So over several years he regularly irradiated yeast cultures to accelerate mutations, and then subjected them to increasing levels of alcohol, selecting for tolerance.
I can think of no better way to make truly mad science socially acceptable. This man is a hero.
Why? Why do you want to understand these people? I'm serious. Why deliberately fill your head with hatred and evil and seek to know what motivates these people? Can you? Is it possible? To what end?
Most "hate sites" aren't filled with cartoon villains spewing senseless hate. The media presents a caricature of insane mass murderers, but if you're adventurous enough to go and have a look for yourself, they're often filled with reasonable people who are raising some good points about how they've been marginalized in some way or another. There's more than one side to every story, but the news isn't ever going to tell the side of such a politically unpopular group for fear of being branded as hate-group-sympathizers.
So how are you going to know which ones are just bigots and assholes and which ones are intelligent people who actually have something interesting to say? You go listen for a bit with an open mind. Even if you don't agree with the conclusions they draw it's sometimes quiet enlightening to find out how they got there.
I just think 10-15 years to get completely ahead when they're currently where NASA was in the mid-'60s (initial manned suborbital and LEO exploration) is a bit optimistic.
Longer term I agree. Hands down, the commercial side with greatly outpace NASA and ESA.
"My vision is for a fully reusable rocket transport system"... NASA had that vision with the Space Shuttle, but even excluding all R&D and capital purchases, just the incremental costs per launch were orders of magnitude higher than $500k per seat. And that's just to LEO! OK, that's "halfway to anywhere", but maintenance is a bitch, the staff required is huge, on and on... NASA isn't a role model for efficiency, but I seriously doubt that the commercial sector is going to be able to outdevelop them in just 10-15 years.
Yes, it's backwards. array.join(',') is more sensible: join is always performed on an array (so why is it in class string instead of class array?), and ',' is a detail on how you're doing it. You are far more likely to: createlistofstuff().join(',') than: createseparator().join(listofstuff) .
Somewhere between the growing totalitarian hell of the US and UK and the apparently overly-respectful approach of the Dutch... Somewhere, is there a sane country where I can live?
... or mid-size financial institution, trying to achieve slightly more efficient arbitrage by communicating market data from one side of the earth to the other a few milliseconds faster by going through instead of around.
Just suggest it to them and they'll have it in mass production next month and be working on compact, cost-reduced versions in a year.
So you'll handle it much like most hotplug PC hardware is these days, with latches and mechanical disconnects that ensure + and - are disconnected simultaneously.
You don't want to disconnect both simultaneously. The idea is to disconnect + first and leave ground connected. The voltage across the whole component falls to ground level instead of potentially floating up to + briefly.
This is a different problem from what the GP was talking about: When you hot-unplug a device drawing lots of DC, it starts to draw an arc. The arc will continue to draw until it gets too long to be stable, forms a big rainbow, and then extinguishes itself. The distance depends on the voltage. Thus you need non-mechanical, solid-state ways to shut off the current, or mechanical switches with an arc-extinguishing enclosure for really big applications like transmission lines.
When you hot-unplug a device on AC, it starts to draw an arc, but within 1/120 of a second the current (possibly a little out of phase with voltage) will cross zero, thus extinguishing the arc; and it won't reignite except if the contacts are VERY close and then will promptly be extinguished again on the next cycle.
If you could use any permutation of the 9 numbers (including 1 to 9, 3 to 7, etc), you would have 9!==362,880 permutations. Excluding all the moves you can't do, it supposedly comes down to 100,000 or so (I haven't worked it out myself).
Even if it was 2 million possibilities, and assuming it's key-strengthened so that each try takes one second of arm-CPU time, it would only take 23 days to brute force on a similar CPU. With a multicore AMD64 processor, GPU, cluster, etc, it could be done much faster.
Regarding "950Hz in, 1000Hz plus some lower frequency components back out."
Let's say I play back the 30 samples from s(0.98,30) as shown here:http://sagenb.org/home/pub/4502/cells/1/sage3.png . It's a single sine wave, it's below the Nyquist frequency... But if I play that back, I will hear a 1000Hz tone increasing in amplitude.
a) is there any reason that would sound like 980Hz?
b) How WOULD you create a 980Hz tone through a 1000Hz DAC? What samples would you use that give a better result than the ones shown above?
No. You added in something which isn't true: the bit about "a mixed signal that doesn't change over time". The sampling theorem states that any arbitrary mathematical function f(t) can be perfectly reconstructed from an infinite series of samples, so long as the sampling frequency is >2X the highest frequency component of f(t). It does not matter if that function is as simple as sin(t), or a ridiculously complex function which generates the exact same waveform output by an orchestra playing Beethoven.
You are mostly correct, but you're missing a finer point: you can't accurately reconstruct the original signal/function from the samples (unless that function is in a specific set) - you can only reconstruct the approximation as sampled. More on that in a moment.
Real world implementations of sampling approximate the theoretical ideal of perfect reconstruction much more closely than you seem to think, despite having finite sampling windows.
I never said it was bad. Aliasing distortion isn't huge until you're in the neighborhood of half the Nyquist frequency, and our ability to perceive it decreases well before it becomes significant.
Er, what? Fourier transforms do not output a number of frequency bands equal to the number of samples used as input.
This isn't my field, so correct me if I go wrong here. Usually FFT gives you back bins from the sample period to the nyquist limit, spaced by sample-length periods. There's no reason you couldn't use non-integer periods for your bins in a DFT, but it's just an alternate representation of the same data.
No. You've got a very distorted (heh) idea of how sampling systems work.
Think of it this way. Make a set of all functions f(t) which have no frequency components greater than F. Shannon proved mathematically that if you sample any member of that set at a sampling frequency greater than 2*F, the resulting sequence of points uniquely corresponds to that function, and no other member of the set. Another thing which fell out of this result was a method of mathematically reconstructing the original continuous waveform from the samples -- a form of curve fitting, if you will.
This is where we disagree. That sequence of points will uniquely correspond to that function if that function is a sum of sines. It's not true for arbitrary audio. You can find a sum of sines that will recreate the sampled waveform, but the aliasing distortion is already present in the sampled data. Unless you can assume that the function is in the set you described, you can only create a function, not the function, that generated those samples; and the one you get isn't necessarily going to accurately represent frequencies near the Nyquist frequency. You just get back the distorted waveform, and that distorted waveform won't sound like the original when played back, even though the original had no components above the Nyquist frequency. See my link below for what that distortion looks like.
When I run that thought experiment in my head, I get a 999Hz tone with no warble. You need a lot more explanation of your chain of reasoning as to why one should expect a 1000 Hz tone back!
950Hz in, 1000Hz plus some lower frequency components back out.
The thing I originally found objectionable in your post was the idea that things rapidly go downhill after 5KHz or so.
I actually said "5kHz pretty well, up to 10kHz passably, and up to the high teens with increasing levels of distortion". I didn't mean it started to die after 5KHz; I meant that it was pretty damn clean at that point, and it wasn't really falling apart until the high teens. So perhaps we're just arguing over what I meant by "passably".:)
Thanks for taking the time to reply in depth. If you still think I'm fucked in the head, could you point me to what I should be reading to straighten myself out?
The trouble is that there is a whole lot of ambiguous territory. Sure, everyone's cool with being harvested once they're brain dead, but what if you're probably brain dead? What if your brain seems to be vaguely functioning but you've been in a coma for ten years and your body is slowly deteriorating? What if there's a one in a thousand chance you could recover?
The unfortunate thing is those ambiguous cases are the majority, not the minority.
For what it's worth, I'm perfectly comfortable with being used for spare parts if there's a one in a thousand chance that I'll recover but a one in two chance that something recovered from my body could give someone else life. Many people aren't, and they want to donate everything only after there's "absolutely no chance" they'll come back, which is as good as saying "I'm not a donor".
I find it disappointing how often people keep trying to duct tape a textbook and a video game together. It inevitably ends up the worst of both: a boring game with some shallow facts littered around.
Compare this to the Exploratorium: lots of things set up that are perfectly fun to play with in their own right, and only a minimal amount of writeup and lecture surrounding them in a nonimposing way... But all of them intriguing in a way that makes you stop and think: "wait... How does that WORK?". Once that moment strikes, there is absolutely nothing that can keep a kid from wandering around the machine a dozen times, pulling the lever and watching carefully to see the result, sometimes given a hint by someone wandering by, until they figure it out.
My idea of the best education ever is to just keep tantalizing kids with something neat that's suited to their level to get their curiosity going, then just keep giving them the resources they need to learn about it.
75 year old wiring tends to be pretty good. Back then they were doing K&T https://en.wikipedia.org/wiki/Knob_and_tube_wiring style. It looks scary as hell but it tends to be pretty safe - since your hot and neutral are separated by a considerable air gap and the wires are thermally insulated from the structure, an insulation failure (abrasion or overheating) usually doesn't burn the house down. It doesn't have modern safety features like a safety ground, but the actual wiring is fine.
The switch away from it has more to do with cost than safety. The guys installing it usually knew what they were doing and paid much better attention to detail than the average contractor dragging romex these days. It took a lot of time, but it was a reliable and safe system, and if it's installed, there's no reason to rip it out and replace it just because it's old.
every non-sinusoid wave is expressed as the sum of sinusoids..... Sampling theory says that a system which samples at 44.1 KHz can reconstruct (with near perfection) all sinusoid components below 22.050 KHz from the original signal. Not just one at a time. All of them, no matter how many.
I'm actually quite familiar with Nyquist-Shannon, and I believe you're mistaken.
Here's the key thing you're missing: A sufficiently long sample can recover all of the frequency components of a mixed signal that doesn't change over time. IE, if you sum 50 sine waves at different frequencies, then sample them at 200 points at a frequency at least double the highest frequency waves, you can recover 100% of the information.
If you only take 25 samples, you cannot recover all of the components. Taking the fourier transform of 25 samples will give you only 25 frequency bands. How could it give you more? There simply isn't enough information there.
In audio, all of the frequency components are changing all the time (or else it would be a very bland track). You don't have enough samples to recover all frequency components at all times in the track, and information gets lost - primarily in the form of distorting high frequency components.
The visual equivalent of this is the Moire pattern: large objects are represented cleanly, but small details near 2x the pixel size (analogous to sample length) are displayed with all kinds of distortion.
For a simple thought experiment, imagine what happens if you sample a 999Hz tone at 2000Hz and then play it back. You will not get 999Hz - instead you get a 1000Hz tone that warbles at 1Hz.
Please do point out where you think I'm mistaken in any of this.
What possible placebo can you give somebody that they won't figure out it wasn't LSD?
A sugar pill, the same as always. Given a sugar pill and told that it's a new experimental drug that may cure their problem, a significant portion of people will either be cured or report being cured. It's a large percentage for diseases with a significant psychological component like alcoholism. A significant percentage will also report a wide range of side effects, regardless of the nature of the disease.
You also don't tell either the LSD or the sugar group that they're receiving LSD. You just tell BOTH of them "its a new drug that we're trying; it can have side effects like completely fucked up hallucinations, but don't worry, we'll comfort you through it and you'll be fine tomorrow".
And sure enough, a bunch of people will report tripping balls on sugar pills, a bunch of people dropping acid report that nothing happened, but after you check all the statistics, you find that people who took the LSD had a better cure rate than those who didn't.
Here's the beauty of it: It doesn't matter that they figure out if it's acid or not. It doesn't matter what the mechanism is at all. All that matters is if it works or not.
Slashdot Mirror
So over several years he regularly irradiated yeast cultures to accelerate mutations, and then subjected them to increasing levels of alcohol, selecting for tolerance.
I can think of no better way to make truly mad science socially acceptable. This man is a hero.
Why? Why do you want to understand these people? I'm serious. Why deliberately fill your head with hatred and evil and seek to know what motivates these people? Can you? Is it possible? To what end?
Most "hate sites" aren't filled with cartoon villains spewing senseless hate. The media presents a caricature of insane mass murderers, but if you're adventurous enough to go and have a look for yourself, they're often filled with reasonable people who are raising some good points about how they've been marginalized in some way or another. There's more than one side to every story, but the news isn't ever going to tell the side of such a politically unpopular group for fear of being branded as hate-group-sympathizers.
So how are you going to know which ones are just bigots and assholes and which ones are intelligent people who actually have something interesting to say? You go listen for a bit with an open mind. Even if you don't agree with the conclusions they draw it's sometimes quiet enlightening to find out how they got there.
I just think 10-15 years to get completely ahead when they're currently where NASA was in the mid-'60s (initial manned suborbital and LEO exploration) is a bit optimistic.
Longer term I agree. Hands down, the commercial side with greatly outpace NASA and ESA.
"My vision is for a fully reusable rocket transport system" ... NASA had that vision with the Space Shuttle, but even excluding all R&D and capital purchases, just the incremental costs per launch were orders of magnitude higher than $500k per seat. And that's just to LEO! OK, that's "halfway to anywhere", but maintenance is a bitch, the staff required is huge, on and on... NASA isn't a role model for efficiency, but I seriously doubt that the commercial sector is going to be able to outdevelop them in just 10-15 years.
There are two kinds of Hungarian Notation:
Good: centigradeDegrees
FAIL: integerDegrees
Yes, it's backwards. array.join(',') is more sensible: join is always performed on an array (so why is it in class string instead of class array?), and ',' is a detail on how you're doing it. You are far more likely to: createlistofstuff().join(',') than: createseparator().join(listofstuff) .
Object oriented is apparently subjective too. How else do you explain: ','.join(array)
The way Pythonistas usually try to defend this is also enlightening.
On the other hand, I can certainly agree that it's faster than Ruby. Ruby is miserable in that regard.
That is very cool. If anyone who makes decisions is reading along, I would love to see an API like this made available!
Is there a datasheet for the chip that I can read in the meantime?
Somewhere between the growing totalitarian hell of the US and UK and the apparently overly-respectful approach of the Dutch... Somewhere, is there a sane country where I can live?
more like underwater city and moon colony
... or mid-size financial institution, trying to achieve slightly more efficient arbitrage by communicating market data from one side of the earth to the other a few milliseconds faster by going through instead of around.
Just suggest it to them and they'll have it in mass production next month and be working on compact, cost-reduced versions in a year.
Is there a way we could also store our encryption keys in the SE?
I wish you could teach this to the people in charge of education.
Excellent job nonetheless.
So you'll handle it much like most hotplug PC hardware is these days, with latches and mechanical disconnects that ensure + and - are disconnected simultaneously.
You don't want to disconnect both simultaneously. The idea is to disconnect + first and leave ground connected. The voltage across the whole component falls to ground level instead of potentially floating up to + briefly.
This is a different problem from what the GP was talking about: When you hot-unplug a device drawing lots of DC, it starts to draw an arc. The arc will continue to draw until it gets too long to be stable, forms a big rainbow, and then extinguishes itself. The distance depends on the voltage. Thus you need non-mechanical, solid-state ways to shut off the current, or mechanical switches with an arc-extinguishing enclosure for really big applications like transmission lines.
When you hot-unplug a device on AC, it starts to draw an arc, but within 1/120 of a second the current (possibly a little out of phase with voltage) will cross zero, thus extinguishing the arc; and it won't reignite except if the contacts are VERY close and then will promptly be extinguished again on the next cycle.
If you could use any permutation of the 9 numbers (including 1 to 9, 3 to 7, etc), you would have 9!==362,880 permutations. Excluding all the moves you can't do, it supposedly comes down to 100,000 or so (I haven't worked it out myself).
Even if it was 2 million possibilities, and assuming it's key-strengthened so that each try takes one second of arm-CPU time, it would only take 23 days to brute force on a similar CPU. With a multicore AMD64 processor, GPU, cluster, etc, it could be done much faster.
Numeric lock codes can be used to encrypt; there's no reason the pattern locks couldn't be used that way as well, though I haven't tried it.
If it's not encrypted, I'm REALLY surprised the FBI can't figure it out. Flash chips are very easy to dump.
I'm surprised the FBI can't just dump the flash and brute force it. There are only about 100,000 possible patterns.
Oh, forgot one thing.
Regarding "950Hz in, 1000Hz plus some lower frequency components back out."
Let's say I play back the 30 samples from s(0.98,30) as shown here:http://sagenb.org/home/pub/4502/cells/1/sage3.png . It's a single sine wave, it's below the Nyquist frequency... But if I play that back, I will hear a 1000Hz tone increasing in amplitude.
a) is there any reason that would sound like 980Hz?
b) How WOULD you create a 980Hz tone through a 1000Hz DAC? What samples would you use that give a better result than the ones shown above?
No. You added in something which isn't true: the bit about "a mixed signal that doesn't change over time". The sampling theorem states that any arbitrary mathematical function f(t) can be perfectly reconstructed from an infinite series of samples, so long as the sampling frequency is >2X the highest frequency component of f(t). It does not matter if that function is as simple as sin(t), or a ridiculously complex function which generates the exact same waveform output by an orchestra playing Beethoven.
You are mostly correct, but you're missing a finer point: you can't accurately reconstruct the original signal/function from the samples (unless that function is in a specific set) - you can only reconstruct the approximation as sampled. More on that in a moment.
Real world implementations of sampling approximate the theoretical ideal of perfect reconstruction much more closely than you seem to think, despite having finite sampling windows.
I never said it was bad. Aliasing distortion isn't huge until you're in the neighborhood of half the Nyquist frequency, and our ability to perceive it decreases well before it becomes significant.
Er, what? Fourier transforms do not output a number of frequency bands equal to the number of samples used as input.
This isn't my field, so correct me if I go wrong here. Usually FFT gives you back bins from the sample period to the nyquist limit, spaced by sample-length periods. There's no reason you couldn't use non-integer periods for your bins in a DFT, but it's just an alternate representation of the same data.
No. You've got a very distorted (heh) idea of how sampling systems work.
Think of it this way. Make a set of all functions f(t) which have no frequency components greater than F. Shannon proved mathematically that if you sample any member of that set at a sampling frequency greater than 2*F, the resulting sequence of points uniquely corresponds to that function, and no other member of the set. Another thing which fell out of this result was a method of mathematically reconstructing the original continuous waveform from the samples -- a form of curve fitting, if you will.
This is where we disagree. That sequence of points will uniquely correspond to that function if that function is a sum of sines. It's not true for arbitrary audio. You can find a sum of sines that will recreate the sampled waveform, but the aliasing distortion is already present in the sampled data. Unless you can assume that the function is in the set you described, you can only create a function, not the function, that generated those samples; and the one you get isn't necessarily going to accurately represent frequencies near the Nyquist frequency. You just get back the distorted waveform, and that distorted waveform won't sound like the original when played back, even though the original had no components above the Nyquist frequency. See my link below for what that distortion looks like.
When I run that thought experiment in my head, I get a 999Hz tone with no warble. You need a lot more explanation of your chain of reasoning as to why one should expect a 1000 Hz tone back!
Here, a picture is worth a thousand words.
http://sagenb.org/home/pub/4502/
950Hz in, 1000Hz plus some lower frequency components back out.
The thing I originally found objectionable in your post was the idea that things rapidly go downhill after 5KHz or so.
I actually said "5kHz pretty well, up to 10kHz passably, and up to the high teens with increasing levels of distortion". I didn't mean it started to die after 5KHz; I meant that it was pretty damn clean at that point, and it wasn't really falling apart until the high teens. So perhaps we're just arguing over what I meant by "passably". :)
Thanks for taking the time to reply in depth. If you still think I'm fucked in the head, could you point me to what I should be reading to straighten myself out?
The trouble is that there is a whole lot of ambiguous territory. Sure, everyone's cool with being harvested once they're brain dead, but what if you're probably brain dead? What if your brain seems to be vaguely functioning but you've been in a coma for ten years and your body is slowly deteriorating? What if there's a one in a thousand chance you could recover?
The unfortunate thing is those ambiguous cases are the majority, not the minority.
For what it's worth, I'm perfectly comfortable with being used for spare parts if there's a one in a thousand chance that I'll recover but a one in two chance that something recovered from my body could give someone else life. Many people aren't, and they want to donate everything only after there's "absolutely no chance" they'll come back, which is as good as saying "I'm not a donor".
I find it disappointing how often people keep trying to duct tape a textbook and a video game together. It inevitably ends up the worst of both: a boring game with some shallow facts littered around.
Compare this to the Exploratorium: lots of things set up that are perfectly fun to play with in their own right, and only a minimal amount of writeup and lecture surrounding them in a nonimposing way... But all of them intriguing in a way that makes you stop and think: "wait... How does that WORK?". Once that moment strikes, there is absolutely nothing that can keep a kid from wandering around the machine a dozen times, pulling the lever and watching carefully to see the result, sometimes given a hint by someone wandering by, until they figure it out.
My idea of the best education ever is to just keep tantalizing kids with something neat that's suited to their level to get their curiosity going, then just keep giving them the resources they need to learn about it.
Fantastic. Thanks for writing this up.
Guess where the bad guys are going to focus their efforts.
75 year old wiring tends to be pretty good. Back then they were doing K&T https://en.wikipedia.org/wiki/Knob_and_tube_wiring style. It looks scary as hell but it tends to be pretty safe - since your hot and neutral are separated by a considerable air gap and the wires are thermally insulated from the structure, an insulation failure (abrasion or overheating) usually doesn't burn the house down. It doesn't have modern safety features like a safety ground, but the actual wiring is fine.
The switch away from it has more to do with cost than safety. The guys installing it usually knew what they were doing and paid much better attention to detail than the average contractor dragging romex these days. It took a lot of time, but it was a reliable and safe system, and if it's installed, there's no reason to rip it out and replace it just because it's old.
every non-sinusoid wave is expressed as the sum of sinusoids. .... Sampling theory says that a system which samples at 44.1 KHz can reconstruct (with near perfection) all sinusoid components below 22.050 KHz from the original signal. Not just one at a time. All of them, no matter how many.
I'm actually quite familiar with Nyquist-Shannon, and I believe you're mistaken.
Here's the key thing you're missing: A sufficiently long sample can recover all of the frequency components of a mixed signal that doesn't change over time. IE, if you sum 50 sine waves at different frequencies, then sample them at 200 points at a frequency at least double the highest frequency waves, you can recover 100% of the information.
If you only take 25 samples, you cannot recover all of the components. Taking the fourier transform of 25 samples will give you only 25 frequency bands. How could it give you more? There simply isn't enough information there.
In audio, all of the frequency components are changing all the time (or else it would be a very bland track). You don't have enough samples to recover all frequency components at all times in the track, and information gets lost - primarily in the form of distorting high frequency components.
The visual equivalent of this is the Moire pattern: large objects are represented cleanly, but small details near 2x the pixel size (analogous to sample length) are displayed with all kinds of distortion.
For a simple thought experiment, imagine what happens if you sample a 999Hz tone at 2000Hz and then play it back. You will not get 999Hz - instead you get a 1000Hz tone that warbles at 1Hz.
Please do point out where you think I'm mistaken in any of this.
What possible placebo can you give somebody that they won't figure out it wasn't LSD?
A sugar pill, the same as always. Given a sugar pill and told that it's a new experimental drug that may cure their problem, a significant portion of people will either be cured or report being cured. It's a large percentage for diseases with a significant psychological component like alcoholism. A significant percentage will also report a wide range of side effects, regardless of the nature of the disease.
You also don't tell either the LSD or the sugar group that they're receiving LSD. You just tell BOTH of them "its a new drug that we're trying; it can have side effects like completely fucked up hallucinations, but don't worry, we'll comfort you through it and you'll be fine tomorrow".
And sure enough, a bunch of people will report tripping balls on sugar pills, a bunch of people dropping acid report that nothing happened, but after you check all the statistics, you find that people who took the LSD had a better cure rate than those who didn't.
Here's the beauty of it: It doesn't matter that they figure out if it's acid or not. It doesn't matter what the mechanism is at all. All that matters is if it works or not.