- Site menu:
- Latest Discussions
- Latest Newswire
- Archives
- Submit Story
- Story HOWTO
- Syndicate
- E-mail Alerts
The most important open source system: Voting
|
|
|
Comments (48)
Missing two thirds of the equation.
While good, his article misses the fact that you also need open hardware and an auditable paper trail. Open source by itself is not a complete solution. The entire system must be open.
When at the university, I learned about at least one electronic voting algorithm that *seemed* to not require any of these two (not that having them is a bad thing). Each voter would get a code, with which they could verify that their vote was tallied correctly on a page *publicly* listing all these codes. This way it was fully anonymous, verifiable by all voters and everyone could sum up the scores. And it wasn't that complex, too... But I ain't security expert so I won't say if it's really secure :)
In all this fuss with voting machines it looks to me like the main failure of producers of those machines is that they think it's just a matter of incrementing counters when a button is pushed. I can only cringe at that thought and pray it ain't so.
In all this fuss with voting machines it looks to me like the main failure of producers of those machines is that they think it's just a matter of incrementing counters when a button is pushed. I can only cringe at that thought and pray it ain't so.
Quoting:While good, his article misses the fact that you also need open hardware and a auditable paper trail.
That's what they thought in my country after the proprietary-voting machines debacle, so they went searching. Finally, after a very long search, they found the solution. It's very open, and dubbed: "Red pencil and paper". It is really brilliant: The "hardware" is open (well, in fact during the elections the ballot boxes are closed), there is an auditable paper trail and there is no software needed, meaning no software can be closed. Procedures are open of course, and it even seems this system is currently not covered by patents/copyright meaning no copyright/patent holder can exert control over the democratic process. Moreover, they calculated this system is cheaper than using the old proprietary voting machines. Wow, what a great insight!
> It's very open, and dubbed: "Red pencil and paper".
Which is exactly the same one we used (with black pens, admittedly) until last election. Then the "experts" decreed that we must use electronic machines. :( Sometimes the old ways really are the best ways.
Which is exactly the same one we used (with black pens, admittedly) until last election. Then the "experts" decreed that we must use electronic machines. :( Sometimes the old ways really are the best ways.
> ...they could verify that their vote was tallied correctly on a page *publicly* listing all these codes.
If the codes are made public then it's not really a secret ballot. With a public listing, It's fairly easy for those voting one way to determine who must have voted the other. That's not really a good trade-off.
If the codes are made public then it's not really a secret ballot. With a public listing, It's fairly easy for those voting one way to determine who must have voted the other. That's not really a good trade-off.
Quoting: If the codes are made public then it's not really a secret ballot. With a public listing, It's fairly easy for those voting one way to determine who must have voted the other. That's not really a good trade-off.
How on earth are you going to do that? Perhaps if the listing was only for a really small area with only a few people, but if the listing contains many people (like the entire US) it's impossible to find out.
They could simply create a website with all the codes on them, and offer free access to that website in certain government offices for people who can't go online.
It occured to me that I was stretching things a bit with this claim of lack of need of trusted hardware. Sorry :)
Pen-and-pencil voting is fine for me and I don't advocate changing it, BTW.
Well, I'd like to see 10 million voters revealing their votes just to nail the other 10 million voters... [it's my country's scale] This may be a hole, but probably it's the best we can get. Debian does this way, BTW (see for example http://www.debian.org/vote/2008/vote_001_tally.txt).
Pen-and-pencil voting is fine for me and I don't advocate changing it, BTW.
Quoting:If the codes are made public then it's not really a secret ballot. With a public listing, It's fairly easy for those voting one way to determine who must have voted the other. That's not really a good trade-off.
Well, I'd like to see 10 million voters revealing their votes just to nail the other 10 million voters... [it's my country's scale] This may be a hole, but probably it's the best we can get. Debian does this way, BTW (see for example http://www.debian.org/vote/2008/vote_001_tally.txt).
born incompetent, or do they get special training?
For deity's sake, is there any simpler function for a modern computer than COUNTING?? As the article says, it's either incompetence or corruption. My vote goes to both.
TC, Nobody is THAT dumb. Gotta be corruption.
Unfortunately, TC, voting is not JUST counting...
But producers of voting machines are too incompetent and corrupt to do even that ;)
But producers of voting machines are too incompetent and corrupt to do even that ;)
Never assign to malice that which can be adequately explained by stupidity.
I'm with TC. It's stupidity, in that Diebold et al., and their government enablers, thought they could cavalierly replace a system with known problems, with another system, untested, bringing in a whole new set of known and unknown problems.
And it was malicious, in that they didn't care, as long as they got their $$$.
And it was malicious, in that they didn't care, as long as they got their $$$.
I worked for one of Diebold's companies for a couple of years. Trust me, the phrase fits.
Why is any e-voting necessary in the first place?
14 earlier comments in the full thread.
Quoting:WHY it is necessary to e-vote
Good question. It isn't. BUT That's only because we're still using a system developed for a world before railroads and telegraph. People, we no longer live in this world.
The thing is, it is not necessary, but, if done right (and I mean: REALLY right, if that's at all possible), it could *enable* us to do better.
You think our current system works good enough? That's because <stab-in-the-dark> you've never tried anything else </stab-in-the-dark>. That's the only thing we have so we stick to it. I say it's not good.
Two things I hate about our current system is that it: - uses primitive either-or logic (you can imagine how many times I wanted really bad a "-1" vote instead of a "+1" one; and Condorcet method would be like Nirvana) - is discrete in time domain with insanely large quanta (I mean -- someone needs to be very popular for a single day every few years and after he/she gets elected doesn't give a ****)
I'd like both of these changed (continuous elections with a possibility to change your preferences any time you want with immediate effect?). Electronic voting could change it IF done right.
Of course, this is a really REALLY big IF.
There, I've said it. You can bring out your stakes and pitchforks.
jezuch, are you suggesting we fix the bugs as we encounter them instead of hiding from them?
That is just ridiculous.
That is just ridiculous.
Quoting:Doing as you say, if you mess with the machine to get a time-ordered list of the verification codes, all you have to do is to write down the order in which people entered the booth, print the code list,, put the two pieces of paper side by side and secrecy goes out of the window.
Your assumption is false. You assume a machine that can be tampered with. That means that the security experts who designed it fouled up. It remains an implementation detail.
Sander, the core of the problem is we can't trust anyone involved with these dratted machines.
Quoting:I propose returning to a more definitive, much simpler form of selecting leaders: cage fighting. Stick the whole lot in the cage, sic 'em on each other, and whoever is left standing after 30 minutes wins.
Not only simpler, but more fun to watch, as well.
I'm glad to see so much interest in the issue. I'll answer to all
those who answered after my last contribute in one post:
Me? Me, I don't need it, of course. Please re-read the post where I mentioned that function. I simply said that all you'd need to do to violate secrecy is record that information separately, and then match it against the list of "codes+votes", therefore it is wrong to say that it's impossible to establish " a link between the voter's identity and the code
Jezuch, what you'd like to change is much more and much deeper than the voting system itself, so it's offtopic wrt my original question and, personally, I'll simply ignore it, at least here and now.
Sander, you keep eluding my original question. First of all, I already said that I don't want to know how to patch evoting after it has been imposed. I want to know, by reading some official report, study, whatever, WHY it is necessary to do it, BEFORE going into "implementation details". What I'm saying is that I'd like that nobody officially discussed implementation details before the first, right question has got a satisfactory answer.
Secondly, I'm a bit baffled by your answer. Of course I "assume a machine that can be tampered with", because that is the very core of the question, it's far from being an implementation detail. And there should have been no e-voting whatever before such a really tamper-proof implementation had been 100% disclosed and discussed in the open.
I'm no security professional, yet yesterday I could figure out in 30 seconds after reading your comment a way to try to tamper with it: I'm not going to dismiss such worries only on unproved assumption that "the experts are experts, they surely know how to prevent this and have done it" tuxchick already explained this to you:
In any case, this discussion is fascinating but risks to become quite time consuming. Personally I'd really rather stick to the original question and ignore for the moment anything else, if possible:
"Does ONE independent report exist which PROVES that it is absolutely necessary to give up paper voting because e-voting has such huge real advantages that we can't do without? I've never read one, please provide links if it exist."
Thanks!
Quoting:theboomboomcars: Why do you need to record the time of the vote?
Me? Me, I don't need it, of course. Please re-read the post where I mentioned that function. I simply said that all you'd need to do to violate secrecy is record that information separately, and then match it against the list of "codes+votes", therefore it is wrong to say that it's impossible to establish " a link between the voter's identity and the code
Quoting:jezuch: Two things I hate about our current system is that it:
- uses primitive either-or logic... - someone needs to be very popular for a single day every few years and after he/she gets elected doesn't give a ****)
I'd like both of these changed (continuous elections with a possibility to change your preferences any time you want with immediate effect?).
Jezuch, what you'd like to change is much more and much deeper than the voting system itself, so it's offtopic wrt my original question and, personally, I'll simply ignore it, at least here and now.
Quoting:Sander_Marechal: Your assumption is false. You assume a machine that can be tampered with. That means that the security experts who designed it fouled up. It remains an implementation detail.
Sander, you keep eluding my original question. First of all, I already said that I don't want to know how to patch evoting after it has been imposed. I want to know, by reading some official report, study, whatever, WHY it is necessary to do it, BEFORE going into "implementation details". What I'm saying is that I'd like that nobody officially discussed implementation details before the first, right question has got a satisfactory answer.
Secondly, I'm a bit baffled by your answer. Of course I "assume a machine that can be tampered with", because that is the very core of the question, it's far from being an implementation detail. And there should have been no e-voting whatever before such a really tamper-proof implementation had been 100% disclosed and discussed in the open.
I'm no security professional, yet yesterday I could figure out in 30 seconds after reading your comment a way to try to tamper with it: I'm not going to dismiss such worries only on unproved assumption that "the experts are experts, they surely know how to prevent this and have done it" tuxchick already explained this to you:
Quoting:Sander, the core of the problem is we can't trust anyone involved with these dratted machines.
In any case, this discussion is fascinating but risks to become quite time consuming. Personally I'd really rather stick to the original question and ignore for the moment anything else, if possible:
"Does ONE independent report exist which PROVES that it is absolutely necessary to give up paper voting because e-voting has such huge real advantages that we can't do without? I've never read one, please provide links if it exist."
Thanks!
> Does ONE independent report exist which PROVES that it is absolutely necessary to give up paper voting because e-voting has such huge real advantages that we can't do without?
Well, since pen and paper voting can demonstrably do the job, I don't think such a study could be believed if it was found. Now, that doesn't mean that e-voting, if done correctly, might not be better than pen and paper. But there's now way it's absolutely necessary.
And as I noted earlier in another comment, if you want e-voting open software is only part of the picture. You also need open, verified hardware and a fully auditable paper trail. The entire process must be open to peer review and public scrutiny. Given those, I think it could be made to work. I disagree with Sander about the problems with making the votes publicly accessible. I think voter data mining is not only possible, but inevitable in such a situation; up to and including identifying individual voters. Verifying the paper trail at the time you place your vote should be sufficient.
Well, since pen and paper voting can demonstrably do the job, I don't think such a study could be believed if it was found. Now, that doesn't mean that e-voting, if done correctly, might not be better than pen and paper. But there's now way it's absolutely necessary.
And as I noted earlier in another comment, if you want e-voting open software is only part of the picture. You also need open, verified hardware and a fully auditable paper trail. The entire process must be open to peer review and public scrutiny. Given those, I think it could be made to work. I disagree with Sander about the problems with making the votes publicly accessible. I think voter data mining is not only possible, but inevitable in such a situation; up to and including identifying individual voters. Verifying the paper trail at the time you place your vote should be sufficient.
Quoting: Does ONE independent report exist which PROVES that it is absolutely necessary to give up paper voting because e-voting has such huge real advantages that we can't do without?
Well, since pen and paper voting can demonstrably do the job, I don't think such a study could be believed if it was found.
precisely my point, and I also agree with you on the risks of voter data mining, thanks. However, I'm still interested in papers on that theme, so if something even just similar to them exists, please let me know by email at marco, at digifreedom; dot net, thanks
Quoting:And as I noted earlier in another comment, if you want e-voting open software is only part of the picture.
of course, but me, I don't want e-voting period (see again http://digifreedom.net/node/52). IF we really can't get rid of e-voting, then of course every part has to be as open as possible.
But the reason why engaged in this and similar threads in other places is that I'm frankly baffled to see so many FOSS supporters advocate FOSS for e-voting without, at least apparently, stopping for a moment to ask themselves whether the basic thing makes any sense or not.
It sounds like "I like FOSS for the sake of FOSS, and finding bugs in source code so much that I want e-voting just to have one more container of source code I could play with;even if the thing is flawed at its roots no matter what the source code (or HW design) license is"
Marco
> ...so if something even just similar to them exists...
The closest I could find with a quick Google search was an article by Bruce Schneier. While somewhat negative, it does discuss some of the reasons electronic voting may be preferred.
You can find it at http://www.schneier.com/blog/archives/2004/11/the_problem_wi...
The closest I could find with a quick Google search was an article by Bruce Schneier. While somewhat negative, it does discuss some of the reasons electronic voting may be preferred.
You can find it at http://www.schneier.com/blog/archives/2004/11/the_problem_wi...
Marco- E-voting isn't necessary.
But we live in a digital age and people seem to think everything needs to be computerized in some form or another.
Having a quick look at your site I would say that all of your concerns would be addressed with an open system. Is it necessary, no, but we could enjoy the convenience without the security issues with an open system.
I would agree that the current system is broken and shouldn't be used.
Having a quick look at your site I would say that all of your concerns would be addressed with an open system. Is it necessary, no, but we could enjoy the convenience without the security issues with an open system.
I would agree that the current system is broken and shouldn't be used.
@mfioretti
One advantage is that, theoretically, e-voting can reduce cases of double (or more) voting, and rigging in general. Another is, of course, reduced chance of mistakes. This is assuming that it has been verified by third-party that the e-voting machine works as advertised. Using FOSS for e-voting will allow verification of this (again assuming that this is what actually is going into the machines) in a much broader and transparent way.
Quoting:I'm frankly baffled to see so many FOSS supporters advocate FOSS for e-voting without, at least apparently, stopping for a moment to ask themselves whether the basic thing makes any sense or not.
You're preaching to the wrong choir. We're FOSS. We take proprietary system and build open systems that work better. Whether that's servers, mobile phones or e-voting machines. You should talk to the people who decide between e-voting and paper&pen voting, not to us.
Exactly Sander, we are the people who you talk to about how hard is it to add 1+1 (which the evoting folks seem to have problems with) and why it would be easier fix all the associated problems with Open Source. The idea and ideals surrounding e-voting itself ... thats not our gig.
Quoting: The idea and ideals surrounding e-voting itself ... thats not our gig.
Why not? We're citizens who presumably have a say in these matters. Do you really think that should be left up to people who can't add 1+1?
If they can't figure out 1+1, they shouldn't be using Intel hardware.
/rimshot
/rimshot
TC, might as well expand it to world hunger, teenage pregnancy, global warming, etc etc. These things may interest one or another of the readership, but is pertinent in this venue only as OSS is relevant to the discussion.
i.e. passing moral judgment on the voting practices and habits of the lemmings at large ... not our gig.
i.e. passing moral judgment on the voting practices and habits of the lemmings at large ... not our gig.
> ...passing moral judgment on the voting practices and habits of the lemmings at large ... not our gig.
More accurately, this is not the appropriate forum, though we often touch on such matters in passing. And occasionally linger longer than is appropriate. :)
More accurately, this is not the appropriate forum, though we often touch on such matters in passing. And occasionally linger longer than is appropriate. :)
Quoting:You're preaching to the wrong choir. We're FOSS. We take proprietary system and build open systems that work better. Whether that's servers, mobile phones or e-voting machines. You should talk to the people who decide between e-voting and paper&pen voting, not to us.
this is just the attitude that makes me sad. In this forum I have only discussed technical problems and asked for technical, concrete, business-like justifications for e-voting. If something is, technically speaking, inherently insecure, is it an ethical issue offtopic for this forum to say let's avoid it?
The way you say it, "We're FOSS" is equal to "we're citizens with enough technical skills to recognize a dumb idea, but we won't use those skills to stop the dumb idea, we'll just turn it into a FOSS dumb idea". Bah
E-voting should have been treated by geeks as some Linux hackers treated winmodems when they first appeared: "I'm FOSS, I'm surely skilled enough to reverse engineer this proprietary winmodem sw into a FOSS one which would work better. But I won't waste my time on it, because a software modem is an idiot idea at the purely technical level, without ever going to bother about the morality of its license".
Thanks for the link to Schneier's article, both it and its comments are worthwhile reading.
Marco
> Thanks for the link to Schneier's article, both it and its comments are worthwhile reading.
De nada. Glad to be of assistance.
De nada. Glad to be of assistance.
One of the reasons that when "voting" finally makes it to a country where they haven't had that particular stage of evolution yet, they always demand paper ballots, is a matter of trust.
Without a paper trail, there is no way to audit. If the machine says N votes were cast for X, that's it, and whoever has access to the machine wins.
E-voting gains traction in the US because most people trust the system. Fewer every year, but still enough.
I see no way to reconcile "secret ballots" and "e-voting". One or the other, ok, but both is just too easy to corrupt.
Without a paper trail, there is no way to audit. If the machine says N votes were cast for X, that's it, and whoever has access to the machine wins.
E-voting gains traction in the US because most people trust the system. Fewer every year, but still enough.
I see no way to reconcile "secret ballots" and "e-voting". One or the other, ok, but both is just too easy to corrupt.
Post a comment
A free account is needed to post. Write your comment here first: your text is kept in this browser while you sign up or log in, and is waiting for you when you come back.
Your message:Markdown: **bold**, *italic*, [text](https://link), - lists, > quotes, $x^2$ for math. Formatting help
Sign up and post (it is free) or Log in and post