Ah, but remember, Andrew Johnson was impeached during Reconstruction and he didn’t step down. The reason why Clinton didn’t step down and Nixon did, I personally would argue, was that Nixon was facing some SERIOUSLY heavy charges-and most likely prison. Clinton wasn’t.
My guess is that small scale fraud would be impossible to detect. Large scale fraud though should be detectable.
Having the paper trail wouldn’t help. After all, with the old paper system there was voter fraud. Also, trying to verify manually would be a task beyond the resources of most states in any case (i.e. if they could just count the ballots manually they WOULD be doing that as it would be cheaper). Think about what happened in Florida with the famous butterfly ballots and the hanging chads and such.
It would take weeks to manually go through the paper trail and in the end whatever results you arrived at would STILL be controversial…and probably inaccurate or flawed by the same fraud that you originally suspected.
Test the systems of course and then certify them and lock them down to prevent tampering. Similar to how databases are tested that hold critical data in OTHER fields…you don’t think they print everything out and then have folks go through the printouts to verify the data do you?
Like I said, it’s probably impossible to eliminate small scale fraud…but you could certainly do a lot to ensure none of the wide spread fraud people are afraid of (of course, since we are talking about multiple vendors today, those fears get into pretty much CT’s at this point).
They have people who are there to judge the vote already. All you need is to ensure the machines are validated and then locked down…and then procedures to ensure they stay that way. I shouldn’t think this would be THAT technically challenging…not if the government actually writes the specs to get what they and we NEED for this job. We go off on paper trail tangents instead of addressing the crux of the issue and getting the machines we need.
-XT
In my voting precinct, our electronic voting machines have a little receipt that prints out as you go, built into the machine and behind glass. I can see the paper record of my votes as they’re cast. Once I even voted for the wrong guy and then immediately changed it on the screen, just so I could see how it printed out on the receipt. I don’t understand why all the machines aren’t like that–I certainly felt a lot more secure seeing the printout.
Ok, I’ll try to explain this as clearly as I can:
If you lock the machines down completely NOBODY will be able to verify that they will in fact do what they are supposed to do.
If you use paper ballots in a simple container ANYBODY with half a brain can verify that it works.
If you suspect fraud with a machine after the voting, you will NOT (and please trust me on this) be able to prove it, if the fraudulent party has done his homework. You WILL have to take the word of whoever did whatever he did to check the machines at some time before the votes were cast, and the word of everybody who’s been in any kind of contact with the machines. Just verifying that a single machine works as specified will be VERY difficult quite expensive and won’t prove anything about the other machines.
I know how computers work. I’ve been coding for a living for over 10 years. I cannot verify a voting computer works as specified. You won’t be able to find enough qualified people with enough time to check every machine.
Verifying every machine thoroughly would be orders of magnitude more difficult and expensive than counting ballots by hand.
There is no cheap, fast and easy way to do anonymous voting correctly. Paper ballots are easy to check and cheap. It will take a couple of hours more for the results to roll in. Do you really want to give up reliability for that?
:dubious:
Sure…that’s why there was never any voting fraud before there were computers…
Er, what?
I don’t trust you on this. You are laboring under the unwarranted impression that I don’t know anything about how secure systems and databases work. Let me assure you…you are wrong. I concede there are ways I could rig the programming…however, I could rig the paper trail as well (I can think of several ways to do so in fact). And this leaves aside the possibility of fraud or error in the manual counting phase…or of fraud in whatever readers you would use to automate the counting of the paper ballots.
No, you won’t (well, ok…yeah, you would, to a certain degree, though I’m pretty sure one could devise a way to certify the box for the vote with a minimum of trust needed, at least at the polls), but that’s beside the point. You would, after all, have to trust the manual counters in the event of paper ballots, or the automated readers, in any event.
Horseshit. The machines could be certified in a central location using a verified image that has already been tested and approved, and then locked down for the vote. Randomly they could be tested after the fact to ensure none of the code was tampered with. I’m starting to question whether YOU actually know anything about this stuff to be honest.
I fail to see how a paper trail is going to make things better in any event. You’d have to somehow certify that the paper was stored properly and un-tampered with, and then you’d either need to use a reader to scan the paper ballots or you’d need to use people to do so…which would be a pretty large undertaking.
I was a database PA for 10 years and have been an IT engineer now for 20, so please don’t try and bullshit me with your creds. I can think of several ways to verify a electronic voting system off the top of my head, and I haven’t done this stuff in over a decade. Either you don’t know what you are talking about or you are trying to make this more complex than it is in order to make some point. There are secure systems out there with critical data, and they don’t do a core dump every time they want to make sure the friggin data is correct. If you are a programmer who works on secure systems you should KNOW this already.
The key would be to certify the CODE, and ensure it was doing what it was supposed to do. Propagating that code to the boxes, then running diags on them and locking them down would not be a big challenge. Nor would running diags on the boxes AFTER the election to verify that the code wasn’t tampered with be a huge challenge.
Horse. Shit.
-XT
I think you’re underestimating the stakes and the problems anonymous voting poses. Also, the basic voting mechanism needs the trust of the public. Not just the trust of the experts.
How do you propose to verify the boxes?
edit: see also: http://www.wijvertrouwenstemcomputersniet.nl/English
Both as a computer programmer and a reasonably sensible person it seems obvious to me that the best way to do this is have the computer print out a copy of its record which the voter can verify (or not), which is then placed in a bin. Further, I would suggest printing a non-consecutive unique ID on the printed copy do that it can be cross-referenced against the computer data later, and futher desiging the entire printed copy to be readily scanned back into a computer and interpreted.
Such a printout would of course not be subject to any funny business about hanging chads or whatnot, since the voting machine would print unambiguously; further, with the unique ID, the results could be readily cross-checked against the printed returns after the fact, making it essentially impossible to fudge the data or fabricate or quietly lose printed copies. It would be the best of all possible worlds.
And there is no practical reason not to do this, to leave out paper printing entirely. None. Except, of course, if you don’t want the votes to be secure against tampering…
The Irish government spend some huge amount of money implementing e-voting. The machines were scrapped after concerns were raised over security. In the case of these machines I think it was shown that their innards could be accessed very easily using a generic key.
http://www.bradblog.com/ The Brad blog found problems in voting a long time ago. They are in contacy with a programmer who said he programmed computers to read 51-49 Repub. he was zeroed in on voter fraud ,bad machines and programming for over 5 years.
Remember that electronic voting is actually not very common. Most places still have paper ballots. What non-Americans sometimes fail to understand is that there are no country-wide elections in the US. Every election is handled on the state level, certified by state election officials. Even the presidential election is really 50 state elections, and each state chooses a slate of electors who will eventually choose the president.
So the spectre of massive country-wide vote fraud is overblown. Election fraud has to take place state by state. And election fraud is only worthwhile in swing states. If you’ve got an overwhelming party majority in a state there’s no reason to rig the vote because you’ve already got the vote sewed up. But rigging the election in a swing state is harder because there exists another party that is just as interested in winning. You need a single party with no oversight to really rig the election, and in a swing state those conditions aren’t available.
That said, electronic voting should be done away with. The simplest method is a paper ballot that can be scanned electronically…fill in the bubble for the candidate of choice, then it gets scanned like the familiar standardized tests we all took in school. This way a paper ballot exists that can be counted by hand if the results are in question, yet the results are available instantly just like all-electronic voting.
While there may possibly exist methods of making electronic voting tamperproof, that doesn’t matter. The voting system not only has to actually BE tamperproof, it has to be demonstrably tamperproof, otherwise the voters don’t have confidence in the system. It’s not enough to demonstrate that the system is tamperproof in a way that only people with a background in database administration can understand, it’s got to be understandable by my grandma. If my grandma can’t understand why the system is tamperproof, then the system is a bad system.
I agree and it was pretty much what I was going to post before I saw yours. As an additional check random locations should be audited to confirm that everything is kosher. Candidates could also call for a check in close elections. This stuff should be simple with current technology.
No, you are wrong. For starters, ATMs have printers, and video cameras. Two, even with printers and video cameras, they also have modems which call in. Even with all that, there are people who routinely review ATM activity over the course of operations just to look for anything fishy as well as to see if the machines need something. And, there was at least in my bank, a dot matrix printer which spewed out reams of paper to document real time various logs. Even if someone manages to defraud an ATM or the wire transfer system, they can’t fully cover their tracks and manipulate the logs because some of those logs are sitting in hard copy in some bank. The wire transfer system has encryption methods that outstrip SSL, yet, at least as late as 1998, they wanted that printer going realtime just to be sure.
With all that physical security and ATM has, it is sometimes necessary to review the video from the ATMs to see what the hell happened. I don’t recall ever having to check the dot matrix print outs while I was there.
You can’t put the video camera on the voting machine, not and keep it anonymous, but that is the final check for an ATM to see if what it is telling us is happening is really happening. You can’t put a modem in the voting machine so some real person can watch the activity and send someone out if it looks fishy. We routinely did that with ATMs. I know, I got stuck several Saturdays babysitting the ATMs and I made a few of the calls.
Diebold makes fine ATMs. They had a reputation for quality when I was at the bank. I don’t trust their voting machines.
In my case because I’ve been a scrutineer and I am stupidly honest.
In addition to that with two poll-clerks, at least one scutineer (often two or three) there are a lot of people that would have to be in collusion in order for them to pull off a fraud. This is not to say that it doesn’t happen but that at least there are checks against it.
Even if there is fraud, if one of the candidates insists on a recount then there is a recount/audit done by people that know what they are doing.
I suppose in my OP I should have stayed away from specifically mentioning the voting computers (that just happened to be the thing that made my girlfriend and I both exclaim WTF) and stuck with the shennanigans alleged to have occured in the 2000 and 2004 elections.
The consensus of opinion worldwide seems to be that there were significant irregularities in both elections. These were widely reported on by American news, the BBC, Rueters, the ABC, the CBC and countless other “reputable” news organizations.
Not being American I can only guage by what is said by varous sources and I have been left with the distinct impression that American democracy is in dire shape; but not being American I have no real feel for it nor do I have any direct experience with it.
I am concerned about this because what my southerly neighbours do directly affects me and the integrity of the electoral system and results is therefore very important to me.
Simple solution to the voting computer problem - print out a reciept (like an ATM) that indicates which candidate the voter chose, have the voter fold the slip so the choice is not visible and drop it in a box.
That way if someone calls bullshit on the computer results there is a hard copy that can be used to verify or refute the results. No loss of anonimity and an increase in integrity.
And with a record you could do exactly the same with no one the wiser - unless part of the deal forthe $20 is that you write your name on the ballot (which would spoil it anyway) so that your benefactor can verify that they got their sawbuck’s worth.
http://www.cbsnews.com/stories/2004/07/28/sunday/main632436.shtml There are a lot of problems with voting machines. We can question the manufacturers, the programmers and those running them.
Yeah, I knew that. Here’s another bit of trivia for ya: did you know that the president of Diebold, a “staunch” Republican, promised in a speech to deliver Ohio to the Republican Party prior to the 2004 election?
Can you tell me what you think was going on here? If it was hanky panky, it lasted for years and was coordinated by several people:
At one precinct I voted in, they always wrote a number on your page in the big book of voters, (the number was what voter you were in order that day, so if you were the 5th voter they put a 5 by your name and a 5 on the ballot) and then put the same number on your ballot. I tried asking questions about this and found out nothing other than it did not happen at most precincts. It stopped when they changed to electronic voting with alternate paper ballots. I chose a paper ballot and no one wrote on it. I have sinced moved.
NY: 50 Percent of Sequoia Voting Machines Flawed | WIRED The machines flunk when they are tested. The original machines did not come with printers but they could be plugged in. That is if you knew enough to remove the plate covering the access hole.