Memme: Computers count votes so no one can see, They claim Bush won...

I say prove it to me.

People do not understand that tabulation fraud is invisible. Even in punch card counties, a software program tabulates the vote. One computer programmer is all that is needed.

I don't buy it. Quite a lot of people see each line of a commercial computer program. And they do test the tabulators and direct-record machines. A hack clever enough to be unnoticed on testing runs, but to show up on the real thing, that's tricky. Half the company would have to be in on it, and just because the CEOs of ESS and Triad are republicans--aren't most CEOs?--doesn't really make me think you could keep a secret this big.

Message removed by moderator. Click here to review the message board rules.

people much more confidence in the machines. Unfortunately, under U.S. election laws, that information is confidential to the companies that make them, which as you conveniently noted, have Republican execs. This is done for "security" reasons for the corporations.

I simply can't comprehend how any sane person can see the mountain of "glitches" and "irregularities" and come to any conclusion but that the election was stolen?

and I can vouch for how easy it can be to hide code. Apps get very large, with millions of lines of code.

It is simple to hide an "easter egg" into an app, and easy to keep it hidden.

An easter egg won't do it. This has to change the results of a national election--and be completely invisible for every tabulation run when results are crosschecked, and steal an identical number of votes on each other tabulation.

Also, field operatives across the country need to know how to trigger it. It's a different ballot in each location and in each election, so you can't hardcode things. It needs to be configured who to steal how much from, separately in each jurisdiction.

The results of such a hack wouldn't need to match numerically or even statistically, and in fact would not match, as demographics for various areas are often vastly different. Thus, you have one (1) single formula in the code to handle it all. It could be as simple as Kerry_vote = (Bush_vote - 1).

Field operatives wouldn't even need to touch anything to trigger it; it could be triggered by the machine's internal clock, for example. Something like

if datetime = X
Kerry_vote = etc...

And it doesn't need to be calibrated if a candidate appears on each ballot. All you need to know is how the code refers to each candidate, which the programmers would know.

What you're saying is akin to saying that the easter eggs in Half-Life 2 will look different and appear in different places (or not appear at all) depending on what system specs each user has. Simply, transparently, untrue.

The names 'kerry' and 'bush' appear NOWHERE in the system as produced by the manufacturer. All the system knows is "votes for candidate 1"--and which guy appears as candidate 1 on the ballot depends on the state, and maybe even the precinct. So you can't easily hardcode it.

There's easy no way that the code can figure out which array index in its table of total votes is for which candidate.

I'm sorry, but if you're looking to the voting machines for fraud, you'll looking in the wrong place. My theory is absentee ballots.

Which they were.

Open the source. Let's see if we're right or wrong.

I'm not just being stubborn here. Those companies really do own their sourcecode, and making it public really would deprive them of this.

It really is unconstitutional, no matter how much you and I want to see it.

I don't buy it, not for a skinny second.

I'm a goose, Diebold is a gander.

Open the fucking source.

Companies are free to keep their code hidden, but it should be in bid specs that only machines with open code will be allowed to bid.

That's a good policy for the future. I'd write my state government pushing it, except that New York doesn't use electronic voting or tabulation.

and they have almost all mechanical lever machines.

However, this doesn't change the fact that the vote is completely unverifiable in states without VVPBs and random audits. It just shows that we shouldn't blindly rely on these exit poll anomalies to prove fraud.

suggestive of tabulation fraud.

Especially in Butler, Warren Clermont and Hamilton County in SW OH.

The programs used by any computer may easily be changed to produce a
fraudulent result by any competent programmer who has unfettered access.

The only absolute way to eliminate fraud is a hard-copy produced by the voter that is counted by hand.)


A vendor would not have to be in on it to achieve fraud. Only a single programmer with access to the tabulation equipment would have to be in on it. The fraudulent tabulation program could easily use a "switch" triggered by an unusual vote combination, a time-based trigger, a threshold of votes cast, or any combination of the above. Bu using a trigger such tabulation software would be able to pass pre election testing.

ALmost nobody has 'unfettered access' to this sort of thing. Every change would be logged in the source control system and everyone gets to see the changelogs.

The names on the ballot aren't hardcoded, so it's hard to have vote combinations trigger much of anything.

And remember that for tabulators, there is always the possibility of a manual recount--and we've done recounts in a lot of jurisdictions, and somehow they never seem to differ significantly from the tabulator results.

You are not a programmer obviously.

I'm a computer science major at Cornell. I know a fair bit about programming.

I'm aware that the Ohio recount was somewhat flawed--but there was no way of knowing in advance what would happen there, or which precints would be counted how. And there were recounts in many places, not just Ohio.

Nobody would run a risk like that, when it's so unnecessary.

counties. County BoEs in conjunction with tabulation vendors appear to be complicit in choosing certain precincts. Whether this is because of laziness or fraud, we'll never know unless their is a hand recount. Of course with chain of custody issues being violated in some counties, even hand recounts of punch cards is problematic.

They did a whole-state recount in Washingon--and the tabulators did fine.

Ohio, Florida and Pennsylvania were the make or break states

it but the companies. Not even the election officials. And If the tabulators are easily accessible to hacking then anything can happen.

And, by the way, money talks. How much money would you need to keep a secret?

doesn't make much...

and you own expensive creatures called horses...

Anyone could be "on the take" the proverbial "dirty cop"

Bribery or extortion

I wouldn't keep this sort of secret, and doubt that very many other programmers would either. "The company" is a huge number of people--they can't all be corrupt or evil.

Most tabulators, as far as I know, are entirely sealed off from the outside world, and thus hackproof. A few have modems that are only turned on to phone in results--but the machines are still inaccessable except for an unpredictable few minutes.

All it takes is a corrupt board. EnronTycoKmartArthurAndersenWorldcom.

If you told me that Diebold was looted to funnel money to the GOP, I'd believe you. Evil or corrupt boards can steal money, no problem.

But it's much harder for them to rig an election. They can't do it with their own hands, and it would take a lot of people to do it right.

It's as though Worldcom weren't merely looted, but were conducting espionage on a vast scale.

modems. Triad dialed in before the recount to modify software. You can call the director of the board of elections in Van Wert County and ask her yourself. She sees nothing wrong with that. I say it puts the whole election in question.

I'd like to know more about this--can you give us details?

Who knows the phone numbers for the machines? How long are they connected to a phone jack? What authentication mechanisms are there?

I'd like to know more before I start a letter-writing campaign.

complete picture is to contact them.

What these companies are counting on- if I may make a pun- is the ignorance so many people have toward computers, how they work, and what is and is not possible when programming them. That's the single biggest reason "they" got away with this- people just. Don't. Know.

Example: I have a friend who is almost completely computer illiterate. He assumes all PC products are made by Dell, HP, and Compaq. He wants to know what the "brand name" of my custom case is. When I told him it doesn't have one, because it's not from a major computer manufacturer, he said, and I quote, "Oh, but it HAS to have a brand. I bet Dell made it."

He thinks a faster computer will make his web pages load faster. I tell him it's mostly RAM and modem speed. So, he gets himself a newer, faster machine... and then complains because his web pages load at the same speed as before.

He uses a 2GHz machine... for email and chat. He has not less than FOUR chat programs open at once, and can't understand why his "computer is slow". "Shut your chat programs down," says I.

"Oh, but I need them on," he says.

He runs ADBOT SOFTWARE voluntarily!!! He says he NEEDS THE ADS!!!!! :angry:

I'm to the point of refusing to help this guy any longer with his PC, simply because he refuses to take the advice and then keeps right on complaining.

And it's people like him that truly allowed this- uninformed, and honestly legitimately ignorant people who don't have a clue about the inner workings of these machines, who implicitly trust that most all programmers are unwilling to do this.

Bollocks. Look at all the virii, trojans, adbot software, doalers, keystroke recorders, and on and on that people inadvertantly install on their own machines. They simply don't know what to look for, wouldn't know how to fix it if they did, and thus are willing to spend $$$$$$ getting something fixed (over a course of at times DAYS), when they could easily take care of the problems themselves.

I'm glad I'm smart enough with these to have never needed to call a tech support line in a decade. Unfortunately, most people aren't so savvy- like my friend I talked about, above.

People would likely be horrified that their votes were going out over a modem. I hope it wasn't wireless.... that's yet another glaring security hole.

But- people don't know, and it would take an intensive crash course in computers and viruses and such for them to learn.

Common ignorance allowed this theft.

illiterate.

And that nice trust worthy Triad rep, such a nice white young man.

As far as OH tabulation fraud, even more disturbing is the connection to DataMaxx and LEADS

into buildings. I don't want to believe that they planned all of this for years right under our noses. I don't want to believe that Dubya sat there in the school and did nothing or that he is hiding the truth about 911 from us.

Just because something is so horrible that you don't WANT to believe it, is no reason to turn your back on finding the truth.

I hadn't realized that. Hmm. I'll do some research.

Don't make the assumption that simply because a software company is relatively small and private that it can't accomplish biiiig things. Look at VLAVe- Half-Life, Half-life 2. Hundreds of millions of dollars in that company- and they have a full staff of less than fifty, I believe. Half-Life is, for all intents and purposes, Gabe Newell's baby, and he kept Half-Life 2 as secret as possible for a very long time.

Until the source code for VALVe's "Source" engine got stolen because someone got Gabe's password, that is.

daughter Cheryl Bellucci.

They probably have more sales reps than any other type of employee.

And blood does not rat.

Suppose one of those CEOs- whose company was set to make the machines doing the counting in one particular state- promised to deliver that state's electoral votes to the sitting incumbent... PRIOR to the election?

And it's very, very easy to hide three or four lines of code in an application with thousands or hundreds of thousands of lines of code. And don't forget those patches that were installed, with .dll files nobody except the company making them can look at.

No, it would be very, very easy indeed to insert malicious code into these programs, and just as easy to hide it. That's why we need to be able to independently examine the actual source code- which, by the way, is considered "intellectual property" and thus off-limits.

What do they have to hide?

Release the source. Unless you're guilty, of course, in which case, it's in your own best interest to hide it for as long as possible.

I fail to understand why the concept of malicious code eludes so damn many people... we're talking Captain Obvious here, folks...

Blood doesn't snitch...

These are large complex systems. There's way way way more than one programmer who works on each election systems. They can't all be family.

ones and not in every state, just relevant ones. That guy from Triad no problem coming and going as he pleased. It's not just one company either and all of the companies were pro Republican. It's not impossible because they did it.

All the CEOS were republicans--like CEOs generally. I don't believe all the employees were.

We have one guy from Triad who came in and mucked around, in one county. How do you know he was a Republican? How many solidly republican technicians can they hire, without ONCE making a mistake and spilling the beans to someone honest?

that only comes out years- or even decades- later?

My Lai comes to mind...

You'll find the answer to your question by considering how many people in our armed forces perform "dirty deeds" they can't talk about to anyone. Mind, there's nobody holding a leash on the guy I work with who was in a special forces unit... but he still isn't spilling the beans, except to say some of the things would make a LOT of people angry if they found out about them.

Nobody's holding his mouth shut; he's doing that all on his own. Oh, he took an oath? Meaningless words- unless you're willing to back them up, which he is.

How much effort would it take to find Republicans willing to lie, cheat, stal, and keep it quiet? Not much, given that I work with one who claimed he would have happily fiddled with the vote if he thought his guy was going to lose.

Although he's a bad example- he can't keep his mouth shut about anything for more than five minutes at a time- it only goes to show that there are people out there who can and will do anything they feel is necessary- even if it's illegal- to win.

I fail to understand why this concept is so difficult to grasp. People like this live in every county of every state in the country; it only takes finding them.

Blind, unthinking loyalty to a cause isn't too tough to find.

Rethugnakin to collect a paycheck. Guaranteed, there will be whistle blowers coming out, there always are sooner or later. There is always impeachment when that moron * screws up again, and again. Sooner or later people are going to get sick of these retards. Personally I think they already are sick of them and this election that they DID rig proves that they can't win honestly. Most scumbags can't.

It is indeed a very small operation.

Actually Triad's tabulation program was written in COBOL and runs on DOS 6 on a 386 with floppies for file transfer.

These companies make their money writing software. Having them release their source code like this really would be taking away their intellectual property--which is unconstitutional: it's an illegal taking without due process or compensation.

We of course should require independent outside code audits. I think we have a good chance of getting that in the next year or so.

It's not a four line fix, given that candidate names aren't hardcoded. You need a way to dynamically configure it who to steal from--and that isn't four lines.

Also--Ohio doesn't use Diebold electronic voting machines.

"Also--Ohio doesn't use Diebold electronic voting machines"

If you want to debate at DU you better be up to date with your info. Go research a little more before you start spouting off. We're all way ahead of you here.

it had sounded as though you were saying otherwise. My bad. *bows*

but you give yourself away by talking about intellectual property and closed code and how it's important to keep the code closed... and then make the declarative "It's not a four line fix, given that candidate names aren't hardcoded."

How do YOU know this? Easy- you don't. Because neither you, nor me, nor anyone else here has seen the source code. So, you can't say it "can't be done" because you can't know if it WAS done or not.

And yet, you're advocating keeping the source closed. That's an... interesting position, given that these programs are being used for a public purpose.

They use these machines in many different elections in many different jurisdictions, and don't write custom software for every county. Since it's a different ballot in each place, the candidate names can't be compiled in.

That's why I say it couldn't be done.

A lot of software is used for public purposes--every copy of Windows on any government computer, to start with. You can't just say "let's put Microsoft out of business" because they're writing software used for "public purposes".

Be fun to try though. ;)

could easily fit on the head of a very small pin, but thanks for your explanations and reasonable tone. You've discovered, no doubt, that some folks are so invested in the belief that there was a successful conspiracy to commit fraud on a massive scale, that they simply won't consider arguments that run up against that belief system.

that uses punch cards.

As I understand, they have to generate some configuration files, specifying which names go where and suchlike.

They can't compile separate binaries for every precint and county, every election, can they?

The tabulation computer in Hocking County was a 386. The files are transferred via floppy. Only the Triad tabulation program can interpret the "meaning" of the punched holes in the punched cards. In addition Ohio has candidate rotation by precint, so the precinct code and the ballot label are neccessary to determine if fraud has occured.

With all physical ballot systems (punched card, for example), one
possible rigging is to tamper with the ballots.

With Votomatics, installing the wrong ballot label for the precinct
can make trouble. On the Votomatic, the ballot label puts names by
holes in the card. If you tabulate hole one as a Bush hole, but your
ballot label says hole one is a Kerry hole, of course, you'll switch
votes. So, recounters should demand to see the actual ballot labels
used on the votomatics in that precinct and check the holes against
the tabulation.

In addition punch card systems have bad candidate positions due to braces under certain hole positions that result in a build up of chad.

"These companies make their money writing software. Having them release their source code like this really would be taking away their intellectual property--which is unconstitutional: it's an illegal taking without due process or compensation."

Then they should get the fuck out of the business of making software for the government/election process. If they cannot understand that the importance of transparency and the democratic ideals of "count every vote," and the importance of having everything beyond reproach, then they have no business in this business. Period.

Solemity, I appreciate your statements. And IF there was not LITERALLY the control of the world at stake, and IF there was not a virtually bottomless pit of money to throw around to help hide the wrongdoing, then I might say your points are well made. But they simply leave out these massive parts of the equation. I simply DO believe people are that corrupt. I had a CEO of a company I co-owned take my house and everything I had just to get his hands on a million-and-a-half dollars in stock I was holding. Imagine was crooks like Bush would do for trillions of dollars and control of the world.

Wake up, dude. This is hardball.

> These companies make their money writing software. Having them
> release their source code like this really would be taking away
> their intellectual property--which is unconstitutional: it's
> an illegal taking without due process or compensation.


If you read a book, does the author lose his intellectual property?

Which intellectual property are they losing? Copyrights? No. Patents? No. Those have to be disclosed anyway. Trademarks. I don't think so. Trade secrets? I certainly hope there aren't any trade secrets involved in counting votes.

> We of course should require independent outside code audits.

You've never developed commercial software, have you?

Independent code audits are as good as useless. And this was clearly demonstrated results of testing and auditing performed on behalf of the Ohio SOS.

If I'm an independent auditor, I'm going to do the least possible work to get paid for an audit. There is no possible way to force me to do a meaningful audit.

If you give me a blanket requirement like: ensure that there are no trojans in the code, any knowledgeable auditor is going to refuse point blank to provide this kind of assurance. You can't ensure that there are no trojans in the code any more effectively than you can ensure that there are no bugs. Anyone who believes that all bugs can be removed from a non-trivial peice of software really doesn't know much about the current state of the art in software development.

The actual code audits that were performed by independent auditors were along the lines of:

- Check that functions don't have multiple return points. (An archaic and obsolete coding dictum, which is a wickedly bad thing, not a good thing these days). Big deal.

- Check that variables and function names follow industry standard conventions. There isn't an industry standard convention, and even if you do arbitrarily choose one, what have you really tested here?

We've seen the also point made handily on the independent software testing reports. 83 requirements. 83 tests performed. What kind of code coverage do you think that provides? Maybe 0.5% of all source code lines. It worse than doing nothing, because it gives a false sense of assurance. Becuase the vendors write the test plans, it's easy to avoid problems. Just delete test plan steps for anything you don't want tested. Nobody will notice, and even if they do, there's nothing they can do about it.

Look at the security testing that was performed on voting machines. People with no particular security expertise poking randomly at machines they know little about. A tester can poke away at machine under test for days without breaking in. I, on the other hand, can break into a diebold optical scanner because I happen to know that the default password is 'global', and I can break into a Diebold central tabulator machines in seconds, because I know that the the default password is 'psdb', and I can break into the DREs in seconds because I know that the default password is '1111'. In this case, testing provides no meaningful protection at all.


> Also--Ohio doesn't use Diebold electronic voting machines.

Right. They use Diebold optical scanners, and Diebold vote tabulators.

development. Things get hidden all the time--from back doors, to easter eggs, to "bugs" that can be exploited elsewhere. People can and do steal large amounts of money this way, and there's no reason that votes couldn't be stolen as well.

If management doesn't push for the kind of detailed, black & white box testing and code review that it takes to find them, nobody takes it upon themselves to look. (Anybody that starts trying to do things "beyond their assigned scope" obviously needs some more work to do--and guess what kinds of jobs they get?).

--MarkusQ

the tabulation fraud was occuring where no one expected in punch card counties

Everyone seems to ignore the fact that there was enough voter suppression to flip the outcome. Who needs fancy software magic (which I also suspect may have been used) when you can force your opponent's supporters to stand around in the rain rather than voting? It's so efficient, it boggles the mind: there is no way you could mobilize enough white supremacists to keep as many black voters out of their polling places as you can by just making them stand in long lines, thus keeping each other out.

--MarkusQ

The programming is simple enough to be done by a junior high kid and could easily be written to pass pre-testing--you'd only need a date trigger, for example, to make the code lay dormant until election day.

Source code is not rigorously examined, as evidenced by the fact that leaked software HAS revealed numerous, obvious backdoors that would never pass a real examination. The testing that occurs is extremely superficial and it's conducted by subcontractors to the election software companies!

Not only that, changes to operating systems are not even covered by the rules regarding election software. Thus, someone could write a windows patch (e.g., a .dll) that does the dirty work, and there would be nothing illegal about installing this patch even on election day itself! In fact, such uncertified patches are ROUTINELY installed after certification, with company reps given unfettered access to voting machines in person, via the internet, or via modem, up to and including election day.

You're just dead wrong about this.

There are reasons why Diebold machines were barred from being used in California and there are reasons why Diebold was sued and had to come to a $2.6 million. I suggest you read up on that.

You know, republicans have done fairly well in California and Ohio even without Diebold. This is a bad talking point.

but I think it's pretty obvious you don't belong here.

Yes, fair assessment...most CEOs are GOP:

Pharmaceutical Companies?--That's why you can't get cheaper drugs from Canada...

Health-care companies?--That's why you can't get coverage without paying exorbitant fees...

Brokerage firms?--That's why Social Security is "broken" and needs fixing...

Conservative religions industry?--That's why we have "faith-based" funding...

Retail stores?--That's why we have the "Wal-marting" of employee working conditions and rights...

Defense contractors?--That's why we have the Iraq war...

Oil companies?--That's why we have the Iraq war...

The Bush Administration is government "of the corporation, by the corporation, for the corporation," and its continued presence depends on GOP-CEO voting machine companies...

They're also big campaign contributors, are members of elite republican clubs, and one was even quoted as saying that he was "committed to delivering the state of Ohio to Bush."

They're also felons, having been convicted of felony theft, planting backdoors in their software, and having been involved bribery and kickback schemes.

Yeah, nothing to worry about here.

and why did Republicans prevent the issue of securing a verifiable paper trail on machines from getting to the floor of Congress for a vote. I can't find any other explanation for that except that they intended to commit fraud.

What are they hiding?????

How to hide vote stealing code? Put it in the compiler. Have the compiler watch for a particular sequece of code, and patch in new code when it sees it. A well known technique, first proposed by Dennis Ritchie in the dark ages of computing, and a technique that has actually been seen in the wild.

Curiously, all the guys with criminal records in the Diebold software team seem to work in the compiler group. And you have to ask... why in God's name would anyone write a custom compiler these days. Compilers are cheap and easy to come by. And if you did write a compiler, why would you write a Basic compiler? Basic is neither an easy language to write a parser for, nor a desirable language from a programming point of view. There's something fishy there.

if you start to question ALL the computers involved in the counting process, the Big Guns will be trained on you.

We have heard these reports several times from various states:

DREs programmed to default to Bush, not allowing straight party Dem tickets, causing undervotes for Pres, etc. If this causes a high enough error rate, and there's no paper trail, you don't have to hack anything. You just let the voters make mistakes and in a close election, Bush wins. You don't have to do it by name. You can do it by incumbant, party, ballot order, or any other parameter.

Who had this experience? Post it here!

I still can't imagine that there is any obvious default setting, but reputable sources such as Chuck Herrin in NC have said it is so, and it's not even hidden. It's up to the voter to erase this choice if he/she doesn't like it. But of course, who wouldn't like to see Shrub get a second term, right? ;)

Oh, and what about hacking into the machines via modem? All you need is the phone number and the passwords.