An Amazon driver stole UPS pkg from my house. Unlucky for them my cameras caught the event. UPS delivered the package at around 3 and Amazon delivered another package and picked up the UPS package and took the proof of their delivery picture.
The driver and "assistant" both got caught, fired and charged. Amazon was pretty difficult to deal with getting things made right, at least until the Sheriff showed up at their facility with the pictures and license places and talked to the driver.
So amazon probably does need to do this. It was not clear the assistant was a amazon employ, but the assistant did not get diversion while the driver did. The assistant had a previous theft record so this was not his first crime.
The provider he is probably talking about has some traffic shaping software on it, before the traffic shaping was introduces the ping times were around 600-800ms. About in range for speed of light delay, after the traffic shaping/modem firmware upgrade pings were now 1200, with once in a while pings dropping to the old range
First of all satellite is going to be horrible on latency, the *BEST* that can be done is around 600ms (speed of light), and at best that is going to be horrible on VOIP and VPN. And with most for *ONLY* $100 per month you will have bandwidth usage limits (the consumer plans have low bandwidth limits 20-40GB/month for that kind of money), to not have bandwidth limits and not be shared it will cost quite a bit more.
You might want to investigate using a point-to-point connection of your own, back before the TCOM companies had connections everywhere large companies did their own microwave links, and I understand that the new point-to-point wireless stuff (not consumer grade-commercial stuff) can do at least 20-30 (or more) miles point-to-point, if you have a line of sight to something with TCOM access you should be able to do it, or you could setup a string of relays to do it.
I never said it was a good idea, just add the *FACT* that almost all hydrogen that anyone currently uses came from natural gas as it is cheaper to produce hydrogen via that method than with electricity.
And in most parts of the world any electrolysis also produces carbon in the production of the electricity at some point in time.
It seems possible that the energy to separate the hydrogen could be coming from suppling the water at high pressure to the membrane, if so then the source of the energy to separate the water would be the original pump that is being used to supply the water to the membrane. One of the articles I found mentions that the pump supplying the water to the membrane was run from a dry-cell battery, and then turned off after things were started, so they could just be moving the energy from the dry-cell battery to separate things.
It would be useful if the membrane was more efficient at separating things than using electrolysis, but it would not be making any real power, the power for the original pump would still need to come from somewhere.
The amount of energy required to separate the molecular bond is equal (or greater after losses) to the amount of energy you get back when you run them back through a fuel cell, you don't gain anything. The question is where are they getting the energy to separate things from.
It costs more to produce hydrogen through the electrical method than by reforming natural gas to make hydrogen, so almost all hydrogen the world currently uses is made by reforming natural gas.
Actually,
Distinguishing between standard encrypted data streams and jpeg files (or most other non-encrypted binary file) is trivial. If you calculate something called entropy (used in data compression) over 100k or more of data, the encrypted data stream will almost always be completely random (7.99-8.00), and the jpeg stream will not be as random (7.96 or less).
Though if one works to de-randomize the encrypted data by adding non-random stuff in it to make this not work, it gets more difficult, and if one changes the encrypted data stream to instead place a common word in place of a number (of say 0-65536 a different work for each) it gets much harder to determine that the string of nonsense is actually an encrypted data stream, and there are probably a number of schemes like that that could be used to hide encrypted data streams so that it would take a lot of work to figure out what was going on.
If you want a certain answer on a poll, if you ask the question in the correct way, you can usually get the answer that you want.
Like: Does it bother you that the US govt increased domestic spying to keep you safe from the terrorists?
Rather than: Does it bother you that the US govt increased domestic spying is keeping track of everything that you do?
The first one will get a more positive answer against domestic spying than the second one, and I would bet the polls questions being used are heavily loaded to get the answer the poll taker wants.
I have satellite, the one thing that will annoy you with working remote is that there is a round trip delay of around a 1/2 second at best (22k miles up then back down and then both again for the return packet-1/2second at speed of light). Usually unless it is the middle of the night the delay is closer to a full second, this is painful with ssh connections and other similar interactive work. VPN at best work sometimes, and at worst does not work at all. And all of the satellite providers have bandwidth limits on a per month basis, though if you are used to dial-up this is probably not a big issue. Dial-up is much better for low-bandwidth interactive work, and the high bandwidth interactive work (VPN, X-windows) gets very ugly on satellite if it works at all.
Around here the Long range dedicated point-to-point wireless goes for $100/MB/month+equipment, and the point-to-point towers can go quite a distance (15+ miles I understand). If you could get a location that had LOS with a provider and LOS to your location or physically close enough, you could pay to put up a tower and then get that signal to your location either through fiber (copper cable is dangerous to run outside for long distances as close lightening strikes can put large voltages on everything) or again wireless. If you have a neighbor in the correct location and they also want internet you could probably work out a deal with them. I understand though the tower+equipment can be at least several thousand $.
Power consumption depends on the type of CPU's and the type of RAM.
I work for an OEM and have tested one of those quad motherboards with 4 - 852 cpus, and 64GB of ram, and the power consumption was under 600 watts, power consumption
varies depending on the amount and type of ram and manufacturer of the actual ram
chips, and the type of cpu used.
They do seem to work nicely, the 64G setup won't fit in a 1u setting, the 32G setup will fit in the 1u chassis.
That would do it too, so would someone figuring out to cheaply and easily convert 90-100% of the energy in any fuel to electricity, that to would kill everyone getting the normal 30-50%. Fuel cells meet all of the above except for the "cheaply and easily" part.
If a perpetual motion machine were actually possible:
We won't know about it until the guy who invented it quietly competes with and puts out of business the oil and power companies. If one were to actually work there would be way too much money in it to even think about letting someone else have a piece of your billions. If it actually worked it would easy to go into the power generation business, since no one yet has then none work.
And the arguements about supression don't hold water, which ever oil/power company that bought it would use it to eliminate their competition.
Electric motors are already 80-90% efficient, while this might make it closer to 100% it won't go over, unless someone discovered some new laws of physics. Given that they attempt to make the claim of greater than 100% I suspect the entire thing is full of crap.
So produce an EMP near electroinic detonators?
If you could get a big enough EMP (would would be hard unless your missle contains a small nuke itself), then that would for sure detonate the incoming, maybe not as a full yield device, but certainly the explosive used to initiate the nuclear explosion would go off and at at least make it a dirty bomb, and probably a part nuclear explosion.
The problem with false positivies is that even if the number of false positives is low, the absolute number of events is high.
If the test is only wrong.001% of the time (99.999 correct) then you have 10 false positivies for each million people through the system. If you assume that the actual correct positive account is 1 in a million that are trying to bypass something, then given the huge number of people being run through this system this will mean that if you get a "hit" it is more likely a false positive than a real positive, 10 falses for each 1 real.
If the test is 99.99% correct then the failures go up by a factor of 10.
Even with high correctness, false positives are a real problem espeically when the number of correct positives is a given population is very low.
Slashdot Mirror
An Amazon driver stole UPS pkg from my house. Unlucky for them my cameras caught the event. UPS delivered the package at around 3 and Amazon delivered another package and picked up the UPS package and took the proof of their delivery picture.
The driver and "assistant" both got caught, fired and charged. Amazon was pretty difficult to deal with getting things made right, at least until the Sheriff showed up at their facility with the pictures and license places and talked to the driver.
So amazon probably does need to do this. It was not clear the assistant was a amazon employ, but the assistant did not get diversion while the driver did. The assistant had a previous theft record so this was not his first crime.
The provider he is probably talking about has some traffic shaping software on it, before the traffic shaping was introduces the ping times were around 600-800ms. About in range for speed of light delay, after the traffic shaping/modem firmware upgrade pings were now 1200, with once in a while pings dropping to the old range
First of all satellite is going to be horrible on latency, the *BEST* that can be done is around 600ms (speed of light), and at best that is going to be horrible on VOIP and VPN. And with most for *ONLY* $100 per month you will have bandwidth usage limits (the consumer plans have low bandwidth limits 20-40GB/month for that kind of money), to not have bandwidth limits and not be shared it will cost quite a bit more.
You might want to investigate using a point-to-point connection of your own, back before the TCOM companies had connections everywhere large companies did their own microwave links, and I understand that the new point-to-point wireless stuff (not consumer grade-commercial stuff) can do at least 20-30 (or more) miles point-to-point, if you have a line of sight to something with TCOM access you should be able to do it, or you could setup a string of relays to do it.
The big question is how remote is it?
I never said it was a good idea, just add the *FACT* that almost all hydrogen that anyone currently uses came from natural gas as it is cheaper to produce hydrogen via that method than with electricity.
And in most parts of the world any electrolysis also produces carbon in the production of the electricity at some point in time.
It seems possible that the energy to separate the hydrogen could be coming from suppling the water at high pressure to the membrane, if so then the source of the energy to separate the water would be the original pump that is being used to supply the water to the membrane. One of the articles I found mentions that the pump supplying the water to the membrane was run from a dry-cell battery, and then turned off after things were started, so they could just be moving the energy from the dry-cell battery to separate things.
It would be useful if the membrane was more efficient at separating things than using electrolysis, but it would not be making any real power, the power for the original pump would still need to come from somewhere.
The amount of energy required to separate the molecular bond is equal (or greater after losses) to the amount of energy you get back when you run them back through a fuel cell, you don't gain anything. The question is where are they getting the energy to separate things from.
It costs more to produce hydrogen through the electrical method than by reforming natural gas to make hydrogen, so almost all hydrogen the world currently uses is made by reforming natural gas.
Actually, Distinguishing between standard encrypted data streams and jpeg files (or most other non-encrypted binary file) is trivial. If you calculate something called entropy (used in data compression) over 100k or more of data, the encrypted data stream will almost always be completely random (7.99-8.00), and the jpeg stream will not be as random (7.96 or less). Though if one works to de-randomize the encrypted data by adding non-random stuff in it to make this not work, it gets more difficult, and if one changes the encrypted data stream to instead place a common word in place of a number (of say 0-65536 a different work for each) it gets much harder to determine that the string of nonsense is actually an encrypted data stream, and there are probably a number of schemes like that that could be used to hide encrypted data streams so that it would take a lot of work to figure out what was going on.
If you want a certain answer on a poll, if you ask the question in the correct way, you can usually get the answer that you want. Like: Does it bother you that the US govt increased domestic spying to keep you safe from the terrorists? Rather than: Does it bother you that the US govt increased domestic spying is keeping track of everything that you do? The first one will get a more positive answer against domestic spying than the second one, and I would bet the polls questions being used are heavily loaded to get the answer the poll taker wants.
I have satellite, the one thing that will annoy you with working remote is that there is a round trip delay of around a 1/2 second at best (22k miles up then back down and then both again for the return packet-1/2second at speed of light). Usually unless it is the middle of the night the delay is closer to a full second, this is painful with ssh connections and other similar interactive work. VPN at best work sometimes, and at worst does not work at all. And all of the satellite providers have bandwidth limits on a per month basis, though if you are used to dial-up this is probably not a big issue. Dial-up is much better for low-bandwidth interactive work, and the high bandwidth interactive work (VPN, X-windows) gets very ugly on satellite if it works at all.
Around here the Long range dedicated point-to-point wireless goes for $100/MB/month+equipment, and the point-to-point towers can go quite a distance (15+ miles I understand). If you could get a location that had LOS with a provider and LOS to your location or physically close enough, you could pay to put up a tower and then get that signal to your location either through fiber (copper cable is dangerous to run outside for long distances as close lightening strikes can put large voltages on everything) or again wireless. If you have a neighbor in the correct location and they also want internet you could probably work out a deal with them. I understand though the tower+equipment can be at least several thousand $.
Power consumption depends on the type of CPU's and the type of RAM. I work for an OEM and have tested one of those quad motherboards with 4 - 852 cpus, and 64GB of ram, and the power consumption was under 600 watts, power consumption varies depending on the amount and type of ram and manufacturer of the actual ram chips, and the type of cpu used. They do seem to work nicely, the 64G setup won't fit in a 1u setting, the 32G setup will fit in the 1u chassis.
That would do it too, so would someone figuring out to cheaply and easily convert 90-100% of the energy in any fuel to electricity, that to would kill everyone getting the normal 30-50%. Fuel cells meet all of the above except for the "cheaply and easily" part.
If a perpetual motion machine were actually possible: We won't know about it until the guy who invented it quietly competes with and puts out of business the oil and power companies. If one were to actually work there would be way too much money in it to even think about letting someone else have a piece of your billions. If it actually worked it would easy to go into the power generation business, since no one yet has then none work. And the arguements about supression don't hold water, which ever oil/power company that bought it would use it to eliminate their competition.
Electric motors are already 80-90% efficient, while this might make it closer to 100% it won't go over, unless someone discovered some new laws of physics. Given that they attempt to make the claim of greater than 100% I suspect the entire thing is full of crap.
So produce an EMP near electroinic detonators? If you could get a big enough EMP (would would be hard unless your missle contains a small nuke itself), then that would for sure detonate the incoming, maybe not as a full yield device, but certainly the explosive used to initiate the nuclear explosion would go off and at at least make it a dirty bomb, and probably a part nuclear explosion.
The problem with false positivies is that even if the number of false positives is low, the absolute number of events is high.
.001% of the time (99.999 correct) then you have 10 false positivies for each million people through the system. If you assume that the actual correct positive account is 1 in a million that are trying to bypass something, then given the huge number of people being run through this system this will mean that if you get a "hit" it is more likely a false positive than a real positive, 10 falses for each 1 real.
If the test is only wrong
If the test is 99.99% correct then the failures go up by a factor of 10.
Even with high correctness, false positives are a real problem espeically when the number of correct positives is a given population is very low.
Roger