I agree that trying to come up with a good engineering solution to the electronic / cryptographic voting problem is a worthwhile endeavor - primarily just so that everyone has a clear understanding of what a system would look like and what properties it actually ends up having.
But anyone who does that should also go through the effort of optimizing a classical paper ballot / ballot box / hand count system for comparison. You should be able to produce a system with no trusted parties, perfect technical transparency, a very low cost per voter, and most of the useful properties that you can get out of an electronic system too. If you're going to say that crappy DREs aren't representative of electronic voting machines then butterfly ballots aren't representative of paper.
Personally, I tend to think that the public won't see beyond "paper ballots vs. voting computers" - and that an arbitrarily chosen paper ballot / hand counted system is likely to be better than an arbitrarily chosen machine-assisted system. But even if that's true, it doesn't reduce the benefit of scholarship in trying to produce the best system possible.
As an aside, here's another point to consider about the system you suggest vs. an engineered paper system:
Your system requires every voter to buy themselves a voting device. My guess would be that it'd be damn hard to get these devices much cheaper than $50 at retail - graphing calculators cost $100 and they don't need to do any networking. With a thousand voters at a precinct, that's $50,000.
Consider what you could do with that much money for a paper ballot election: if a certified public accountant costs $100/hour and there are seven parties in the election (in the US, say Democrat, Republican, Green, Libertarian, Constitution, Socialist, Reform) - you could give each party the money to hire 4 CPAs they trust (that's 4 CPAs per political party for every thousand voters) to spend 2 days counting the votes and auditing the election.
Seriously? That high on the list? If it's that high on the list, my guess is that most of the entries are really generic. They probably have items like "transmissible disease", "accident", and "wild animal" rather than "rabid weasel", "choked on condom during oral sex", and "bled to death after scratching genital warts" - because I bet those last three are all more likely causes of death in the western world than terrorism.
Well, not to get all nerd, but it was a reference to Cpt Kirk and Star Trek II.
Which was then re-referenced throughout the franchise. And then referenced by fans of the show commonly enough that I'd expect some people to pick up on it without ever having seen any Star Trek. Sure, it's a moderately obscure reference - but it's not something that you'd have to be an obsessive Star Trek fan to recognize.
And for most people (including myself) Star Trek II is junk culture.
That's just boorish snobbery (which is one of the most annoying forms of nerdiness, btw). There are no worthwhile cultural works that someone won't dismiss offhand as "junk culture" or similar. Sure, Star Trek isn't any sort of high art - but ignoring its cultural impact is foolish. Cultural literacy isn't about whether you've read the works of Shakespeare or whatever (unless you're talking about the English professor subculture), it's about sharing cultural experiences with other actual people.
Never attribute to malice that which is adequately explained by stupidity.
That's a decent first approximation, but it's not good enough if actual maliciousness is reasonably likely. Any competent person doing something malicious will make damn sure it looks like an honest stupid mistake.
i feel like these negative comments i get are in response to some really crappy protocol that i'm not talking about. please actually see what it is that i'm talking about before you poo-poo it.
My criticisms apply to all the cryptographic protocols that I've seen and to any of the straightforward variants that haven't been explicitly described - including the systems described in the PDF and video that you linked to.
the thing that makes this voting system great is the public list of all the votes. most attacks are hard against this.
This means one of two things. You might be giving up secret ballots - which keeps getting suggested but is still a bad idea that a sane population won't accept. The other possibility is that you have some complex cryptographic verification protocol that allows you to prove to yourself that your vote was properly counted but not prove to anyone else what your vote was. The latter schemes are neat, but no-one will actually uses them because they're too complicated.
now i propose that each voter needs some sort of 'voting device' that they buy at the store. these devices must be available from multiple vendors, and all information needed to make one from scratch must be available. (the protocol must be open)
That's easy to say, but the realities of implementing it would be an utter mess and the properties that it needs would be utterly destroyed by even a single political compromise.
while you complain alot, you present no way to make this perfectly trustworthy process possible. here's why: it's REALLY FUCKING HARD!
Here's where we disagree. You merely think it's really hard. I think it's impossible among a population non-experts and have given up on "perfectly trustworthy".
paper ballots systems have huge security problems.
Sure. But the errors with a well designed paper ballot system are almost never systematic - they're small random errors at the precinct level or, occasionally, really blatant fraud that will get caught by any impartial election observer (and still only effect a single precinct if not caught).
--
We're really talking at cross purposes here. My givens are as follows:
The paper ballot system has certain properties that are essential to democracy - those properties must be preserved in any new system. These qualities include technical transparency and secret ballots.
Any new system must be not just technically possible but practically and politically possible. I'm not really convinced that the general public or voting officials could tell the difference between Diebold DRE machines, the system your proposing, and some random other system - and they're the ones who will be making the decision. With paper ballots there's no chance of an election official following the logic "fringd's system is good, it uses electronic devices, so I need to get electronic devices, I'll get DRE machines from a well known vendor".
Now it very well may be that we can do better than the traditional paper ballot / voting booth / ballot box / hand count publicly / archive ballots protocol - but restricting ourselves to transistors that implement complex mathematics isn't a good idea since any such system throws away technical transparency.
Git is the belt fed, compressed air, master carpenters nail gun; it's really, really cool, but is entirely overkill for my needs and does not fit my tool belt at all. Furthermore, it comes with a heavy manual that I'd better read if I'd like to keep all my fingers!
I don't know about that. I've been using git on my latest single-developer project just as a way to keep a log of changes and roll back stupid mistakes - basically the same stuff you're using SVN for.
Git has one huge advantage over SVN - it doesn't need an external repository to commit to because every checkout is also a complete repository. This does mean that "committing" and "creating an off-machine backup" are different operations. Also, the "IDE" I use is vim in a terminal window, so integration isn't an issue at all.
In conclusion, you're right - Git is a different tool from SVN, but you seem to have overrated how different and imagined a massive complexity difference that isn't there.
But a detail from a 40 years old TV show/25 year old movie, that most people have forgotten?
That's factually incorrect. It's a detail from a 40 year old TV setting that was last referenced in a new episode seven or eight years ago. And yes, all of the Star Trek offshoots have still been "popular TV shows", in spite of the fact that both fans and anti-nerds rip on them.
The vast majority of the world won't get the reference - it's sub-culture literacy, at most.
The vast majority of the world wouldn't get *any* single literary reference. That doesn't mean that literate people won't pick up literary references from their own culture. For Star Trek, that culture is pretty wide: people who have TVs and watch American shows.
No. Recognizing fictional references is an example of "cultural literacy". When the reference is a popular TV show, it's more like "basic cultural literacy".
My mom might not understand it, but she trusts me, and I understand it, and that should be enough.
No, it shouldn't. If we're going to claim to have a democracy, all the voters should be equally able to personally understand that their vote is trustworthy. Having only a small elite that understands the voting system and requiring everyone else to accept their word simply isn't democratic (and I say that even when I personally am a member of the proposed "small elite").
And it's a very small elite who are actually qualified to evaluate a cryptographic protocol, or even to evaluate if it was properly followed by election officials. You'd be lucky to get one per polling station even if every qualified person volunteered - and that means that just one defector wins attackers a whole polling station worth of votes (if you think that your protocol stops that sort of attack, you explain the protocol and I'll explain the "ignorant user, malicious expert" attack it doesn't stop - there will always be one).
Maybe not, but maybe trusting me, and my cousin, and all of the newspapers, and all of the political parties, and anybody else, anywhere in the world who feels like it and knows enough, maybe that is enough.
Not everyone personally knows and trusts even one computer security expert, not every computer security expert will take the time to audit the election procedures, and very few of the people who claim to be computer security experts are actually competent to judge a cryptographic voting system.
As for newspapers, are you crazy? Have you seen the coverage of DRE voting problems? The problems with a cryptographic voting system will be even harder to explain to a newspaper audience. Even the reporters won't be able to understand them. We will *always* get the headline "Computer Security Researcher Claims Voting System Not Secure, Voting Official Denies Claim" whether the system is secure or not - and the average newspaper reader won't be able to tell the difference.
You see, there will by many independent verifiers with cryptovoting, and even if I don't understand exactly how to verify it, it is extremely unlikely that every single independent verifier is lying to me.
It might be possible to create a voting system that acts that way, and it might even have useful security properties in practice. On the other hand, it might not. But in any case, a "cypherpunk voting system" isn't going to get implemented - and even if it was, no one but cypherpunks will actually validate it and if they say it doesn't validate no-one will listen to them.
Cryptographic protocols that are secure tend to break when *anything* goes wrong. Voting systems are executed by millions of people, and millions of people make some mistakes. That means that any useful cryptographic assurance will be more complicated than "tampered / not tampered" - and therefore we'll be right back to "experts" arguing in newspaper articles reguardless of the validity of the election.
you must be the privileged few who have the political power to do the recount. there is limited access, which makes it much more opaque.
Maybe we should make our paper ballot system a bit more transparent. Sounds good to me. I'd personally be happy to go to my local voting station and help scan all the paper ballots into a computer as they get counted so they can be posted to the town website or whatever.
please read more about cryptovoting
I've got a reasonable crypto background, and I've been watching the discussion about using crypto in voting for years.
Cryptography is great stuff. There is a cryptographic voting protocol where a group of mathematicians could get together with scratch paper and PDAs and vote in an election with all the properties we would want in an election, and such that any one of them could detect fraud in the protocol eve
But if that issue was fixed? Then, hell, yeah, I'd be willing to bet that the vast majority of Americans could audit an encrypted, authenticated and verified e-voting system.
Absolutely. If math education were fixed to the point where most people could understand RSA and catch the flaw in an intentionally fraudulent 99% error margin calculation then we'd have a radically different situation on our hands.
Unfortunately, for the moment, we're stuck with the voting population that we have - utter incompetence at math and all.
The only thing that Microsoft ever did was take existing trends, embrace and extend, and make a lot of money in the process. If Microsoft had never existed the IBM PC would have run CP/M. The only functionality we might not have is the ability to embed Lotus 123 spreadsheets in our Word Perfect documents - and GNU+Linux would probably have trashed OS2 all over the place in the late 90's.
If the election was close enough that margin of error could come into play, then count both of them and compare the results.
The margin of error for a black-box computer voting system is 100%. As long as we accept that fact, your system works fine - although we can optimize away the electronic count step.
The concept can - in principle - be implemented as well as, or better than, alternatives.
It can, and then the very small fraction of the population that is capable of understanding the security properties of cryptographic protocols will be convinced that the election was legit if they personally act as election observers (through the audit mechanism included in this well-designed e-voting system).
There's a problem though: One of the properties that any voting system should have is that *all voters* should be able to understand that the election was legitimate. Any voter should be able to act as an election observer or auditor. This simple requirement immediately eliminates any sort of DRE voting system.
Even the best-practice that most "voting experts" suggest, optical scan with statistical sampling, isn't good enough because an arbitrarily selected observer can't follow the statistics. Hell, in the 2004 Ohio recount the voting officials couldn't even get the concept of a *random sample* right and most of the people involved didn't realize anything was wrong.
The traditional paper ballot / ballot box / hand count protocol isn't perfect, but it's the only system that's been suggested that meets the "any voter can observe or audit" requirement - and without meeting that requirement, it'd be a stretch to call the resulting system democratic.
That's the biggest problem that I see with PC gaming.
The problem is with your mindset, not with PC gaming.
Crysis looks *beautiful* on medium settings. The fact that it will look even better on new hardware a year from now is an advantage for people who buy that hardware and completely irrelevant to anyone who doesn't. At least for people who don't have some sort of massive jealousy issue that makes it so they can't handle the idea that someone might, at some point in the future, have nicer toys than they do.
Even look at NASA and the DOD. NASA uses COTS parts, whereas the DOD sees nothing wrong with demanding some widget be special made.
Look at the budget difference, and even with their horrible budget NASA still manages to order custom parts when they actually need them.
The claim that having high-budget goal-oriented public organizations is technologically beneficial is an interesting one. The claim that that organization producing bombs and fighter planes is innately better than if they were producing space probes and moon bases isn't interesting - it's absurd. Seriously, blowing up the infrastructure of poor countries so their populations starve is basically a solved problem.
Iceland started out with little science, less engineering, and one of the world's poorest economies in the 1980s.
Iceland is the *perfect case* for geothermal. Perhaps there are specific places in the US that are equally lucky, but Iceland's story is basically irrelevant to the US as a whole in the same way that tidal power is irrelevant to people who live in Kansas.
... Solar also can be built out fast.
My entire post was explaining why solar (at least in the form of photo-voltaic panels) *cannot* be built out fast. You just can't fab silicon that quickly.
And conservation can reduce big chunks... converting commuter highways to rail conveyors (and just using mass transit instead of individual transit).
You might as well be talking about ponies and rainbows. If conservation was going to save the world, it would have happened in the 70's, and rail conveyors (hell, mass transit in general) is about as realistic in the near-future USA as Gravity trains.
You seem to be operating in a "If I were crowned dictator of the world" fantasy. In the real world, what happens is constrained by the existence of other people (the study of which is frequently called "economics"). Government policy can manipulate that to an extent, but even the most drastic of realistic government interventions still take time.
If massive government intervention started now, we could probably start installing some of the technologies you mention in bulk in a decade or two (sooner in lucky cases, but there are less of them than you seem to think). In three or four decades we could start replacing old power plants. By 2100 we could be off fossil fuels and nuclear entirely.
But for the new power plants that *will go in* in the next decade or two, the choice really is nuclear or coal - and modern nuclear is arguably better than most of the solutions that you suggest.
Do you realize how many people have died from mining the nuke fuel?
Ok. Let's find some numbers then. What ground rules do you want to set? All time mining deaths? Coal will be way higher. Deaths after basic safety precautions were understood? Coal will still be higher simply due to volume.
And how many will die when there's an accident (or sabotage) at the waste storage dumps, sometime in the next several hundred/thousand years?
Radiation accidents tend to expose like two people who then have an increased cancer risk when they get old. As industrial risks go, that's pretty boring. And we can make almost your whole issue go away just by using modern fuel recycling techniques.
What happens when they hit a nuke plant?
My bet: The power company will lose millions of dollars to damages on expensive equipment and to the massive PR hit. There will be no deaths, significant radiation exposures, or significant release of radioactive material.
What happens when they just rob one?
The DOW will fall 150 pts as the media go into a frenzy over the "nuclear terrorist" threat. Four hours later the FBI will catch the perpetrators and realize that they just stole two cement blocks. They'll still get 5 years in jail for "making terrorist threats".
Seriously, nuclear power plants aren't dangerous as industrial plants go and they aren't interesting as terror targets aside from the media hysteria that surrounds them. Nuclear power plants aren't bombs, and the radioactive material that they use isn't any more interesting than your average highly poisonous industrial material.
If terrorists ever do execute a plot involving a nuclear power plant, just be thankful that they got distracted by the overhyped target and didn't hit something legitimately dangerous like an urban chemical plant.
NO RADIATION WAS RELEASED. The safety devices functioned as expected and safed the plant. There was no radiation release outside of the containmnt vessel. The other reactor at the 3MI site is still in functional use today generating nearly 1GWe.
Not quite. The best information is that there was a very small radiation exposure for anyone within a mile or so - less than being in a doctor's waiting room when someone gets an X-Ray.
The anti-nuke activists will claim that there was an unknown amount of radiation released because the detectors were tampered with, but given that there's no evidence that there was a dangerous radiation dose even for the personnel on site arguing any further than that seems silly - if it was a billion times less than a chest X-ray or a million times less really doesn't matter to anyone but the technicians involved.
How unfortunate we are that we can choose between only nukes and coal.
What other option do we have the industrial capacity to build out to keep up with the expected increase in demand over the next couple decades?
It's nice to imagine that we could solve the problem with photovoltaics, but the entire industry couldn't build enough solar panels to handle the increase in power consumption for *next month* in *10 years*. It comes down to silicon fabrication, which we really don't have the technology to do in volumes relevant to large scale electricity generation. The other "alternative" power generation schemes either have similar problems or they only work in very specific places (geothermal, solar-thermal, tidal, etc).
Even if the US government started tomorrow to massively subsidize every alternative power generation scheme that had any chance of helping, we'd still be building new coal and nuclear plants for two decades. You just can't magic into existance orders of magnitude of industrial capacity overnight - so the choice really is coal or nuclear (or natural gas, which is slightly cleaner and quite a bit stupider than coal).
ALL future reactors will be breeders in advanced countries will be breeders. It is far too expensive for them not to be.
If people were rational, that would be true. Unfortunately, they rarely are. For example, almost all of the new reactors being proposed in the USA today are the same obsolete 1970's PWR reactors that we already have.
If the anti-nuke activists would accept that modern breeding nuclear technology is safe, clean, and sustainable and protest obsolete reactor designs we might get somewhere. As it is, all we're going to get is coal plants and a couple new PWRs (which will turn a bunch more uranium ore into highly radioactive 1%-burned nuclear fuel).
Have a look at the "Nuclear fuel cycle" and you'll see that it is in effect not a cycle at all, but a process in which at the end we have to dig up more and more holes to put the waste material into and hope it stays there to the end of time.
Yea, when you skip the "reprocessing" step in the fuel cycle you do have that problem. When you *don't* skip the reprocessing step, you can end up with an actual fuel cycle where only a small amount of fresh Uranium ore need go in and a moderate amount of mildly radioactive waste is removed on each cycle.
Note that when discussing fuel cycles, obsolete thermal reactors (like the ones in the USA) are basically useless. You need breeder reactors (with fast neutron breeder reactors like the one mentioned in the article being especially nice).
Take a look at the fuel cycle for the Integral Fast Reactor for an example of what I'm talking about.
Birds are the primary predator for most insects (and fish feed on mosquito larvae), and insect populations recover from DDT sprayings much more quickly than birds or fish. And since they can now breed with impunity, as they don't have to worry about predators anymore, those insects will be a much bigger threat to the human population than they were before, and just spraying more DDT doesn't work, since at that point it starts to affect other wildlife (including humans), and the insects build up resistance quickly.
Nice story. If you can back it up with facts (solid references), it'll even be interesting.
Also, banning DDT didn't cause "millions of people" to die, no matter how popular that meme is with the anti-environmentalist crowd. Just repeating something over and over again doesn't make it true.
More than a million people die every year from Malaria.
DDT was a key component in eliminating Malaria in the United States, according to This New Yorker Article.
With poverty that could potentially have prevented some countries from buying DDT and the existence of DDT resistant mosquitoes it's impossible to say how many lives could have been saved by heavier DDT usage for malaria prevention over the past 30 years - but given the two basic facts I stated above it seems likely that the number is, literally, millions.
You don't have to be an "anti-environmentalist" to agree with me, but you probably can't be part of the "ban dihydrogen-monoxide because chemicals are evil"/"cellphones cause cancer because they emit electromagnetic radiation" crowd.
From an evolutionary standpoint we should probably let people with diseases die from those diseases so as to strengthen the gene pool and keep the population in check
Should we? Is the ability to survive Malaria with poor medical care a trait we want to be selecting for? Probably the best genetic trait for that is sickle-cell anemia - which isn't necessarily something I'd want the whole population to have.
As for controlling the population, how does a high death rate impact birth rates? How does it impact wealth and education level in the population? (Since the best birth control is apparently a rich and well educated population)
but population is probably the most important long-term human issue after pollution
Do you have an argument to support this wild claim? The estimations that I've seen (population stabilizes at ~10 billion in 2050) don't really seem like that big a deal.
If you want a real problem to worry about, try this one: People form strongly held opinions on issues without having gone to the effort to understand the questions they're answering at all.
You know what happens if you fuck up the ecosystem? Millions of people die horrible deaths thanks to famine, landslides, brush fires, etc.
Great. Two plans have been proposed. Both result in millions of people dying. Now for the next step - smart people actually have to consider the scenarios in detail, as well as the other possible plans, and figure out which is best. And no, you can't rationally say that the one where the people die of famine is obviously better than the one where they die of dengue fever without actually looking at the details.
Slashdot Mirror
I agree that trying to come up with a good engineering solution to the electronic / cryptographic voting problem is a worthwhile endeavor - primarily just so that everyone has a clear understanding of what a system would look like and what properties it actually ends up having.
But anyone who does that should also go through the effort of optimizing a classical paper ballot / ballot box / hand count system for comparison. You should be able to produce a system with no trusted parties, perfect technical transparency, a very low cost per voter, and most of the useful properties that you can get out of an electronic system too. If you're going to say that crappy DREs aren't representative of electronic voting machines then butterfly ballots aren't representative of paper.
Personally, I tend to think that the public won't see beyond "paper ballots vs. voting computers" - and that an arbitrarily chosen paper ballot / hand counted system is likely to be better than an arbitrarily chosen machine-assisted system. But even if that's true, it doesn't reduce the benefit of scholarship in trying to produce the best system possible.
As an aside, here's another point to consider about the system you suggest vs. an engineered paper system:
Your system requires every voter to buy themselves a voting device. My guess would be that it'd be damn hard to get these devices much cheaper than $50 at retail - graphing calculators cost $100 and they don't need to do any networking. With a thousand voters at a precinct, that's $50,000.
Consider what you could do with that much money for a paper ballot election: if a certified public accountant costs $100/hour and there are seven parties in the election (in the US, say Democrat, Republican, Green, Libertarian, Constitution, Socialist, Reform) - you could give each party the money to hire 4 CPAs they trust (that's 4 CPAs per political party for every thousand voters) to spend 2 days counting the votes and auditing the election.
Seriously? That high on the list? If it's that high on the list, my guess is that most of the entries are really generic. They probably have items like "transmissible disease", "accident", and "wild animal" rather than "rabid weasel", "choked on condom during oral sex", and "bled to death after scratching genital warts" - because I bet those last three are all more likely causes of death in the western world than terrorism.
Which was then re-referenced throughout the franchise. And then referenced by fans of the show commonly enough that I'd expect some people to pick up on it without ever having seen any Star Trek. Sure, it's a moderately obscure reference - but it's not something that you'd have to be an obsessive Star Trek fan to recognize.
That's just boorish snobbery (which is one of the most annoying forms of nerdiness, btw). There are no worthwhile cultural works that someone won't dismiss offhand as "junk culture" or similar. Sure, Star Trek isn't any sort of high art - but ignoring its cultural impact is foolish. Cultural literacy isn't about whether you've read the works of Shakespeare or whatever (unless you're talking about the English professor subculture), it's about sharing cultural experiences with other actual people.
That's a decent first approximation, but it's not good enough if actual maliciousness is reasonably likely. Any competent person doing something malicious will make damn sure it looks like an honest stupid mistake.
My criticisms apply to all the cryptographic protocols that I've seen and to any of the straightforward variants that haven't been explicitly described - including the systems described in the PDF and video that you linked to.
This means one of two things. You might be giving up secret ballots - which keeps getting suggested but is still a bad idea that a sane population won't accept. The other possibility is that you have some complex cryptographic verification protocol that allows you to prove to yourself that your vote was properly counted but not prove to anyone else what your vote was. The latter schemes are neat, but no-one will actually uses them because they're too complicated.
That's easy to say, but the realities of implementing it would be an utter mess and the properties that it needs would be utterly destroyed by even a single political compromise.
Here's where we disagree. You merely think it's really hard. I think it's impossible among a population non-experts and have given up on "perfectly trustworthy".
Sure. But the errors with a well designed paper ballot system are almost never systematic - they're small random errors at the precinct level or, occasionally, really blatant fraud that will get caught by any impartial election observer (and still only effect a single precinct if not caught).
--
We're really talking at cross purposes here. My givens are as follows:
Now it very well may be that we can do better than the traditional paper ballot / voting booth / ballot box / hand count publicly / archive ballots protocol - but restricting ourselves to transistors that implement complex mathematics isn't a good idea since any such system throws away technical transparency.
I don't know about that. I've been using git on my latest single-developer project just as a way to keep a log of changes and roll back stupid mistakes - basically the same stuff you're using SVN for.
Git has one huge advantage over SVN - it doesn't need an external repository to commit to because every checkout is also a complete repository. This does mean that "committing" and "creating an off-machine backup" are different operations. Also, the "IDE" I use is vim in a terminal window, so integration isn't an issue at all.
In conclusion, you're right - Git is a different tool from SVN, but you seem to have overrated how different and imagined a massive complexity difference that isn't there.
That's factually incorrect. It's a detail from a 40 year old TV setting that was last referenced in a new episode seven or eight years ago. And yes, all of the Star Trek offshoots have still been "popular TV shows", in spite of the fact that both fans and anti-nerds rip on them.
The vast majority of the world wouldn't get *any* single literary reference. That doesn't mean that literate people won't pick up literary references from their own culture. For Star Trek, that culture is pretty wide: people who have TVs and watch American shows.
No. Recognizing fictional references is an example of "cultural literacy". When the reference is a popular TV show, it's more like "basic cultural literacy".
No, it shouldn't. If we're going to claim to have a democracy, all the voters should be equally able to personally understand that their vote is trustworthy. Having only a small elite that understands the voting system and requiring everyone else to accept their word simply isn't democratic (and I say that even when I personally am a member of the proposed "small elite").
And it's a very small elite who are actually qualified to evaluate a cryptographic protocol, or even to evaluate if it was properly followed by election officials. You'd be lucky to get one per polling station even if every qualified person volunteered - and that means that just one defector wins attackers a whole polling station worth of votes (if you think that your protocol stops that sort of attack, you explain the protocol and I'll explain the "ignorant user, malicious expert" attack it doesn't stop - there will always be one).
Not everyone personally knows and trusts even one computer security expert, not every computer security expert will take the time to audit the election procedures, and very few of the people who claim to be computer security experts are actually competent to judge a cryptographic voting system.
As for newspapers, are you crazy? Have you seen the coverage of DRE voting problems? The problems with a cryptographic voting system will be even harder to explain to a newspaper audience. Even the reporters won't be able to understand them. We will *always* get the headline "Computer Security Researcher Claims Voting System Not Secure, Voting Official Denies Claim" whether the system is secure or not - and the average newspaper reader won't be able to tell the difference.
It might be possible to create a voting system that acts that way, and it might even have useful security properties in practice. On the other hand, it might not. But in any case, a "cypherpunk voting system" isn't going to get implemented - and even if it was, no one but cypherpunks will actually validate it and if they say it doesn't validate no-one will listen to them.
Cryptographic protocols that are secure tend to break when *anything* goes wrong. Voting systems are executed by millions of people, and millions of people make some mistakes. That means that any useful cryptographic assurance will be more complicated than "tampered / not tampered" - and therefore we'll be right back to "experts" arguing in newspaper articles reguardless of the validity of the election.
Maybe we should make our paper ballot system a bit more transparent. Sounds good to me. I'd personally be happy to go to my local voting station and help scan all the paper ballots into a computer as they get counted so they can be posted to the town website or whatever.
I've got a reasonable crypto background, and I've been watching the discussion about using crypto in voting for years.
Cryptography is great stuff. There is a cryptographic voting protocol where a group of mathematicians could get together with scratch paper and PDAs and vote in an election with all the properties we would want in an election, and such that any one of them could detect fraud in the protocol eve
Absolutely. If math education were fixed to the point where most people could understand RSA and catch the flaw in an intentionally fraudulent 99% error margin calculation then we'd have a radically different situation on our hands.
Unfortunately, for the moment, we're stuck with the voting population that we have - utter incompetence at math and all.
Bullshit.
The only thing that Microsoft ever did was take existing trends, embrace and extend, and make a lot of money in the process. If Microsoft had never existed the IBM PC would have run CP/M. The only functionality we might not have is the ability to embed Lotus 123 spreadsheets in our Word Perfect documents - and GNU+Linux would probably have trashed OS2 all over the place in the late 90's.
The margin of error for a black-box computer voting system is 100%. As long as we accept that fact, your system works fine - although we can optimize away the electronic count step.
It can, and then the very small fraction of the population that is capable of understanding the security properties of cryptographic protocols will be convinced that the election was legit if they personally act as election observers (through the audit mechanism included in this well-designed e-voting system).
There's a problem though: One of the properties that any voting system should have is that *all voters* should be able to understand that the election was legitimate. Any voter should be able to act as an election observer or auditor. This simple requirement immediately eliminates any sort of DRE voting system.
Even the best-practice that most "voting experts" suggest, optical scan with statistical sampling, isn't good enough because an arbitrarily selected observer can't follow the statistics. Hell, in the 2004 Ohio recount the voting officials couldn't even get the concept of a *random sample* right and most of the people involved didn't realize anything was wrong.
The traditional paper ballot / ballot box / hand count protocol isn't perfect, but it's the only system that's been suggested that meets the "any voter can observe or audit" requirement - and without meeting that requirement, it'd be a stretch to call the resulting system democratic.
The problem is with your mindset, not with PC gaming.
Crysis looks *beautiful* on medium settings. The fact that it will look even better on new hardware a year from now is an advantage for people who buy that hardware and completely irrelevant to anyone who doesn't. At least for people who don't have some sort of massive jealousy issue that makes it so they can't handle the idea that someone might, at some point in the future, have nicer toys than they do.
Look at the budget difference, and even with their horrible budget NASA still manages to order custom parts when they actually need them.
The claim that having high-budget goal-oriented public organizations is technologically beneficial is an interesting one. The claim that that organization producing bombs and fighter planes is innately better than if they were producing space probes and moon bases isn't interesting - it's absurd. Seriously, blowing up the infrastructure of poor countries so their populations starve is basically a solved problem.
Iceland is the *perfect case* for geothermal. Perhaps there are specific places in the US that are equally lucky, but Iceland's story is basically irrelevant to the US as a whole in the same way that tidal power is irrelevant to people who live in Kansas.
My entire post was explaining why solar (at least in the form of photo-voltaic panels) *cannot* be built out fast. You just can't fab silicon that quickly.
You might as well be talking about ponies and rainbows. If conservation was going to save the world, it would have happened in the 70's, and rail conveyors (hell, mass transit in general) is about as realistic in the near-future USA as Gravity trains.
You seem to be operating in a "If I were crowned dictator of the world" fantasy. In the real world, what happens is constrained by the existence of other people (the study of which is frequently called "economics"). Government policy can manipulate that to an extent, but even the most drastic of realistic government interventions still take time.
If massive government intervention started now, we could probably start installing some of the technologies you mention in bulk in a decade or two (sooner in lucky cases, but there are less of them than you seem to think). In three or four decades we could start replacing old power plants. By 2100 we could be off fossil fuels and nuclear entirely.
But for the new power plants that *will go in* in the next decade or two, the choice really is nuclear or coal - and modern nuclear is arguably better than most of the solutions that you suggest.
Ok. Let's find some numbers then. What ground rules do you want to set? All time mining deaths? Coal will be way higher. Deaths after basic safety precautions were understood? Coal will still be higher simply due to volume.
Radiation accidents tend to expose like two people who then have an increased cancer risk when they get old. As industrial risks go, that's pretty boring. And we can make almost your whole issue go away just by using modern fuel recycling techniques.
My bet: The power company will lose millions of dollars to damages on expensive equipment and to the massive PR hit. There will be no deaths, significant radiation exposures, or significant release of radioactive material.
The DOW will fall 150 pts as the media go into a frenzy over the "nuclear terrorist" threat. Four hours later the FBI will catch the perpetrators and realize that they just stole two cement blocks. They'll still get 5 years in jail for "making terrorist threats".
Seriously, nuclear power plants aren't dangerous as industrial plants go and they aren't interesting as terror targets aside from the media hysteria that surrounds them. Nuclear power plants aren't bombs, and the radioactive material that they use isn't any more interesting than your average highly poisonous industrial material.
If terrorists ever do execute a plot involving a nuclear power plant, just be thankful that they got distracted by the overhyped target and didn't hit something legitimately dangerous like an urban chemical plant.
Not quite. The best information is that there was a very small radiation exposure for anyone within a mile or so - less than being in a doctor's waiting room when someone gets an X-Ray.
The anti-nuke activists will claim that there was an unknown amount of radiation released because the detectors were tampered with, but given that there's no evidence that there was a dangerous radiation dose even for the personnel on site arguing any further than that seems silly - if it was a billion times less than a chest X-ray or a million times less really doesn't matter to anyone but the technicians involved.
What other option do we have the industrial capacity to build out to keep up with the expected increase in demand over the next couple decades?
It's nice to imagine that we could solve the problem with photovoltaics, but the entire industry couldn't build enough solar panels to handle the increase in power consumption for *next month* in *10 years*. It comes down to silicon fabrication, which we really don't have the technology to do in volumes relevant to large scale electricity generation. The other "alternative" power generation schemes either have similar problems or they only work in very specific places (geothermal, solar-thermal, tidal, etc).
Even if the US government started tomorrow to massively subsidize every alternative power generation scheme that had any chance of helping, we'd still be building new coal and nuclear plants for two decades. You just can't magic into existance orders of magnitude of industrial capacity overnight - so the choice really is coal or nuclear (or natural gas, which is slightly cleaner and quite a bit stupider than coal).
Do you realize how many deaths would need to have been suppressed to even bring nuclear up to the death count of hydroelectric (much less coal)?
If we built more fast breeding plants like Monju, we could run off the already-mined nuclear material for hundreds of years.
If people were rational, that would be true. Unfortunately, they rarely are. For example, almost all of the new reactors being proposed in the USA today are the same obsolete 1970's PWR reactors that we already have.
If the anti-nuke activists would accept that modern breeding nuclear technology is safe, clean, and sustainable and protest obsolete reactor designs we might get somewhere. As it is, all we're going to get is coal plants and a couple new PWRs (which will turn a bunch more uranium ore into highly radioactive 1%-burned nuclear fuel).
Yea, when you skip the "reprocessing" step in the fuel cycle you do have that problem. When you *don't* skip the reprocessing step, you can end up with an actual fuel cycle where only a small amount of fresh Uranium ore need go in and a moderate amount of mildly radioactive waste is removed on each cycle.
Note that when discussing fuel cycles, obsolete thermal reactors (like the ones in the USA) are basically useless. You need breeder reactors (with fast neutron breeder reactors like the one mentioned in the article being especially nice).
Take a look at the fuel cycle for the Integral Fast Reactor for an example of what I'm talking about.
Nice story. If you can back it up with facts (solid references), it'll even be interesting.
More than a million people die every year from Malaria.
DDT was a key component in eliminating Malaria in the United States, according to This New Yorker Article.
With poverty that could potentially have prevented some countries from buying DDT and the existence of DDT resistant mosquitoes it's impossible to say how many lives could have been saved by heavier DDT usage for malaria prevention over the past 30 years - but given the two basic facts I stated above it seems likely that the number is, literally, millions.
You don't have to be an "anti-environmentalist" to agree with me, but you probably can't be part of the "ban dihydrogen-monoxide because chemicals are evil"/"cellphones cause cancer because they emit electromagnetic radiation" crowd.
Should we? Is the ability to survive Malaria with poor medical care a trait we want to be selecting for? Probably the best genetic trait for that is sickle-cell anemia - which isn't necessarily something I'd want the whole population to have.
As for controlling the population, how does a high death rate impact birth rates? How does it impact wealth and education level in the population? (Since the best birth control is apparently a rich and well educated population)
Do you have an argument to support this wild claim? The estimations that I've seen (population stabilizes at ~10 billion in 2050) don't really seem like that big a deal.
If you want a real problem to worry about, try this one: People form strongly held opinions on issues without having gone to the effort to understand the questions they're answering at all.
Great. Two plans have been proposed. Both result in millions of people dying. Now for the next step - smart people actually have to consider the scenarios in detail, as well as the other possible plans, and figure out which is best. And no, you can't rationally say that the one where the people die of famine is obviously better than the one where they die of dengue fever without actually looking at the details.