Dead is dead.

Too many points of failure. Too complex a system.

Other malfunctions have happened in other plane systems, one resulting in the plane shaking the passengers around like dice. All three computers received the same input and made the same mistake. The question is: can we understand that we've overcomplexified systems to the point that they are too unstable to use? We made the same mistake with cars and roadways in the past century once; we kept doubling down on the system's complexity as the carnage mounted, and to this day, we think the answer is better cars rather than toss the original solution out and make something simpler. We're addicted to complexity. MAkes more money, for one thing.

Airbus created the first commercial craft that were completely computer controlled. QED: they go boom often, and we hear about them often. It's not the brand, it's the belief that computers are the best and only solution in every system case, voting machines to cars to pacemakers to trucks to planes. Those systems will fail spectacularly because the paradigm is to treat them like PCs, updated frequently to fix endless streams of errors, when they should have been working correctly in the first and only place. Computers are "infinite" machines - they can be operated in an infinite number of ways, and that is really bad news when you are trying to control a simple and finite process. We are over-complexifying systems because we can. Every nail gets the same hammer. Bad engineering and it will fail.The question is whether the computer-addicted generation will be able to understand what the problem is.

An anecdote can be defined as "evidence we don't want to hear about."

Hyper-complex software, sensor arrays, and mechanical systems will fail. They will always fail; humans cannot anticipate all errors, all possible combinations of factors that can cause death and destruction. Humans can't build autonomous complex systems (no, really, they can't. We've barely started making such things) that can't fail. In this case, can't say that a human pilot or a mechanical backup would have made a dfference, but as the world goes forward, gleefully firing truck drivers and converting cars into remote-controllable computer complexes, such things will be so commonplace as not to be worth reporting. Which will feed back our certainty that all is well. It isn't.

RAM prices stayed high long after production costs dropped. It took government action - overseas, I recall- to punish the collusion and price-fixing and force the prices to drop. There ain't no such thing as a free market.

Disney not two weeks ago fired their IT department and replaced them with H1B workers. They just didn't want to pay American salaries to American workers. The arguments is over.

The NSA ain't god, and this is *me* saying this. The NSA is an API for the governments's paranoid would-be supergovernment. They just do what they're told, being Slashdotty geeks, and then they go home. Hacking the evoting machines would be... problematic even for their kind of organization.

Much easier if billionaires just buy the voting machine companies and tell a few trusted IT people to install backdoors on the main accumulation points. Auditing is useless if the original votes are uncountable because of anonymity requirements. We trust the accumulated totals. And if you think the Koch bros. and others like them aren't capable of that, then you ain't paying attention.

Canada DID had (last I heard) a nationwide paper voting system. You draw an "X" with a pencil next to your candidate. They counted each card on a table with both parties looking on (It works if one side is not constantly contesting the reading to slow it down intentionally so their boy can win by running down the clock). They finish the count nationwide in three hours. Works, and you can recount in less than a day, if you have to (if one side isn't trying to delay, delay, delay).

I understand the Harper Government is rapidly shoving e-voting down Canada's throat.

Guess why. Go on, guess. Why replace a system that works and can't be cheated with a series of black boxes that can't be trusted? Why would you possibly want to do that. Golly. I can't. Why would they do that... it's like they *want* a system that can be hacked to win, esp. since their side owns the companies, one way or another. Nah, that can't be right. That would be dishonest, and plays to the public's idea that computers are always awesome and better than people.

If a company has a person who has access to the accumulated counts, then that person can change the vote, downstream, upstream, or final destination. The individual computers don't matter; there are no ways, given the constraints, the verify who voted for what - yes, there are established, effective ways of controlling the data, but if you can't audit the entire chain, it doesn't matter. The code, the counts, the data are owned by private companies, and they have established that all of that are their protected trade secrets. Past lawsuits show us that they refuse court orders and countersue, and that they destroy the data, such as it is. Puttering around with geeky analysis of the browsers and malware isn't addressing the problem which is: they who control the machines control the elections. Worrying about the "hackers" on the outside is specious. It's the hackers on the inside.