Normally, insurance companies aren't allowed to drop coverage or raise rates based on information that becomes available after your coverage starts. (If you have information that you withhold when signing up, that's a different story.)
There's another interesting potential use for genetic disease screening. I don't know if it applies to any known diseases now, but I'm certain it will eventually.
Suppose there's a disease for which there is a genetic test, but also good preventative care options. Ideally, you'd want to take the test, find out that you're at risk, and have the insurance company pay for the prevantative care. In an ideal world, that's what they would want too -- the preventative care would cost them less than paying if you actually did get sick. Both parties would then want the test administered, and want the other party to know the result.
Insurance companies should charge the correct amount, given all knowledge of risk factors -- insurance should be spreading risk among your insurance pool, not about getting free medical care. The problem is that this results in morally repugnant things like denying people care for unlikely conditions they're predisposed to but had no control over.
In an ideal world, this would be an argument for socialized medicine, or at least goverment-run health insurance for such factors. At a societal level, a certain fraction of people will have genetic anomalies, and at a societal level we shouldn't discriminate against them. It's no different from a car accident -- even if you're a good, careful driver, you roll the dice every time you get behind the wheel. The difference is that the dice get rolled (at least in part) long before the healthcare bills start to appear. But, that doesn't make it any less of a "risk" -- it just means that people can attempt to opt out of paying their share after they've checked the dice roll.
The ideal solution would be to cover everyone based on a pool composed of everyone -- aka mandatory government funded insurance (or socialized healthcare, depending on how you want to look at it).
(And to wander further afield -- there are lots of issues with our current healthcare system. I'm skeptical that socialized healthcare is the answer, and I don't know that implementing government insurance for genetic conditions would actually improve anything. At the same time, I find it ethically wrong to allow people to use genetic results they know about in decisions about what healthcare coverage to buy, but not allow the providers to use the same information in decisions about what to charge. It's a difficult and many-faceted problem, and I won't pretend to have even some of the answers.)
Well yes, the cheap unshielded meter didn't belong. What amazed me, though, was how *badly* it didn't belong. It wasn't actually on the same power system as any of the analog stuff, just in the same rack. (The analog system was all powered off various linear supplies from the inverter I mentioned.) The system it was on was a collection of lead-acid batteries, and a large number of valves controlled through relays, commanded by a PLC and control panel at the other end of the cable. It was lighting up LEDs to the point that it interfered with usability. The interference with the analog system was actually relatively minor (though of course we didn't want it). A cheap digital panel meter should be perfectly at home in a circuit whose major components are relays, solenoids, position / pressure sensor switches, and LED indicators, all commanded by a set of mechanical switches and sequenced by an industrial grade PLC (which never showed any signs of caring).
That when the humans and the robots destroy each other in a nuclear war, it will be up to the cockroaches to continue the battle against the robot cockroaches?
Either way you need to do voltage conversion. Transmission grids need to be at high voltage for efficiency (that's why AC was chosen originally -- you can convert it with transformers). These days power electronics have gotten cheap, so you can build an inexpensive DC-DC or DC-AC converter to connect your solar panel to the grid, but either way you do it you'll have to get it to high voltage somehow.
Yep, for ever expanding definitions of "short distances." High voltage high power DC silicon is getting better and cheaper, so we're already seeing a few long-haul DC lines where the reduced radiative losses and increased carrying ability of the cables makes it more efficient. On the other end, DC converters are becoming ubiquitous inside electronics. Google wants to standardize on only one voltage (12V) coming from your computer's PSU, and anything that wants another voltage just has its own converter. All of this will make power distribution and conversion cheaper, smaller, and more efficient. And some places are already moving to distributing DC to the rack-mount cabinets, at (iirc) 48V, and converting DC-DC from there.
On the flip side, I absolutely hate DC converters when trying to do analog work. If you buy a 16-bit analog to digital signal capture board, they run off a switching power supply -- a power supply that generates lots of high frequency noise. I've worked with ones that simply can't do better than about 12-13 bits -- the last 3-4 are basically worthless from the noise. I worked on a project that had need of a voltage monitor on a large battery-powered system involving a large number of sensors and controls. We installed a nice little digital LED voltmeter. It had a tiny, noisy switching converter in it. Not only did it dirty up all our data, it actually kicked back enough noise that it lit up LEDs that should have been electrically isolated from it (capacitative coupling through a long cable). Needless to say, we ripped it out and installed an analog (as in coil, magnet, needle) voltmeter.
Even good (read: expensive, as in $1/W or more) sine wave inverters are a pain. One we worked with produced wonderfully clean AC output -- but also created significant radiated noise, and dirtied up the DC input! Capacitors and ferrite beads helped, but not enough -- we eventually just moved it far away and ran the AC through a long extension cord.
So... DC/DC converters are wonderful for space, efficiency, and cost, but they can be quite the pain to work with if you actually care about having any sort of analog precision anywhere nearby. That sort of noise is amazingly hard to get rid of, and you often can't get rid of all the switching supplies these days.
In the case of a zero-day exploit, 48 hours is an entirely reasonable expectation. As far as I can tell, it is far from the norm, however. For less egregious flaws, feature upgrades, or normal bugs, a slower turnaround would be perfectly reasonable (and what I'd expect). But for a zero day exploit... Well, I firmly believe software companies need to be held more accountable for those by their customers. If that's what your customers are doing, then good for them.
And why would we expect the market to "punish" them? Does anyone actually expect it to cost them sales or other revenue, or increase their costs, or otherwise have a relevant impact on their financial status?
Well, in terraforming terms, finding stuff to make up the Martian atmosphere probably isn't that hard. There are significant CO2 ice caps, and there may be significant water available with modest effort. CO2 plus plants gives you O2. Also, there is some good evidence to suggest that the icecaps' existence is bistable -- that is, if you could mostly evaporate them, the additional greenhouse effect would warm the planet enough to finish the job and keep it that way.
Basically, the problem of terraforming is to find resources that are already available in almost the form you want, and find some way to leverage your input effort. You don't want to have to process every single megaton of atmosphere you want to add / remove. It's far easier to (for example) dust carbon black on the poles and add a few orbiting mirrors.
Of course, the only reference I have handy is Zubrin's The Case for Mars which is a bit dated but (I think) still basically correct. The details may well have changed thanks to newer lander data.
So what? Humanity has shown us repeatedly that such minds exist, and that we know of no way of changing that. Yes, we should absolutely blame the people writing and operating these botnets. But, does that mean we should be giving them as much help as we are? You pose a false dichotomy -- we can most certainly blame the people responsible, while also making it harder for them to do things like this.
This wouldn't be slashdot without a car analogy. You have keys for your car, but clearly they wouldn't be necessary if not for the people who steal cars. Saying you should have keys for your car is like blaming the bullet instead of the killer.
I wonder what would happen if the electric company billed for the volt-amps consumed, instead of the watts, and then reported both numbers (together with your power factor) on the bill. I also wonder what would be required to do whole-house power factor correction? How much cost would it add if you were going to install a grid-tie solar system or something similar? How do these numbers compare to the added cost of power factor corrected power supplies in consumer electronics?
Empirically, they can't. It does not matter why, unless with that answer comes some insight into how to change it. It would appear that simply telling them to do better has no impact. If *you* want to save power, then that method has some hope of success. If a large organization or society wants to save power, that method is almost hopeless. So, given that you can't just tell people to conserve energy and expect it to work, what can you do? Incentives or mandates for more efficient standby modes is one solution that might actually have an impact.
Have you ever tried tossing an RFID tag into the microwave? I've done that with a couple I've gotten in books. Antennas designed to receive microwaves produce an impressive spark show...
Unfortunately, there doesn't appear to be a Debian package, and the downloadable version won't run on my machine. Did they really have to link against a specific version of the libc?
Completely free, no ads, donation supported. Fair warning: a few of the items are easiest to get by donating, but those are nice-to-haves, and it's certainly no big deal if you don't have them.
Technically true, but beside the point. People who talk about fuel cells for cars aren't talking about a closed system. They're talking about distributing fuel and using atmospheric oxygen, and exhausting the waste products. Part of the reason for this, I imagine, is that closed cycle fuel cell system efficiencies are poor compared to batteries and capacitors. So fuel cells should be lumped in with gasoline engines, diesel engines, non-plugin hybrids, etc. in that comparison.
From a public policy standpoint, the interesting questions are system efficiency, and whether the energy is distributed through the electric grid or by carrying fuel around. Unless I'm missing something, these people are talking about the latter.
For now, it's batteries. But in the not too distant future, it may well be supercapacitors. Supercapacitors now are about a factor of ten away from lithium-ion batteries; improvements that are currently in labs appear to be able to remove most or all of that gap. Right now supercapacitors are expensive, but once the market starts growing they should come down in price. There are relatively fundamental limits to how much better traditional batteries can get in terms of capacity, but the apparent limits on supercapacitors are phenomenal. It might be 10 years before they see serious use, but I imagine small-scale use will be here sooner than that, especially if the rumors are to be believed.
Fuel cells are interesting, but I think that direct electrical storage through batteries and later supercapacitors is more likely to actually work out.
Well, 4kHz, because that's the Nyquist frequency for 8kHz sample rate. Unfortuntaley, if you're intending to use the data without doing the dithering work, you need to get rid of the aliasing artifacts -- which, uncoincidentally, are the exact data that would allow the dithering to function.
If it's doing that, then there's a digital AA filter before the resampling for the same purpose as the analog AA filter. It would have the same effect.
The problem with this approach is that audio ADCs have an analog antialiasing filter in front of them. It's not just that you can't see the high frequencies because you don't have enough samples; they're actually *removed* from the analog signal before it's digitized. If they weren't, you could recover them with enough microphones, but you'd also get weird aliasing artifacts. As it is, they're gone, never to return.
Fine by me. As long as Comcast and company are open and clear about what they're doing. That's the sort of industry regulation I'm in favor of -- require it to be very clear to consumers exactly what service they're buying, and require the provider to actually provide the service as advertised. If, with all parties aware of what's happening, Comcast wants to sell a bittorrent-limiting service, and customers want to buy it, then more power to them.
Slashdot Mirror
Normally, insurance companies aren't allowed to drop coverage or raise rates based on information that becomes available after your coverage starts. (If you have information that you withhold when signing up, that's a different story.)
There's another interesting potential use for genetic disease screening. I don't know if it applies to any known diseases now, but I'm certain it will eventually.
Suppose there's a disease for which there is a genetic test, but also good preventative care options. Ideally, you'd want to take the test, find out that you're at risk, and have the insurance company pay for the prevantative care. In an ideal world, that's what they would want too -- the preventative care would cost them less than paying if you actually did get sick. Both parties would then want the test administered, and want the other party to know the result.
Insurance companies should charge the correct amount, given all knowledge of risk factors -- insurance should be spreading risk among your insurance pool, not about getting free medical care. The problem is that this results in morally repugnant things like denying people care for unlikely conditions they're predisposed to but had no control over.
In an ideal world, this would be an argument for socialized medicine, or at least goverment-run health insurance for such factors. At a societal level, a certain fraction of people will have genetic anomalies, and at a societal level we shouldn't discriminate against them. It's no different from a car accident -- even if you're a good, careful driver, you roll the dice every time you get behind the wheel. The difference is that the dice get rolled (at least in part) long before the healthcare bills start to appear. But, that doesn't make it any less of a "risk" -- it just means that people can attempt to opt out of paying their share after they've checked the dice roll.
The ideal solution would be to cover everyone based on a pool composed of everyone -- aka mandatory government funded insurance (or socialized healthcare, depending on how you want to look at it).
(And to wander further afield -- there are lots of issues with our current healthcare system. I'm skeptical that socialized healthcare is the answer, and I don't know that implementing government insurance for genetic conditions would actually improve anything. At the same time, I find it ethically wrong to allow people to use genetic results they know about in decisions about what healthcare coverage to buy, but not allow the providers to use the same information in decisions about what to charge. It's a difficult and many-faceted problem, and I won't pretend to have even some of the answers.)
Well yes, the cheap unshielded meter didn't belong. What amazed me, though, was how *badly* it didn't belong. It wasn't actually on the same power system as any of the analog stuff, just in the same rack. (The analog system was all powered off various linear supplies from the inverter I mentioned.) The system it was on was a collection of lead-acid batteries, and a large number of valves controlled through relays, commanded by a PLC and control panel at the other end of the cable. It was lighting up LEDs to the point that it interfered with usability. The interference with the analog system was actually relatively minor (though of course we didn't want it). A cheap digital panel meter should be perfectly at home in a circuit whose major components are relays, solenoids, position / pressure sensor switches, and LED indicators, all commanded by a set of mechanical switches and sequenced by an industrial grade PLC (which never showed any signs of caring).
That when the humans and the robots destroy each other in a nuclear war, it will be up to the cockroaches to continue the battle against the robot cockroaches?
Either way you need to do voltage conversion. Transmission grids need to be at high voltage for efficiency (that's why AC was chosen originally -- you can convert it with transformers). These days power electronics have gotten cheap, so you can build an inexpensive DC-DC or DC-AC converter to connect your solar panel to the grid, but either way you do it you'll have to get it to high voltage somehow.
Yep, for ever expanding definitions of "short distances." High voltage high power DC silicon is getting better and cheaper, so we're already seeing a few long-haul DC lines where the reduced radiative losses and increased carrying ability of the cables makes it more efficient. On the other end, DC converters are becoming ubiquitous inside electronics. Google wants to standardize on only one voltage (12V) coming from your computer's PSU, and anything that wants another voltage just has its own converter. All of this will make power distribution and conversion cheaper, smaller, and more efficient. And some places are already moving to distributing DC to the rack-mount cabinets, at (iirc) 48V, and converting DC-DC from there.
On the flip side, I absolutely hate DC converters when trying to do analog work. If you buy a 16-bit analog to digital signal capture board, they run off a switching power supply -- a power supply that generates lots of high frequency noise. I've worked with ones that simply can't do better than about 12-13 bits -- the last 3-4 are basically worthless from the noise. I worked on a project that had need of a voltage monitor on a large battery-powered system involving a large number of sensors and controls. We installed a nice little digital LED voltmeter. It had a tiny, noisy switching converter in it. Not only did it dirty up all our data, it actually kicked back enough noise that it lit up LEDs that should have been electrically isolated from it (capacitative coupling through a long cable). Needless to say, we ripped it out and installed an analog (as in coil, magnet, needle) voltmeter.
Even good (read: expensive, as in $1/W or more) sine wave inverters are a pain. One we worked with produced wonderfully clean AC output -- but also created significant radiated noise, and dirtied up the DC input! Capacitors and ferrite beads helped, but not enough -- we eventually just moved it far away and ran the AC through a long extension cord.
So... DC/DC converters are wonderful for space, efficiency, and cost, but they can be quite the pain to work with if you actually care about having any sort of analog precision anywhere nearby. That sort of noise is amazingly hard to get rid of, and you often can't get rid of all the switching supplies these days.
What's to stop it from killing the engine to the police car?
Directional antennas are not exactly new technology. They work just fine for high-power microwave transmitters.
In the case of a zero-day exploit, 48 hours is an entirely reasonable expectation. As far as I can tell, it is far from the norm, however. For less egregious flaws, feature upgrades, or normal bugs, a slower turnaround would be perfectly reasonable (and what I'd expect). But for a zero day exploit... Well, I firmly believe software companies need to be held more accountable for those by their customers. If that's what your customers are doing, then good for them.
Exactly.
And why would we expect the market to "punish" them? Does anyone actually expect it to cost them sales or other revenue, or increase their costs, or otherwise have a relevant impact on their financial status?
Well, in terraforming terms, finding stuff to make up the Martian atmosphere probably isn't that hard. There are significant CO2 ice caps, and there may be significant water available with modest effort. CO2 plus plants gives you O2. Also, there is some good evidence to suggest that the icecaps' existence is bistable -- that is, if you could mostly evaporate them, the additional greenhouse effect would warm the planet enough to finish the job and keep it that way.
Basically, the problem of terraforming is to find resources that are already available in almost the form you want, and find some way to leverage your input effort. You don't want to have to process every single megaton of atmosphere you want to add / remove. It's far easier to (for example) dust carbon black on the poles and add a few orbiting mirrors.
Of course, the only reference I have handy is Zubrin's The Case for Mars which is a bit dated but (I think) still basically correct. The details may well have changed thanks to newer lander data.
So what? Humanity has shown us repeatedly that such minds exist, and that we know of no way of changing that. Yes, we should absolutely blame the people writing and operating these botnets. But, does that mean we should be giving them as much help as we are? You pose a false dichotomy -- we can most certainly blame the people responsible, while also making it harder for them to do things like this.
This wouldn't be slashdot without a car analogy. You have keys for your car, but clearly they wouldn't be necessary if not for the people who steal cars. Saying you should have keys for your car is like blaming the bullet instead of the killer.
I wonder what would happen if the electric company billed for the volt-amps consumed, instead of the watts, and then reported both numbers (together with your power factor) on the bill. I also wonder what would be required to do whole-house power factor correction? How much cost would it add if you were going to install a grid-tie solar system or something similar? How do these numbers compare to the added cost of power factor corrected power supplies in consumer electronics?
Empirically, they can't. It does not matter why, unless with that answer comes some insight into how to change it. It would appear that simply telling them to do better has no impact. If *you* want to save power, then that method has some hope of success. If a large organization or society wants to save power, that method is almost hopeless. So, given that you can't just tell people to conserve energy and expect it to work, what can you do? Incentives or mandates for more efficient standby modes is one solution that might actually have an impact.
Is there actually an article there? All I see is a giant ad a crapload of cookies I had to deny (mostly from 3rd-party sites, too -- Ick).
Have you ever tried tossing an RFID tag into the microwave? I've done that with a couple I've gotten in books. Antennas designed to receive microwaves produce an impressive spark show...
Unfortunately, there doesn't appear to be a Debian package, and the downloadable version won't run on my machine. Did they really have to link against a specific version of the libc?
Doesn't use flash, just standard HTML. And chok full of lovely absurdist humor. And undead elbow macaroni.
Kingdom of Loathing
Completely free, no ads, donation supported. Fair warning: a few of the items are easiest to get by donating, but those are nice-to-haves, and it's certainly no big deal if you don't have them.
Technically true, but beside the point. People who talk about fuel cells for cars aren't talking about a closed system. They're talking about distributing fuel and using atmospheric oxygen, and exhausting the waste products. Part of the reason for this, I imagine, is that closed cycle fuel cell system efficiencies are poor compared to batteries and capacitors. So fuel cells should be lumped in with gasoline engines, diesel engines, non-plugin hybrids, etc. in that comparison.
From a public policy standpoint, the interesting questions are system efficiency, and whether the energy is distributed through the electric grid or by carrying fuel around. Unless I'm missing something, these people are talking about the latter.
For now, it's batteries. But in the not too distant future, it may well be supercapacitors. Supercapacitors now are about a factor of ten away from lithium-ion batteries; improvements that are currently in labs appear to be able to remove most or all of that gap. Right now supercapacitors are expensive, but once the market starts growing they should come down in price. There are relatively fundamental limits to how much better traditional batteries can get in terms of capacity, but the apparent limits on supercapacitors are phenomenal. It might be 10 years before they see serious use, but I imagine small-scale use will be here sooner than that, especially if the rumors are to be believed.
Fuel cells are interesting, but I think that direct electrical storage through batteries and later supercapacitors is more likely to actually work out.
Well, 4kHz, because that's the Nyquist frequency for 8kHz sample rate. Unfortuntaley, if you're intending to use the data without doing the dithering work, you need to get rid of the aliasing artifacts -- which, uncoincidentally, are the exact data that would allow the dithering to function.
If it's doing that, then there's a digital AA filter before the resampling for the same purpose as the analog AA filter. It would have the same effect.
The problem with this approach is that audio ADCs have an analog antialiasing filter in front of them. It's not just that you can't see the high frequencies because you don't have enough samples; they're actually *removed* from the analog signal before it's digitized. If they weren't, you could recover them with enough microphones, but you'd also get weird aliasing artifacts. As it is, they're gone, never to return.
Fine by me. As long as Comcast and company are open and clear about what they're doing. That's the sort of industry regulation I'm in favor of -- require it to be very clear to consumers exactly what service they're buying, and require the provider to actually provide the service as advertised. If, with all parties aware of what's happening, Comcast wants to sell a bittorrent-limiting service, and customers want to buy it, then more power to them.
they are not the only real law breakers here.
There, fixed that for you. There's plenty of blame to go around on this one.