The prime example I usually bring up is the interstate commerce clause. Congress used it to justify writing laws to promote civil rights, beyond the powers granted to them by the Constitution (specifically, the 10th amendment in the Bill of Rights). I'm a big fan of the motivation, but I've always questioned the means...
But... it could be just the right size that it becomes the mass of a 747 right before it collides with Pluto. Of course, then we also have all the radiation issues to worry about, too.
Well, since a 747 is significantly less massive than a black hole (except for very few - if any - primordial black holes, and even then, after swallowing Pluto they'd definitely be more massive), and BBs are significantly bigger than most particles, I'd say trying to deflect a 747 with BBs would actually be much, much easier. Assuming you have the correct angle, of course...:P
(Never mind the fact that by the time said black hole swallowed up Pluto it'd already have totally destroyed our orbital trajectory.)
IIRC, they drank something at Tom's (or maybe it was with the Ents) that made them grow taller, which factored in to the Scouring of the Shire. Of course, without the Scouring, there really was no need for Tom. I, for one, didn't really care for all of Tom's singing, anyway.
But I do agree with you in principle, especially since most major religions teach tolerance (including Christianity... for 100 points, what did Jesus say the greatest commandment was?)
To "Love the Lord your God with all your heart and with all your soul and with all your mind." I'm guessing that you were thinking of #2: "Love your neighbor as yourself." (Cf. Matthew 22:36-40)
As for the whole "choice" question, doesn't that presume the existence of free will?
If the false positive rate is around 200%, and if the false negative rate is 50%, then any given search might tend to yield 2-4 suspects (depending on the variance of the false positives, of course) and about half the time the actual culprit will be included. Of the 2-4 suspects that are not the culprit, most of the time it will be quite easy to eliminate them, either visually by the user, or from simple detective work (i.e., a rock-solid alibi or possibly even common sense in some cases). So, for a fairly low cost, you've got an additional lead. Having a high false positive rate means a lower false negative rate with the added benefit of having the user get used to the fact that the computer is quite fallible.
I'm not arguing for or against the ethics, just the efficiency.
I'm pretty sure that the "blind" angle for our machines was primarily that - just an angle. As to whether or not the losing party can confirm they lost, it depends on what state they were in. Here in Virginia (where Webb beat Allen by 0.3% of the vote), there is no paper trail, so they cannot. Other states have saw the light, however, and have a verifiable paper trail. I really don't have a problem with electronic machines if they have a voter verifiable paper trail.
I obviously can't read an entire act in five minutes, but are you saying that the Help America Vote Act requires the use of electronic voter machine? That's quite unbelievable.
Here in the Australian state of Victoria, electronic voting machines will be in use for the first time at the upcoming election later this month. The reason for their introduction is not to asist with counting, but to allow the blind to cast a secret ballot for the first time.... The use of the machines is entirely voluntary and I expect that few people other than those who actually need to use one will.
Just like with Victoria, the use of electronic voting is not mandated. The availability, however, is. I.e., that's the reason that a town of 80 must have electronic voting machines. Once you have the machines, those in charge are naturally going to encourage their use as "machines don't make mistakes" or "machines are impartial" or some other nonsense. I'll be keen to hear how your expectations play out (i.e., "that few people other than those who actually need to use one will"). Seriously, I will be keen to find out if you're right. If you are right, I (and presumably the majority of slashdot) will be pleasantly surprised. (I can't stress this enough - I am NOT being sarcastic here.)
1) Absolutely. Which is why this case might be important. It's an excellent "test" case.
2) I find it pretty cool. Unfortunately, we're no ways near as well funded as Blue Brain. Most of our funding comes from the NIH. In fact, I'm working on another grant (an SBIR) to NIH right now. What I do like is having about 100 (quality) CPUs at my disposal when I choose to launch a genetic algorithm spanning lots of possible "brain" configurations.
I have a great deal of experience with large batch and MPI clustering; most of it running some variant of linux, but also SGIs and Suns. I have never once seen a very large job come back without some sort of error. Individual nodes crash during compute, memory (even good ECC RAM) flips bits on occasion, but most of the time the error is just not discernible.
I, too, use MPI systems with PBS (and other) queues. OK, first of all, a confession: I do see (frequent) errors on these systems. However, the errors are ALWAYS of the catastrophic type for me. I.e., the program either fails to complete, or it gives me the correct results. (Caveat: all I can actually claim is that when it does complete, the results are always reasonable, and the few times I do check them against the known correct answer the results are always correct - UNLESS there's an error in the code that I've written.) Now, in an election with tens of thousands (I'm guessing) machines, to claim that catastrophic errors are a given is not too big of a claim. However, in most cases recovery from such errors should be possible IF a paper trail is being used. I'm assuming that no catastrophic error happened in this case, or else it would have been reported.
Secondly, if you have NEVER seen a very large job come back without some sort of error, then something is seriously wrong with your system. The majority of time I run jobs on the MPI system they come back without an error - even when I'm running a simulation with 100,000 neurons, a billion synapses, and trillions of synaptic events.
and, of course, the most recent problem of certain voting machines that lack verifiable paper trails (not a problem in this case, thankfully).
Are you sure? The article does not mention a paper trail. It mentions the alternative of paper ballots, and that they were going to open it up to verify the totals, but it did not mention paper trails.
Anyway, I'm gonna go eat my dinner. Thanks for the discussion. Glad it didn't degenerate into a pissed off flame war.:)
Ditto. Sometimes, it's hard to fight the temptation to be nasty, especially if you haven't eaten dinner yet.;)
I've taken several courses in thermodynamics and several courses in statistical mechanics. After all, I also have a degree in astrophysics, and I can honestly tell you that this isn't rocket science, either.;) Simply put, I understand statistical variance quite well. However, deterministic processes are supposed to be deterministic. Sure, it's possible that a bit in a computer will randomly flip (hence the reference to neutrinos, which are usually the humorously blamed party), but if bits flipped that often, then I'd NEVER be able to get the same results in my program that requires more than a billion synaptic events. (Have I not mentioned this program before?)
Let me put it to you this way: if I drop an egg a million times, will you argue that statistics mandates that there's no way the egg will fall every time?
There are several reasons for this. One is just plain probability. The greater the numbers the more likely for error. This is a purely theoretical argument about physical systems and margin of error and does not take into account the specifics of voting systems.
Are you taking a page from the BOFH and blaming neutrinos or something? Voting should be a deterministic process. Probability does not enter into the equation. Again, my program deals with far more events than this system, and I can (and do) easily verify that it has 0 errors out of billions of synaptic events.
When speaking to the US voting system, the situation gets far more complex. There is a complex interaction between individual voters and the interface used to select candidates.
If you're referring to the possibility that the "error resides between seat and keyboard", then I'll admit that's always a possibility, especially considering the butterfly ballot ballyhoo. It's highly unlikely that's the case here, especially considering that 8 or 9 other people have also said they voted for him.
Next, there is the interaction between anywhere from a few to potentially millions of voting machines deployed across disparate geographic boundaries. All these systems must then interlink to central tabulators within each state.
There should be NO interaction between these voting machines. That just opens the door to hackers. No, the ONLY interaction should be between the voting machine and its parent tabulator. My code has similar interactions, when it is running in parallel, which it must do to pass the verification tests. Again, a single error in this situation is unacceptable.
For national elections the Associated Press then handles centralized tabulation for nonofficial election results reported to the press.
Which is not the case here, so even if errors are acceptable there (and they're not), that has no bearing on this case.
TFA doesn't mention any paper, it just says "that a court order would have to be obtained in order to open the machine and check the totals". Hopefully, that means the machine had a paper trail that each voter could see before casting his/her vote. Hopefully, but not likely. Maybe there's an internal paper trail. Maybe not.
I'm curious as to why a town of 80 people needs to be using electronic voting? It wouldn't be a particularly onerous task to hand count <= 80 paper votes.
First of all, how do you know there's an internal paper trail? TFA does not mention it. All it says is "that a court order would have to be obtained in order to open the machine and check the totals". Secondly, if that paper trail was not visible to each person as he/she cast his/her vote, then it cannot disprove fraud or error.
As others have stated, when there's such an obvious error in such a small election, under what grounds can you believe that larger, more significant errors don't exist in larger elections, especially when statisticians have made such claims? We need verifiable paper trails.
I have a "quick test" that simulates billions of synaptic events. This "quick test" must be passed every time before I check in any changes to the program running the quick test. I am not satisfied if a single synaptic event is missed, unless I understand the reason that one (out of over a billion) synaptic event is different.
Barring fraud, why do assert that "Voting machines (or processes) will never reliably count one hundred million votes"? It's not exactly brain surgery (pun intended).
I searched on 'mirror cost "four square feet" solar' and the Google summary for the second hit showed a cost of $5/sq. foot. That's an upper bound, as I'm sure economy of scale would kick in. So, for 2.5 square miles or 70 million square feet, that's about $350 million dollars. As others have said, not bad for a 1GW power plant.
I tried posting a reply to his comment (just to see if it had been fixed), but apparently it has not. Judging by the fact that there are no replies on this thread, I'm guessing it's become somewhat universal.
It's also worth pointing out that even if this did have something to do with brain size, elephants have larger brains than humans. Sure, you can make lots of good arguments about how 20% of your metabolism (or something like that) is devoted to powering your supercooled cpus, but the point is that brain size is an imperfect indicator of intelligence. Also, men have larger brains than women - but smaller brains proportional to mass. And if you find these arguments unconvincing, then I'll just ask you to trust me - I'm an elephant and I've got a larger brain than you do.
(If this looks like it's a duplicate comment it's because the previous comment was orphaned and appears to have disappeared.)
Slashdot Mirror
The prime example I usually bring up is the interstate commerce clause. Congress used it to justify writing laws to promote civil rights, beyond the powers granted to them by the Constitution (specifically, the 10th amendment in the Bill of Rights). I'm a big fan of the motivation, but I've always questioned the means...
But... it could be just the right size that it becomes the mass of a 747 right before it collides with Pluto. Of course, then we also have all the radiation issues to worry about, too.
Well, since a 747 is significantly less massive than a black hole (except for very few - if any - primordial black holes, and even then, after swallowing Pluto they'd definitely be more massive), and BBs are significantly bigger than most particles, I'd say trying to deflect a 747 with BBs would actually be much, much easier. Assuming you have the correct angle, of course... :P
(Never mind the fact that by the time said black hole swallowed up Pluto it'd already have totally destroyed our orbital trajectory.)
IIRC, they drank something at Tom's (or maybe it was with the Ents) that made them grow taller, which factored in to the Scouring of the Shire. Of course, without the Scouring, there really was no need for Tom. I, for one, didn't really care for all of Tom's singing, anyway.
To "Love the Lord your God with all your heart and with all your soul and with all your mind." I'm guessing that you were thinking of #2: "Love your neighbor as yourself." (Cf. Matthew 22:36-40)
As for the whole "choice" question, doesn't that presume the existence of free will?
That's hilarious. I need to know is that a "sanman2" original, or are you quoting someone else?
If the false positive rate is around 200%, and if the false negative rate is 50%, then any given search might tend to yield 2-4 suspects (depending on the variance of the false positives, of course) and about half the time the actual culprit will be included. Of the 2-4 suspects that are not the culprit, most of the time it will be quite easy to eliminate them, either visually by the user, or from simple detective work (i.e., a rock-solid alibi or possibly even common sense in some cases). So, for a fairly low cost, you've got an additional lead. Having a high false positive rate means a lower false negative rate with the added benefit of having the user get used to the fact that the computer is quite fallible.
I'm not arguing for or against the ethics, just the efficiency.
But now I actually have to see this movie. Something that bad has to be good, or something like that.
I'm pretty sure that the "blind" angle for our machines was primarily that - just an angle. As to whether or not the losing party can confirm they lost, it depends on what state they were in. Here in Virginia (where Webb beat Allen by 0.3% of the vote), there is no paper trail, so they cannot. Other states have saw the light, however, and have a verifiable paper trail. I really don't have a problem with electronic machines if they have a voter verifiable paper trail.
Just like with Victoria, the use of electronic voting is not mandated. The availability, however, is. I.e., that's the reason that a town of 80 must have electronic voting machines. Once you have the machines, those in charge are naturally going to encourage their use as "machines don't make mistakes" or "machines are impartial" or some other nonsense. I'll be keen to hear how your expectations play out (i.e., "that few people other than those who actually need to use one will"). Seriously, I will be keen to find out if you're right. If you are right, I (and presumably the majority of slashdot) will be pleasantly surprised. (I can't stress this enough - I am NOT being sarcastic here.)
1) Absolutely. Which is why this case might be important. It's an excellent "test" case.
2) I find it pretty cool. Unfortunately, we're no ways near as well funded as Blue Brain. Most of our funding comes from the NIH. In fact, I'm working on another grant (an SBIR) to NIH right now. What I do like is having about 100 (quality) CPUs at my disposal when I choose to launch a genetic algorithm spanning lots of possible "brain" configurations.
I, too, use MPI systems with PBS (and other) queues. OK, first of all, a confession: I do see (frequent) errors on these systems. However, the errors are ALWAYS of the catastrophic type for me. I.e., the program either fails to complete, or it gives me the correct results. (Caveat: all I can actually claim is that when it does complete, the results are always reasonable, and the few times I do check them against the known correct answer the results are always correct - UNLESS there's an error in the code that I've written.) Now, in an election with tens of thousands (I'm guessing) machines, to claim that catastrophic errors are a given is not too big of a claim. However, in most cases recovery from such errors should be possible IF a paper trail is being used. I'm assuming that no catastrophic error happened in this case, or else it would have been reported.
Secondly, if you have NEVER seen a very large job come back without some sort of error, then something is seriously wrong with your system. The majority of time I run jobs on the MPI system they come back without an error - even when I'm running a simulation with 100,000 neurons, a billion synapses, and trillions of synaptic events.
Are you sure? The article does not mention a paper trail. It mentions the alternative of paper ballots, and that they were going to open it up to verify the totals, but it did not mention paper trails.
Ditto. Sometimes, it's hard to fight the temptation to be nasty, especially if you haven't eaten dinner yet. ;)
I've taken several courses in thermodynamics and several courses in statistical mechanics. After all, I also have a degree in astrophysics, and I can honestly tell you that this isn't rocket science, either. ;) Simply put, I understand statistical variance quite well. However, deterministic processes are supposed to be deterministic. Sure, it's possible that a bit in a computer will randomly flip (hence the reference to neutrinos, which are usually the humorously blamed party), but if bits flipped that often, then I'd NEVER be able to get the same results in my program that requires more than a billion synaptic events. (Have I not mentioned this program before?)
Let me put it to you this way: if I drop an egg a million times, will you argue that statistics mandates that there's no way the egg will fall every time?
Are you taking a page from the BOFH and blaming neutrinos or something? Voting should be a deterministic process. Probability does not enter into the equation. Again, my program deals with far more events than this system, and I can (and do) easily verify that it has 0 errors out of billions of synaptic events.
If you're referring to the possibility that the "error resides between seat and keyboard", then I'll admit that's always a possibility, especially considering the butterfly ballot ballyhoo. It's highly unlikely that's the case here, especially considering that 8 or 9 other people have also said they voted for him.
There should be NO interaction between these voting machines. That just opens the door to hackers. No, the ONLY interaction should be between the voting machine and its parent tabulator. My code has similar interactions, when it is running in parallel, which it must do to pass the verification tests. Again, a single error in this situation is unacceptable.
Which is not the case here, so even if errors are acceptable there (and they're not), that has no bearing on this case.
If you've never read ShrinkLits, you're cheating yourself.
TFA doesn't mention any paper, it just says "that a court order would have to be obtained in order to open the machine and check the totals". Hopefully, that means the machine had a paper trail that each voter could see before casting his/her vote. Hopefully, but not likely. Maybe there's an internal paper trail. Maybe not.
First of all, how do you know there's an internal paper trail? TFA does not mention it. All it says is "that a court order would have to be obtained in order to open the machine and check the totals". Secondly, if that paper trail was not visible to each person as he/she cast his/her vote, then it cannot disprove fraud or error.
As others have stated, when there's such an obvious error in such a small election, under what grounds can you believe that larger, more significant errors don't exist in larger elections, especially when statisticians have made such claims? We need verifiable paper trails.
I have a "quick test" that simulates billions of synaptic events. This "quick test" must be passed every time before I check in any changes to the program running the quick test. I am not satisfied if a single synaptic event is missed, unless I understand the reason that one (out of over a billion) synaptic event is different.
Barring fraud, why do assert that "Voting machines (or processes) will never reliably count one hundred million votes"? It's not exactly brain surgery (pun intended).
I searched on 'mirror cost "four square feet" solar' and the Google summary for the second hit showed a cost of $5/sq. foot. That's an upper bound, as I'm sure economy of scale would kick in. So, for 2.5 square miles or 70 million square feet, that's about $350 million dollars. As others have said, not bad for a 1GW power plant.
For those who are curious, CUDA stands for "compute unified device architecture".
But I did get a message that you posted a reply. Oh, and I can read your reply - which you'd probably guess if you're reading THIS reply.
I tried posting a reply to his comment (just to see if it had been fixed), but apparently it has not. Judging by the fact that there are no replies on this thread, I'm guessing it's become somewhat universal.
Let's see if this comment actually sticks. Perhaps it's been fixed now? (I have no idea why you were marked "Troll", either.)
It's also worth pointing out that even if this did have something to do with brain size, elephants have larger brains than humans. Sure, you can make lots of good arguments about how 20% of your metabolism (or something like that) is devoted to powering your supercooled cpus, but the point is that brain size is an imperfect indicator of intelligence. Also, men have larger brains than women - but smaller brains proportional to mass. And if you find these arguments unconvincing, then I'll just ask you to trust me - I'm an elephant and I've got a larger brain than you do.
(If this looks like it's a duplicate comment it's because the previous comment was orphaned and appears to have disappeared.)