Bitcoin Forum

http://www.youtube.com/watch?v=OynCgwmD-HM&feature=player_embedded

Again no transcript or factual source?

I'll happily write a transcript for a fee. The video includes direct, primary sources.

Thing is, if it was factual, it would already be in writing.  Its not so I'll wait for someone with facts to come along.

Thanks anyway.

Does this guy understand what a representative sample is?  Does he understand statistics at all?

He thinks exit polls don't work because they're "0.3% of the total."  LOL

http://upload.wikimedia.org/wikipedia/en/1/1d/Marginoferror95.PNG

Fair enough but what doesn't make sense is that the same exit polls are used by everybody, and that the exit poll results end up being the final results for the election by every news source.

That is what's fishy. Then there's the fact that the results are verified and sent out to all media outlets by a single private, media company.

http://www.businessinsider.com/ron-paul-just-won-a-caucus-even-though-the-media-is-telling-you-mitt-romney-did-2012-3

Here's a verified incident.

But that isn't fraud.  Admittedly I don't understand how Americans select delegates but on the rules, Ron Paul got more votes and less delegates.  It wasn't hidden - it is just your funny system.

Ron Paul couldn't even win Iowa...

Due to his 8 main districts being exempt from the caucus results.

I'm not familiar with this story, however, I am totally against any Electronic Voting systems; people should outright refuse to use this "scam machines"; I am so surprised (well i should not be) that ppl are quite and do not see the inherent risks and outright thievery of elections that these machines allow.

Good old paper systems; nothing can beat them, especially if you have candidate representatives at each polling station during the vote counting stage;

Quite the opposite, with electronic systems it's not particularly difficult to make fraud almost mathematically impossible. That the electronic systems in actual use don't do that speaks volumes about the priorities of the people who deploy them.

It is possible to devise electronic voting systems that:

1) Make it possible for a person to prove that a vote wasn't counted if a vote was not counted.

2) Make it all but impossible to determine how any particular person voted.

3) Make it all but impossible to coerce a person to vote a particular way.

4) Make it all but impossible for any votes to be altered after they are cast.

5) Make it all but impossible for unauthorized votes to be slipped into the system.

I know you "could" make a machine that is anti-fraud; but as experience has showed over the last 10-20 years, no government is advocating such well designed machines; that is why i personally believe that we should not accept any voting machines, and continue to use the paper based method for elections.

I am not saying that paper based methods are perfect, however with "physical" voting, it is easier to detect and reduce fraud, especially in less developed societies.

A system fulfilling those criteria would need to eliminate most of the reasons why an electronic voting system would be considered beneficial.  it adds cost, complexity, and multiple known and unknown points of failure for no effective gain over a well-implemented paper system.

I have participated in numerous Australian elections (being a delegate for candidates during the counting phase) and i must say it gives me the greatest feeling of pride and comfort that we have such a well run process from start to finish (paper based) which has such simple mechanisms and protocols in place to minimise possible fraud to the minimum.

Now you can't really do that with electronic voting machines; because most components are not "visible" and it would be easy on so many levels to hack such devices, all the way from the chip stage to the final product and software, a dedicated fraudster can infiltrate and rig such elections.

http://www.youtube.com/watch?v=iQXeGtfqCOw

Ran across this.  Seems interesting.  No idea whether it's real or not.

that link is b.s ; it's obvious that it is a cut and paste job.

The major gain is 1, 4, and 5. Even well-implemented paper systems do nothing about 1 and 5. As for points of failure, paper systems have points of failure too. It's hard to handle paper in a redundant way like you can with data.

Once they make an open-source open-design voting machine that is certified by multiple personally accountable technicians and has visible holographic tamper stickers everywhere that I can verify, THEN I might consider supporting electronic voting. Until then, I think the risks from paper ballots are way smaller. Joel's idea is better - but so far just in theory.

oh boy, not this again. Wasn't this debunked on reddit?

The whole point of the design is so that you don't have to trust the machine. You can make it mathematically impractical for the machine to do anything but what it is supposed to, and you can ensure that if it fails to do anything it is supposed to do, it cannot hide that. It's not just theory, the algorithms and methods are known. But nobody has any interest in implementing them.

Perhaps.  But if so, it's obviously not pieced together word-by-word.  And the phrases are interesting enough in and of themselves.

I'm interested! :) Please link some further reading material. I had always assumed a need to either forego secrecy or trust some entity.

There are systems for most of these things, but to my knowledge in the end you still have a box pretending to do this, while it doesn't.

A lot of these systems can be read upon on Wikipedia or books about cryptography. Only problem is that even if it works, I think it is too easy to trick most people or be lied to. You know, it is like Bitcoin. Everything works by mathematical rules and could show them, but also create a program that does something completely different, but still looks like that. And while you can find out by disassembling, having a closer look or something like that it can be hard to do that with a voting machine.

Also you only have to trick some people, to cause that one percent difference. You can trick people, also on stuff that you can (easily) verify.

However, you can also brain wash people with media and in a certain environment you can be as "smart" as you want and still would believe it, because of how our brains work.

That would result in cast votes not in the final count. Anyone in possession of such a cast vote could demonstrate that it was not in the final count. So while a machine could do that, it would be detectable. So long as the system ensures all cast votes come into the possession of some auditors (which is not difficult to do) all such votes would ultimately be counted.

That's why the voting machine wouldn't be the device you'd trust. The voting *machine* would be like the Bitcoin network. You don't have to trust it because it *can't* break the rules. It *can't* generate a spend for your coins without your key and it doesnt' have your key. If it doesn't process a transaction, you have that transaction and you can demonstrate that it wasn't processed.

The idea is not to convince everyone that the voting system is sound. The goal is to product a voting system that actually *is* sound.

So if I'm not trusting the voting machine, where do I cast my vote? Without a technical explanation it's hard for me to understand how physical security is not a problem. With Bitcoin I can take personal responsibility for securing my computer - with voting we need to ENFORCE security so people can't be coerced.

You cast your vote in the voting machine, but you also walk out with your vote (and a collection of other votes) secured in a device that you bring with you that is open source (and auditors also get a copy of the votes). If any vote in your device doesn't appear in the final count (or the auditors don't have it), you can provide the signed output of your device. It's a bit complicated to explain in detail, I'll try to put in more details when I get a chance.

A better choice would be to [ur=https://en.wikipedia.org/wiki/Sortitionl]randomly select people to act as representatives[/url].  Maybe get 3 people randomly selected per district.  Pay them a good wage and only let them server for 2 years.  There is no reason to need voting systems.

No wage is enough. Centralizated representatives - either elected or randomly selected - always provide a clear target for bribery and politics. Especially if they are all lame ducks.