The only appropriate technology for voting is raising your hand. Any device such as a ballot that can be locked in a box and carted into a locked room to be counted by strangers is eminently hackable.
The secret ballot is the single greatest threat to verity in elections since we stopped allowing the outright buying of votes.
Ballot boxes can be stuffed. The counting room can be infiltrated by partisans. The entire process can be a sham run by the state's Secretary of State.
And the issues on which you base your votes can be complete bullshit designed to distract you from the true issues on which you should be making your decisions.
Nobody said plural voting wasn't a logical fallacy in the first place. It's just better than letting a guy make the decisions because he killed the last guy who made the decisions, or was the son, grandson, great^N-grandson, etc. of a guy who killed the last guy (or some guy we dug up in another country because he was a fuzzy cousin of the guy who died while making the decisions (bet you didn't know the British Royal Family is actually German)).
I agree, the science of robotics requires electromechanical control and all the motion-control analysis that implies.
But a "robotic arm" is not a robot. It could be a piece of a robot, but it's not a robot. Unless maybe it can decide to strangle its master. Asimov's laws are not actual laws, yet, that I know of, but in this case the need for them can be a test for whether a thing is a robot or not, although that's a rather stronger test, as all I want is autonomous behavior, not necessarily the ability to decide whether or not to kill someone.
Look, if it has to be tethered to an enormous power and control source, it is not a robot, it is a...what? a mechanical peripheral, at best.
I have the same basic problem with the "robot wars" stuff. If there's a guy with a controller box to one side of the arena deciding what the thing will do, it's an RC toy, not a robot.
Make it autonomous, self-propelled, and self-powered, and it's a robot. Otherwise, it's not any more of a robot than your printer is.
It's a tradeoff metric. If you want it to go faster on the freeway you may have to give up some integer-processing performance. A unit with more MIPS/horsepower would let you get more of either or both.
That may actually be done in hardware. And shouldn't be much of the router's processing. I mean, just how many routes are you trying to resolve in a second, anyway?
But if you get the WNDR-4000, first thing you do is update the firmware. The one I got kept toggling the wireless on and off until it snagged a patch. Pretty sure it will phone home the first time you plug it in, but check manually to see.
I don't know about other models, but this one has gotten some bad customer reviews, and NetGear are proactively responding and trying to get those reviewers to update their comments now that things are working cleaner.
Of course, if your goal is to go open-source, none of this is really a problem for you. It's a nice piece of hardware.
Actually, it's the only thing that executive staff could say.
They don't design the products, they just tell people to go and design the products.
And that's a broad but accurate description of the products they design, with some false confidence slathered on top.
I take it back. They could say one other thing: We know we suck at this relative to our competition and we're tired of spending more money than we make and asking our investors and bankers and competitors to cover our losses, so we're going to shut this thing down and let everyone go do something they'll be proud of some day.
But they won't, because honesty is not an absolute in the executive class.
Steven Wolfram apparently doesn't know the first thing about the practical nature of programming, if the thinks that nature can be programmed practically.
In his world, we'll just mod the universe and it will do what we want.
In my world, I change one line of code and I'm sending a special-purpose computer back to the factory to be disassembled and un-bricked.
Someone ask him where the documentation on the Universe is. Because that "reverse engineering" stuff is precisely how we get into these messes.
Today Pfizer announced results of a new study showing that cholesterol has nothing to do with any health problems whatsoever, but water can kill you. Simultaneously they announced the start of trials of a new drug to control this menace, tentatively named hydroprofitor.
Um, this setup is slightly different from his old one, and if you bother to read the text under TFV you'd see that he is, indeed, working on a way to get airborn from the ground.
Agree. Breitling is the shiznit. As are Omega and IWC. Rolex, on the other hand, is shite.
Breitling have been supporting aviation for a very long time. Chronographs for pilots are one of their specialties. Back when a watch was a critical navigation tool they were the watch to have. They returned the favor by sponsoring all sorts of aviation competition.
And really, it doesn't matter who sponsors it. Yves Rossy is the one doing the science on his equipment. His wing could be covered with a giant Orangina logo and it'd still be the single coolest fucking thing in the air.
In the presence of a static magnetic field (as around a conductor carrying a constant current), electrons in the clamp circuit, which also carries a DC current, will be pushed to one side of the clamp conductor, inducing a voltage relative to the other side. Measure the voltage and you know the current in the wire it's clamped around.
Uh-huh. And if I had a 14-inch penis, a billion-dollar record label, and a brain made of Sakrete(R) I could be the next Mr. Kim Kardashian.
And "an interaction with the vibrational states of the crystal lattice" is a thermodynamic randomness, nothing to do with "quantum vaccuum fluctuation" at all.
So what they really have is a 200-Hz, 6-bit, thermal RNG that isn't fully characterized for bias.
"it's straightforward to transform one distribution into another."
Er, no. A distribution is a distribution. If you have more bits in one bin of the histogram, that bin will represent a larger domain in the output side of your converter. Or else you will have to combine bins to create larger bins in the output distribution. Going from more bins to fewer bins means you lose entropy. Losing entropy is not as good as having the distribution you want in the first place and not losing entropy.
So the distribution is important. Especially if "All that really matter is the entropy." [sic]
Slashdot Mirror
The only appropriate technology for voting is raising your hand. Any device such as a ballot that can be locked in a box and carted into a locked room to be counted by strangers is eminently hackable.
The secret ballot is the single greatest threat to verity in elections since we stopped allowing the outright buying of votes.
All voting systems have holes.
Ballot boxes can be stuffed. The counting room can be infiltrated by partisans. The entire process can be a sham run by the state's Secretary of State.
And the issues on which you base your votes can be complete bullshit designed to distract you from the true issues on which you should be making your decisions.
Nobody said plural voting wasn't a logical fallacy in the first place. It's just better than letting a guy make the decisions because he killed the last guy who made the decisions, or was the son, grandson, great^N-grandson, etc. of a guy who killed the last guy (or some guy we dug up in another country because he was a fuzzy cousin of the guy who died while making the decisions (bet you didn't know the British Royal Family is actually German)).
It means they can root your phone so you can install their, um, rootkit, essentially....
Is a power screwdriver a robot, then?
I agree, the science of robotics requires electromechanical control and all the motion-control analysis that implies.
But a "robotic arm" is not a robot. It could be a piece of a robot, but it's not a robot. Unless maybe it can decide to strangle its master. Asimov's laws are not actual laws, yet, that I know of, but in this case the need for them can be a test for whether a thing is a robot or not, although that's a rather stronger test, as all I want is autonomous behavior, not necessarily the ability to decide whether or not to kill someone.
Not only that, it's a fire-bomb, too.
Look, if it has to be tethered to an enormous power and control source, it is not a robot, it is a...what? a mechanical peripheral, at best.
I have the same basic problem with the "robot wars" stuff. If there's a guy with a controller box to one side of the arena deciding what the thing will do, it's an RC toy, not a robot.
Make it autonomous, self-propelled, and self-powered, and it's a robot. Otherwise, it's not any more of a robot than your printer is.
It's a tradeoff metric. If you want it to go faster on the freeway you may have to give up some integer-processing performance. A unit with more MIPS/horsepower would let you get more of either or both.
And now I'm actually highly curious as to what a table of MIPS/horsepower would look like for current production vehicles. Some new cars have upwards of 100 embedded CPUs.
That may actually be done in hardware. And shouldn't be much of the router's processing. I mean, just how many routes are you trying to resolve in a second, anyway?
But if you get the WNDR-4000, first thing you do is update the firmware. The one I got kept toggling the wireless on and off until it snagged a patch. Pretty sure it will phone home the first time you plug it in, but check manually to see.
I don't know about other models, but this one has gotten some bad customer reviews, and NetGear are proactively responding and trying to get those reviewers to update their comments now that things are working cleaner.
Of course, if your goal is to go open-source, none of this is really a problem for you. It's a nice piece of hardware.
i'm glad i didn't get that joke, and not really happy that i'm starting to infer where it comes from
my lack of ignorance is disturbing
China? AMD already sold it to the Arabs. Who apparently aren't sufficiently impressed with its performance to sink a lot of risky capital into it.
So they've got the keeps-non-ruggedized-computer-in-dangerous-location market sewed up.
Everyone else will still be buying Intel for every other possible reason, but AMD's marketing can check this box, at least.
Actually, it's the only thing that executive staff could say.
They don't design the products, they just tell people to go and design the products.
And that's a broad but accurate description of the products they design, with some false confidence slathered on top.
I take it back. They could say one other thing: We know we suck at this relative to our competition and we're tired of spending more money than we make and asking our investors and bankers and competitors to cover our losses, so we're going to shut this thing down and let everyone go do something they'll be proud of some day.
But they won't, because honesty is not an absolute in the executive class.
Steven Wolfram apparently doesn't know the first thing about the practical nature of programming, if the thinks that nature can be programmed practically.
In his world, we'll just mod the universe and it will do what we want.
In my world, I change one line of code and I'm sending a special-purpose computer back to the factory to be disassembled and un-bricked.
Someone ask him where the documentation on the Universe is. Because that "reverse engineering" stuff is precisely how we get into these messes.
Today Pfizer announced results of a new study showing that cholesterol has nothing to do with any health problems whatsoever, but water can kill you. Simultaneously they announced the start of trials of a new drug to control this menace, tentatively named hydroprofitor.
Yeah, but this is video. Totally new technology for making me wet myself over how fucking cool this dude's schtick really is.
Seriously? You've never seen his other videos? Dude's been doing loops and barrel rolls for years. You can't do those without lift.
Um, this setup is slightly different from his old one, and if you bother to read the text under TFV you'd see that he is, indeed, working on a way to get airborn from the ground.
Titanium. Hollow. Filled with awesome at the same pressure as the core of the Sun.
Agree. Breitling is the shiznit. As are Omega and IWC. Rolex, on the other hand, is shite.
Breitling have been supporting aviation for a very long time. Chronographs for pilots are one of their specialties. Back when a watch was a critical navigation tool they were the watch to have. They returned the favor by sponsoring all sorts of aviation competition.
And really, it doesn't matter who sponsors it. Yves Rossy is the one doing the science on his equipment. His wing could be covered with a giant Orangina logo and it'd still be the single coolest fucking thing in the air.
At what rate? And AC or DC? If AC, he could be inducing an earthquake.
Hall effect.
In the presence of a static magnetic field (as around a conductor carrying a constant current), electrons in the clamp circuit, which also carries a DC current, will be pushed to one side of the clamp conductor, inducing a voltage relative to the other side. Measure the voltage and you know the current in the wire it's clamped around.
Wait. Was that part of that movie? I only saw like 8 seconds of it. Yes, you know which 8 seconds.
Uh-huh. And if I had a 14-inch penis, a billion-dollar record label, and a brain made of Sakrete(R) I could be the next Mr. Kim Kardashian.
And "an interaction with the vibrational states of the crystal lattice" is a thermodynamic randomness, nothing to do with "quantum vaccuum fluctuation" at all.
So what they really have is a 200-Hz, 6-bit, thermal RNG that isn't fully characterized for bias.
I call shenanigans.
"it's straightforward to transform one distribution into another."
Er, no. A distribution is a distribution. If you have more bits in one bin of the histogram, that bin will represent a larger domain in the output side of your converter. Or else you will have to combine bins to create larger bins in the output distribution. Going from more bins to fewer bins means you lose entropy. Losing entropy is not as good as having the distribution you want in the first place and not losing entropy.
So the distribution is important. Especially if "All that really matter is the entropy." [sic]