my dreams.

[ Parent ]

Academics [OT] (4.00 / 1) (#19)
by infraoctarine on Mon Jul 09, 2001 at 02:37:21 PM EST

This has not got anything to do with electronic voting, but I'm a bit disturbed by the following sentence in your story: "Even academics, who are dependent on the discourse and sharing of ideas that used to typify academia, now prefer to patent before sharing - if they ever do share."

I don't think that is true. Most academics still believe in the open dissemination of research results, and most findings are still publicly published. You're right there are more patents from academia now than it used to be, but saying that academics in general prefer patenting to sharing at least goes contrary to my experiece. And "if they ever do share"? Yes, all the time! You have never read a scientific journal or conference proceedings, have you?

Better late than never I guess (4.00 / 1) (#22)
by MfA on Mon Jul 09, 2001 at 03:42:26 PM EST

Sure they want to share, patents are after all a way of sharing the knowledge anyway... its just a little late, usually they will share it a little sooner through a printed publication (which gives them a year to patent it). But at the present pace of innovation publication is a ridiculous slowdown, in physics Ive seen examples given of new research being done on the basis of pre-prints... and articles being written from that research before the original article even went to print.

In a patent sensitive area of research determining when to make something public is a carefull balancing act I imagine, a year is not a long time. If you describe it to anyone else and he puts it in a publication (doesnt have to be print even) that will also trigger the one year limitation...

[ Parent ]
On patents (5.00 / 1) (#24)
by infraoctarine on Mon Jul 09, 2001 at 04:20:39 PM EST

Most scientific work is not patented at all, it's in the public domain. In the academic world, you don't build your reputation by hoarding patents, but rather by making contributions (i.e. being the first to publish results/ideas). That is why I think the notion of academics not wanting to share their work is so absurd. Sharing is how you advance in academia.

I guess it's US patent law you are referring to. European law is different, if you publish or present something before you file your patent application (even the day before) you can't get a patent, the work is then in the public domain. The day after your application is filed, you can go public and still get the patent.

[ Parent ]

Academics sharing... (none / 0) (#31)
by jeep on Tue Jul 10, 2001 at 06:47:58 AM EST

In fact I'm UK based. My experience with Computer Science-based academics was what lead to that remark.

Yes! I've read academic journals. However in CS it is often the case that people don't share for fear of industrial espionage (due to the academic having a commercial spin-off company they want to protect), a misunderstanding of European Patents or just a fear that others in the department will try and claim some of the credit. Most academics do share, but an increasing number don't share as readily as they used to.

Perhaps my statement was too strong, but I cannot deny the trend I have experienced.



--
The FREE e-democracy Project
Promoting Free Software in Government

[ Parent ]
Am I the only one reminded... (4.50 / 2) (#23)
by Kasreyn on Mon Jul 09, 2001 at 04:01:01 PM EST

...of Asimov's short story, "Franchise"?

In it, a future world exists where there is a computer so insanely subtle and powerful that it can accurately determine one "average" american, who ALONE votes for the president. After judging his responses to many subtle psychological questions, the machine returns an answer of who the american people have voted for.

Personally, I think we're heading toward that future anyway. But without free software on that system, the will of the american people will have a strange habit of re-electing Comrade Napoleon 12 times in a row...


-Kasreyn


"Extenuating circumstance to be mentioned on Judgement Day:
We never asked to be born in the first place."

R.I.P. Kurt. You will be missed.
Absolutely! (none / 0) (#30)
by jeep on Tue Jul 10, 2001 at 06:42:50 AM EST

Yep, in my dissertation which lead to the creation of GNU.FREE I referred to Franchise as the future we should aim to avoid...



--
The FREE e-democracy Project
Promoting Free Software in Government

[ Parent ]
Any links? (none / 0) (#39)
by Kasreyn on Tue Jul 10, 2001 at 05:42:19 PM EST

If you could give me a link to that dissertation, I'd be very interested in reading what you thought of this issue. Thanks.

-Kasreyn


"Extenuating circumstance to be mentioned on Judgement Day:
We never asked to be born in the first place."

R.I.P. Kurt. You will be missed.
[ Parent ]
Dissertation (none / 0) (#41)
by jeep on Wed Jul 11, 2001 at 07:59:01 AM EST

Take a look here for the dissertation.

If you have any problems reading them then just drop me an email.



--
The FREE e-democracy Project
Promoting Free Software in Government

[ Parent ]
Some of the criticisms miss the point (5.00 / 1) (#37)
by dpolson on Tue Jul 10, 2001 at 05:32:36 PM EST

For everyone who's pointing out flaws and difficulties with electronic voting systems, while some of the criticisms are well founded, they are also a tad irrelevant to the basic point the author is making. Sure, there are flaws and difficulties thus far, and perhaps for the foreseeable future, with electronic voting systems. However, this just makes it all the more glaringly obvious that if you're going to have them at all, proprietary is clearly the *wrong* way to go. Free Software is the only good route to use for this kind of thing. No, it does not instantly wave a magic wand and make everything better. But the criticisms raised elsewhere in this discussion just underline the central advantage of free software for this kind of application: it allows hackers with criticisms of the system to come up with fixes for the problems they see, and contribute them so that the system is more robust for everyone. Closed source can't do that--the criticisms would remain valid, but wouldn't be addressed. Rufus Polson

Why Electronic Voting Software should be Free Software. | 41 comments (37 topical, 4 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!