According to the whitepaper, the coefficients to weight each transmitter signal to constructively interfere at your location sets up mathematically orthogonal channels (at lease orthogonal to some SNR, with some leakage from other channels depending on the number/location of devices and antennas).
The device can send a signal back which will interfere with other devices, but incoming signals at the antennas can be weighted by the same coefficients (or at least derived from the same) to again cancel all the other signals but your own.
Mathematically, the channel can go both ways with full bandwidth.
Actually, I find it interesting you used the Magic-8-Ball. While it is just a toy, the 20Q toy (which is somewhat similar in my mind) is very interesting because it tries to guess what you are thinking of by asking you yes/no/sometimes/don't-know questions. The neural net was then built by people playing the game and providing 'better questions' for when the AI got the answer wrong.
Could you not do the same for medical diagnosis?
[Do you have a headache?] "No." [Does your stomach hurt?] "Yes." [You have an ulcer?] "No." [What's wrong, and what would have been a better question?] "Food poisoning, and 'Did you eat uncooked meat recently?'" [Noted. Now I am smarter.]
It seems that you could make a diagnosis engine that as you rule things out it could come to as good a conclusion as a typical doctor.
So I simulated your hypothetical experiment. I was going to use a 16 toss sequence, but it was going to take MATLAB 22 hours, so I only had it look for a 8 toss sequence.
A coin is tossed 1 million times. Then we count the total number of times each possible 8-toss sequence occurs. Turns out that each pattern shows up about the correct number of times. No pattern even comes close to showing up 1.1x more than expected. Let alone, 100x.
BILLIONS of women mess around with their hormones every month just for being a human woman. I've had shoes thrown at me because of this. Once she started consciously messing with her hormones: no more shoe projectiles.
Many of the comments on here are "1.8 microseconds, oh no I get less sleep! What a stupid finding."
But seriously, does this have an effect on GPS? GPS satellites need to be corrected for relativistic effects that cause their clocks to tick 38 microseconds/day different than the ground; which would cause error to accumulate at 10km/day. Does 1.8 microsecond difference in our day cause error to accumulate in GPS at the rate of 0.5km/day if not fixed?
Because power, delay, and area are correlated to the effective length of the transistor.
You do want lower power in portable devices.
You also want higher speed.
You can get both if you decrease transistor area, which leads to increased density.
If you still think that getting a computer to not only run a simulation but also generate all the random events that effect that simulation is still a plausible answer with computer technology today - I don't know how you arrived at that answer.
I merely only implied that it is possible to generate random numbers, which is very useful.
Perhaps you want to simulate something very specific and you know that the random part of the simulation has very specific characteristics (e.g. transient noise in SPICE), this is something that is easy to implement in software.
I am no expert in cryptography, but I remember my CS friends talking about 2 locks on a briefcase.
The shared key is encrypted with key A by Adam and sent to Betty (who can not open it without key A). Betty encrypts this with key B and sends it back to Adam. Adam decrypts this with key A and sends in back to Betty. Since the message now is only encrypted with key B, Betty can open it and get the shared key.
Or is this only a educational exercise and in actuality if you eavesdrop all of these transactions you can determine key A and B?
By iteratively solving across the entire solution space and comparing this to the algorithm output.
For a class project, this is possible; however, I understand that you would truly apply optimization algorithms to problems in order to NOT iteratively solve them.
Are you implying that the genetic algorithm doesn't find the global maximum?
optimize - to make as perfect, effective, or functional as possible
"As a general rule of thumb genetic algorithms might be useful in problem domains that have a complex fitness landscape as recombination is designed to move the population away from local optima that a traditional hill climbing algorithm might get stuck in." http://en.wikipedia.org/wiki/Genetic_algorithm
While it is true that sexual crossover of genes will only optimize to the best solution available in the gene pool, mutation allows a global solution to be found. When I tested this algorithm myself, I specifically chose an over-constrained problem that had many local maxima and no clear "best-solution." The genetic algorithm then found the multiple "best" solutions and had 3 distinct gene groups, with the largest being the most fit according to the fitting function. So while the entire gene pool did not converge to a single maximum, it did find the global maxima.
How about "optimize" instead of fashion.
After doing some programming work with genetic algorithm optimization, I realized that although evolution may choose one gene path or another, the end result is always the same: the most optimized solution for survival, tool making, etc.
So I agree, not designed; absolutely optimized.
Banning all liquids was a good idea because of the given threat.
Soon *the terrorists* my find a way to detonate their clothing and all clothing will be banned in the cabin. This sounds pretty cool at first, but keep in mind how often is the random person sitting next to you a slammin' hottie?
After that *the terrorists* will find a brilliant plot to set themselves on fire by rubbing their arms together REALLY FAST. Once this happpens, all PASSENGERS will be banned from being in the cabin. Very smart.
Slashdot Mirror
According to the whitepaper, the coefficients to weight each transmitter signal to constructively interfere at your location sets up mathematically orthogonal channels (at lease orthogonal to some SNR, with some leakage from other channels depending on the number/location of devices and antennas).
The device can send a signal back which will interfere with other devices, but incoming signals at the antennas can be weighted by the same coefficients (or at least derived from the same) to again cancel all the other signals but your own.
Mathematically, the channel can go both ways with full bandwidth.
Actually, I find it interesting you used the Magic-8-Ball. While it is just a toy, the 20Q toy (which is somewhat similar in my mind) is very interesting because it tries to guess what you are thinking of by asking you yes/no/sometimes/don't-know questions. The neural net was then built by people playing the game and providing 'better questions' for when the AI got the answer wrong.
Could you not do the same for medical diagnosis?
[Do you have a headache?]
"No."
[Does your stomach hurt?]
"Yes."
[You have an ulcer?]
"No."
[What's wrong, and what would have been a better question?]
"Food poisoning, and 'Did you eat uncooked meat recently?'"
[Noted. Now I am smarter.]
It seems that you could make a diagnosis engine that as you rule things out it could come to as good a conclusion as a typical doctor.
One of the riders:
"Prohibits NASA from collaborating with China"
So I simulated your hypothetical experiment. I was going to use a 16 toss sequence, but it was going to take MATLAB 22 hours, so I only had it look for a 8 toss sequence.
A coin is tossed 1 million times. Then we count the total number of times each possible 8-toss sequence occurs. Turns out that each pattern shows up about the correct number of times. No pattern even comes close to showing up 1.1x more than expected. Let alone, 100x.
See here: http://i.imgur.com/5F391.png
Pattern number is just dec2bin, i.e., #0 -> TTTTTTTT and #10 -> TTTTHTHT, etc.
BILLIONS of women mess around with their hormones every month just for being a human woman. I've had shoes thrown at me because of this. Once she started consciously messing with her hormones: no more shoe projectiles.
According to TFA, a travelling wave reactor is a type of breed-burn reactor.
http://www.terrapower.com/Technology/Timeline.aspx
BTW: GPS satellites are NOT in geo-synchronous orbit.
Many of the comments on here are "1.8 microseconds, oh no I get less sleep! What a stupid finding."
But seriously, does this have an effect on GPS? GPS satellites need to be corrected for relativistic effects that cause their clocks to tick 38 microseconds/day different than the ground; which would cause error to accumulate at 10km/day. Does 1.8 microsecond difference in our day cause error to accumulate in GPS at the rate of 0.5km/day if not fixed?
Because power, delay, and area are correlated to the effective length of the transistor. You do want lower power in portable devices. You also want higher speed. You can get both if you decrease transistor area, which leads to increased density.
Ha, I thought you meant "Well they had a head start."
I simulated 100 coin flips 1,000,000 times and plotted the percent occurrence for every # of heads:
http://imgur.com/iVLp9.jpg
It looks like about 8% chance to get 50/50 and better than 2% chance of getting either 40/60 or 60/40.
I can't wait for www.celebrityairportscannerpics.com
;)
I mean, they all travel sometime
If you still think that getting a computer to not only run a simulation but also generate all the random events that effect that simulation is still a plausible answer with computer technology today - I don't know how you arrived at that answer.
I merely only implied that it is possible to generate random numbers, which is very useful.
Perhaps you want to simulate something very specific and you know that the random part of the simulation has very specific characteristics (e.g. transient noise in SPICE), this is something that is easy to implement in software.
While pseudorandom number generators are deterministic (LFSR, etc), there is no reason that we can not implement a true random generator in hardware.
There are many ways to do this. For example, amplify the noise from a resistor, quantize it, and use the LSBs. These should be random.
I am no expert in cryptography, but I remember my CS friends talking about 2 locks on a briefcase.
The shared key is encrypted with key A by Adam and sent to Betty (who can not open it without key A). Betty encrypts this with key B and sends it back to Adam. Adam decrypts this with key A and sends in back to Betty. Since the message now is only encrypted with key B, Betty can open it and get the shared key.
Or is this only a educational exercise and in actuality if you eavesdrop all of these transactions you can determine key A and B?
I thought they share a key with the http://en.wikipedia.org/wiki/Diffie-Hellman_key_exchange
So it doesn't matter if someone captures the initial handshake.
With this I absolutely agree and I concede.
There will definately be vestigal stuff left over unless there is evolutionary pressure to remove it, and usually there isn't.
By iteratively solving across the entire solution space and comparing this to the algorithm output.
For a class project, this is possible; however, I understand that you would truly apply optimization algorithms to problems in order to NOT iteratively solve them.
Are you implying that the genetic algorithm doesn't find the global maximum?
optimize - to make as perfect, effective, or functional as possible
"As a general rule of thumb genetic algorithms might be useful in problem domains that have a complex fitness landscape as recombination is designed to move the population away from local optima that a traditional hill climbing algorithm might get stuck in." http://en.wikipedia.org/wiki/Genetic_algorithm
While it is true that sexual crossover of genes will only optimize to the best solution available in the gene pool, mutation allows a global solution to be found. When I tested this algorithm myself, I specifically chose an over-constrained problem that had many local maxima and no clear "best-solution." The genetic algorithm then found the multiple "best" solutions and had 3 distinct gene groups, with the largest being the most fit according to the fitting function. So while the entire gene pool did not converge to a single maximum, it did find the global maxima.
It would be an interesting experiment to make a webpage that presents you two equivalent flash movies with varying frame rates (as you described).
Have the two movies in a random order with a random frame rate (say, 12, 15, 24, 30, 45, 60, etc)
Then have the user merely select which looks "better/smoother."
It would be interesting (as you point out) to see at what frame rate there is no longer a clear winner.
How about "optimize" instead of fashion.
After doing some programming work with genetic algorithm optimization, I realized that although evolution may choose one gene path or another, the end result is always the same: the most optimized solution for survival, tool making, etc.
So I agree, not designed; absolutely optimized.
Or some CD players: http://www.youtube.com/watch?v=gdAmEEAiJWo
Better yet. All passengers are sedated for the duration of the flight.
This is a great idea for a number of reasons.
Banning all liquids was a good idea because of the given threat.
Soon *the terrorists* my find a way to detonate their clothing and all clothing will be banned in the cabin. This sounds pretty cool at first, but keep in mind how often is the random person sitting next to you a slammin' hottie?
After that *the terrorists* will find a brilliant plot to set themselves on fire by rubbing their arms together REALLY FAST. Once this happpens, all PASSENGERS will be banned from being in the cabin. Very smart.
Terrorists win.