My solution is to put a big steel bracket around my brake pedal that would take more than a few minutes of cutting to get through before you can drive away. It doesn't prevent a determined party from taking the car if he really wants to, but it's a layer of actual physical security that prevents someone from duplicating a software key and riding away, just like it prevented someone from picking the lock and hotwiring the ignition and riding away on an older model.
The standard error for a Bernoulli trial (right or wrong are the only outcomes) looks like sqrt(p(1-p)/N). If N=100 (which looks like the trial he ran), then for p anywhere between 0.25 and 0.75 (even though we know it's really 0.5), sigma >= 3.3%.
This means that if you ran the N=100 experiment 100 times, about 33 or those trials will yield you a value of correct between 46.6% and 53.3%, 33 will be between 40 and 46.6%, another 33 will be between 53.3 and 60%, and the last one might be anywhere, but probably closer to 40 or 60.
This is what is known as small number statistics. And I learned it (correctly) in a graduate level engineering course which required several years worth of advanced mathematics in undergrad. My experience is that medical types usually don't take that route, and therefor they make asses of themselves getting excited over random noise.
Are human beings really that incapable of mental abstraction that they need some anthropomorphic gizmo telling them the same thing a normal machine can do, but with 10x the development cost?
My unscientific sampling two of Chinese engineering students rotating through at a good American engineering school before going back suggests to me that it's a larger value of ``soon'' than you might expect given the outward show of progress. Yes, they're dumping zillions into building up institutions, but in trying to catch up on 200+ years of Western experience in under a generation, there's a certain something missing. They're nowhere near cargo-cult level, but they're closer to it than we are in the west, hence my original comment.
Google "average height adult male" and somewhere at the top of the list will be a webpage from the NIH with lots of tables breaking down height and weight by age group.
The columns are:
Age, Subject N, Mean Height, Std. Dev. of Mean, etc...
Now, just glancing at the tables, the avg height for a 20-30yr old male was 69.2 inches. But the next column was 0.012in. I was like "huh?" because the distribution of the population is certainly not 69.2 \pm 0.012 in. But then I read the headings again, and it's the standard deviation of the mean. With an N=1000, 0.012 becomes 0.012 * sqrt(1000) \approx 0.4, which sounds a bit more accurate.
But boy oh boy, 0.012 sure looks like a much tighter measurement than 0.4. And 0.4 is the sigma for the actual population.
No, it's called "Biologists and Med Students Don't Like Math" and never learned about SNR, P_{fa}, P_d, shot noise, and all those lovely topics covered in many good two-semester probability/statistical inference courses at many reputable institutions.
Not that I'm begrudging anyone success or prosperity, but it seems that the way the Chinese are operating in this respect screws everyone over in the long run. The buy-and-clone mentality can put Western manufacturers out of business, and since all the Chinese companies did was reverse-engineer, they don't really build up internal expertise of the level they just quashed. And just like Embrace,Extend,Extinguish, in the long run innovation doesn't happen as fast as it might have.
That's why you don't read just one news source. I regularly read Fox, NYT, CNN, BBC, and whatever random crap comes up on Google News and RealClearPolitics just to guard against any segment of these guys deliberately ignoring something embarrassing about a left-wing pol, a right-wing pol, or something apolitical like this.
Let's be clear. There's 310 million Americans. You can't do anything by committee of 310 million. You couldn't even do anything by committee of ~3 million, which is why the Founding Fathers wisely decided to have those millions elect official representatives who are entrusted to hold the executive departments of the federal government accountable. That's why congressmen get to sit in on briefings Joe Sixpack doesn't.
The answer is to always be on the case of elected officials to make sure bad stuff doesn't happen. Maybe that doesn't happen enough (given that many (most?) House members have racked up 20+ years in office, but that's where it's gotta start. Airing out confidential information, just because, doesn't make government better, it just adds more noise to drown out the bit of common sense that might be in the air.
0. Earn a doctorate in math/science/engineering
1. Land a cushy defense job
2. ???
3. Something (like a small island or country) might have gotten blown up along the way, but profit!!!
My favorite bit is the fellow quoted in the article who laments that he doesn't think there's a solution.
Not to be too arrogant, but anyone who knows basic geometry and how to stick two lines of code together should be able to at least imagine that there exists a solution. Is there really such a wide gap in the Two Cultures that not only does the other side not know how to fix a software problem, they can't even fathom that a fix is possible?
This reminds me of the Cargo Cult mentality mentioned in an article quoted a few days ago, here, where the view of the cult is that technology is an immutable force of nature, not a tool mastered by man, and the idea that man can wield it is so foreign as to be unthinkable.
You'd think that university administrators in the US and their ilk would be advanced beyond that. I feel embarrassed for the poor dumb bastard.
The awesome part will be when you kid gets into 8th, 9th grade and finds that he's in classes with kids who still haven't mastered subtraction or multiplication.
Many moons ago (that is, more than 2), way back when I was in the third grade in the Philadelphia public schools, the teacher asked the class, "who can think of a way to make 25?"
Some kids piped up, "11+14."
Another said "20+5"
I raised my hand and said "5 x 5"
And the teacher answered, "that's good, but some people here don't know about multiplication yet, so we'll just skip over that answer."
And it took me damn near 20 years of elementary, middle, and high school (in a better school district) and an engineering degree or two before I found myself around people for whom basic mathematical manipulations were not a major stumbling block.
Which leads me to conjecture that because mathematics and mathematical reasoning is not a natural behavior (see first link), while talking and telling stories is, there will always be this gap between the people who can get it, and those who can't, and those who can be taught it and those who can't. What to do? Exactly what you've been doing. If the gap is in the genes, then there's no one more qualified to pass on the spark than the parents.
On the other hand, a missile with a man or two within, controlling flight by graphitics,
would be lighter, more mobile, more intelligent. It would give us a lead that might well
mean the margin of victory. Besides which, gentlemen, the exigencies of war compel us
to remember one thing. A man is much more dispensable than a computer. Manned
missiles could be launched in numbers and under circumstances that no good general
would care to undertake as far as computer-directed missiles are concerned
which gets me thinking. How much longer do we have to wait until we can wear spell-check goggles
that automatically highlight wrong spellings in what we look at, or suggest correct spelling in what we
type or handwrite, the way the google search box does? At what point will going in
without a spellchecker be as silly as trying to drive without your glasses on?
It's probably not the power bus directly. Most battery packs and I would assume these devices with exposed leads have built-in current limiters that shut down on an overcurrent condition. I recall seeing a news story a while back about knock-off cell phone batteries that didn't have this circuitry being a fire/safety hazard.
Well, it's a case of known vs unknown unknowns. To me and the people I work with, programming is a tool used to get Machine A to perform Action B with Precision C. An unknown unknown becomes evident very quickly.
But the analogy I like to use is this: the average human can get by without running water, electricity, or even the wheel. And if he does it long enough or never learns about these things, he doesn't even know he's missing anything. Is it any surprise that people can get by with crappy coding born of an incomplete theoretical background without knowing that there's something missing?
but so is arithmetic, (high school) algebra, geometry, calculus, complex analysis, and so forth. And yet there are high schools where calculus is not taught, and the fact that logarithms are the inverse of exponentials never gets mentioned.
The problem is that people who understand what this stuff is learn it, use it, and teach it to their children, but the people who don't understand what this stuff is have and unknown unknown and don't even realize that they're missing anything. Short of spending zillions to buy airtime in the middle of American Idol, I don't think there's any way to make people understand what they're missing.
Sure. Computer scientists are (with exceptions) the ones who can fill a whiteboard full of beautiful abstractions but can't code their way out of a paper bag. Real programmers on the other hand...
Well, yeah. But it's a university, not an R&D lab. And the semi-implicit assumption is that once I'm done cutting my teeth on other people's money, I'll get a real defense job and earn my keep. Works pretty well. The tricky bit (on both ends) is figuring out how to avoid the complete idiots.
Slashdot Mirror
You have it backwards... if we don't write new software, who's going to need more RAM/speed/space?
Gullible idiots, of course.
My solution is to put a big steel bracket around my brake pedal that would take more than a few minutes of cutting to get through before you can drive away. It doesn't prevent a determined party from taking the car if he really wants to, but it's a layer of actual physical security that prevents someone from duplicating a software key and riding away, just like it prevented someone from picking the lock and hotwiring the ignition and riding away on an older model.
The standard error for a Bernoulli trial (right or wrong are the only outcomes) looks like sqrt(p(1-p)/N). If N=100 (which looks like the trial he ran), then for p anywhere between 0.25 and 0.75 (even though we know it's really 0.5), sigma >= 3.3%.
This means that if you ran the N=100 experiment 100 times, about 33 or those trials will yield you a value of correct between 46.6% and 53.3%, 33 will be between 40 and 46.6%, another 33 will be between 53.3 and 60%, and the last one might be anywhere, but probably closer to 40 or 60.
This is what is known as small number statistics. And I learned it (correctly) in a graduate level engineering course which required several years worth of advanced mathematics in undergrad. My experience is that medical types usually don't take that route, and therefor they make asses of themselves getting excited over random noise.
Are human beings really that incapable of mental abstraction that they need some anthropomorphic gizmo telling them the same thing a normal machine can do, but with 10x the development cost?
My unscientific sampling two of Chinese engineering students rotating through at a good American engineering school before going back suggests to me that it's a larger value of ``soon'' than you might expect given the outward show of progress. Yes, they're dumping zillions into building up institutions, but in trying to catch up on 200+ years of Western experience in under a generation, there's a certain something missing. They're nowhere near cargo-cult level, but they're closer to it than we are in the west, hence my original comment.
Google "average height adult male" and somewhere at the top of the list will be a webpage from the NIH with lots of tables breaking down height and weight by age group.
The columns are:
Age, Subject N, Mean Height, Std. Dev. of Mean, etc...
Now, just glancing at the tables, the avg height for a 20-30yr old male was 69.2 inches. But the next column was 0.012in. I was like "huh?" because the distribution of the population is certainly not 69.2 \pm 0.012 in. But then I read the headings again, and it's the standard deviation of the mean. With an N=1000, 0.012 becomes 0.012 * sqrt(1000) \approx 0.4, which sounds a bit more accurate.
But boy oh boy, 0.012 sure looks like a much tighter measurement than 0.4. And 0.4 is the sigma for the actual population.
No, it's called "Biologists and Med Students Don't Like Math" and never learned about SNR, P_{fa}, P_d, shot noise, and all those lovely topics covered in many good two-semester probability/statistical inference courses at many reputable institutions.
Not that I'm begrudging anyone success or prosperity, but it seems that the way the Chinese are operating in this respect screws everyone over in the long run. The buy-and-clone mentality can put Western manufacturers out of business, and since all the Chinese companies did was reverse-engineer, they don't really build up internal expertise of the level they just quashed. And just like Embrace,Extend,Extinguish, in the long run innovation doesn't happen as fast as it might have.
That's why you don't read just one news source. I regularly read Fox, NYT, CNN, BBC, and whatever random crap comes up on Google News and RealClearPolitics just to guard against any segment of these guys deliberately ignoring something embarrassing about a left-wing pol, a right-wing pol, or something apolitical like this.
Let's be clear. There's 310 million Americans. You can't do anything by committee of 310 million. You couldn't even do anything by committee of ~3 million, which is why the Founding Fathers wisely decided to have those millions elect official representatives who are entrusted to hold the executive departments of the federal government accountable. That's why congressmen get to sit in on briefings Joe Sixpack doesn't.
The answer is to always be on the case of elected officials to make sure bad stuff doesn't happen. Maybe that doesn't happen enough (given that many (most?) House members have racked up 20+ years in office, but that's where it's gotta start. Airing out confidential information, just because, doesn't make government better, it just adds more noise to drown out the bit of common sense that might be in the air.
0. Earn a doctorate in math/science/engineering
1. Land a cushy defense job
2. ???
3. Something (like a small island or country) might have gotten blown up along the way, but profit!!!
Or principles or caterpillers?
See, see!? Some tablets are more equal than others!!1!one!eleven
My favorite bit is the fellow quoted in the article who laments that he doesn't think there's a solution.
Not to be too arrogant, but anyone who knows basic geometry and how to stick two lines of code together should be able to at least imagine that there exists a solution. Is there really such a wide gap in the Two Cultures that not only does the other side not know how to fix a software problem, they can't even fathom that a fix is possible?
This reminds me of the Cargo Cult mentality mentioned in an article quoted a few days ago, here, where the view of the cult is that technology is an immutable force of nature, not a tool mastered by man, and the idea that man can wield it is so foreign as to be unthinkable.
You'd think that university administrators in the US and their ilk would be advanced beyond that. I feel embarrassed for the poor dumb bastard.
The awesome part will be when you kid gets into 8th, 9th grade and finds that he's in classes with kids who still haven't mastered subtraction or multiplication.
Many moons ago (that is, more than 2), way back when I was in the third grade in the Philadelphia public schools, the teacher asked the class, "who can think of a way to make 25?"
Some kids piped up, "11+14."
Another said "20+5"
I raised my hand and said "5 x 5"
And the teacher answered, "that's good, but some people here don't know about multiplication yet, so we'll just skip over that answer."
And it took me damn near 20 years of elementary, middle, and high school (in a better school district) and an engineering degree or two before I found myself around people for whom basic mathematical manipulations were not a major stumbling block.
Which leads me to conjecture that because mathematics and mathematical reasoning is not a natural behavior (see first link), while talking and telling stories is, there will always be this gap between the people who can get it, and those who can't, and those who can be taught it and those who can't. What to do? Exactly what you've been doing. If the gap is in the genes, then there's no one more qualified to pass on the spark than the parents.
Dictatorial my ass. Short of making kinda credible threats of physical violence, you can say and do whatever you want.
Clearly tomatoes. Rainbows can't figure out what color they're supposed to be, but tomatoes are resoundingly red. Unless they're green.
On the other hand, a missile with a man or two within, controlling flight by graphitics, would be lighter, more mobile, more intelligent. It would give us a lead that might well mean the margin of victory. Besides which, gentlemen, the exigencies of war compel us to remember one thing. A man is much more dispensable than a computer. Manned missiles could be launched in numbers and under circumstances that no good general would care to undertake as far as computer-directed missiles are concerned
which gets me thinking. How much longer do we have to wait until we can wear spell-check goggles that automatically highlight wrong spellings in what we look at, or suggest correct spelling in what we type or handwrite, the way the google search box does? At what point will going in without a spellchecker be as silly as trying to drive without your glasses on?
It's probably not the power bus directly. Most battery packs and I would assume these devices with exposed leads have built-in current limiters that shut down on an overcurrent condition. I recall seeing a news story a while back about knock-off cell phone batteries that didn't have this circuitry being a fire/safety hazard.
Careful inspection of the picture leads me to conclude that the air gap between the negative probe and the hook probably does have a 2 MOhm resistance
a malfunction in a high tech device that actually can be fixed with duct tape
Well, it's a case of known vs unknown unknowns. To me and the people I work with, programming is a tool used to get Machine A to perform Action B with Precision C. An unknown unknown becomes evident very quickly.
But the analogy I like to use is this: the average human can get by without running water, electricity, or even the wheel. And if he does it long enough or never learns about these things, he doesn't even know he's missing anything. Is it any surprise that people can get by with crappy coding born of an incomplete theoretical background without knowing that there's something missing?
but so is arithmetic, (high school) algebra, geometry, calculus, complex analysis, and so forth. And yet there are high schools where calculus is not taught, and the fact that logarithms are the inverse of exponentials never gets mentioned.
The problem is that people who understand what this stuff is learn it, use it, and teach it to their children, but the people who don't understand what this stuff is have and unknown unknown and don't even realize that they're missing anything. Short of spending zillions to buy airtime in the middle of American Idol, I don't think there's any way to make people understand what they're missing.
Sure. Computer scientists are (with exceptions) the ones who can fill a whiteboard full of beautiful abstractions but can't code their way out of a paper bag. Real programmers on the other hand...
Well, yeah. But it's a university, not an R&D lab. And the semi-implicit assumption is that once I'm done cutting my teeth on other people's money, I'll get a real defense job and earn my keep. Works pretty well. The tricky bit (on both ends) is figuring out how to avoid the complete idiots.