It is difficult to explain the fact that the changes emerge around age 13 as enculturation. Do societal expectations change around then, to favor more differences between the sexes?
It's not only not impossible, but it's pretty much always possible. You just have to think like someone who chases funding.
Everyone who reviews proposals knows the future is uncertain, so they don't currently expect a proposal to accurately predict, say, how someone's research would benefit math education. The key is to explain how what you're proposing could plausibly help. Doing it well comes down to having a reasonable story, having good salesmanship, and wordsmithing.
The new requirements seem very broadly applicable. For example, I could twist scientific literacy, promotion of scientific progress, and possibly national defense into justifying the grant proposal I'm currently working on. "Scientific progress" in particular would be very easy. I expect it would be similarly easy for any other academic who expects to publish at leat one paper on research that he or she intends to support by an NSF grant.
So this probably wouldn't change anything, except to require another section in every proposal, which would just waste everyone's time. It would save exacly zero dollars, and cost a few for every proposal just by a naive conversion from time to money. There are also one-time costs. The only possible way this could save money is by slowing down the overall process.
While I'm railing, I should also mention that active researchers review other people's NSF proposals. Adding another requirement takes time they could use to, I dunno, do useful research?
Everyone who chases funding knows how to play the game. Adding rules won't keep them from getting money, and it'll cost time.
No, it's not. It's Oracle demonstrating that they fail to grasp the idea of a loss leader. They have no idea how to leverage a free product, can't stand the idea of a transaction without dollars changing hands, and as a result, continually piss everyone off.
...responding to the player's own body language rather than mathematical rules.
I've done a lot of machine learning. It's very much mathematical rules. The inferred rules just happen to be so complicated that it's often hard to recognize them as such.
Maybe the submitter meant "hand-crafted decision tree".
(Hint: it is not called lambda CALCULUS because it sounds fancy)
It's because it's a calculus: a method of calculating things. In this case, the things are higher-order functions, and the method is beta-reduction.
Your other points are well-taken, however.
My problem with the question and all the answers so far is that they are focusing on generalities. It's true that most programmers won't need calculus. However, we desperately need the 2% that do, because scientists from other fields don't get taught how to program. Oh, they pick it up on the job. They have to take one programming course from the CS department. But they rarely get rigorous training in software engineering that they need to make reliable stuff.
If there are environmental factors, they're slight. It's as heritable as height.
Plenty of things can cause the same symptoms in the short term without the neurological condition. Examples are the death of a parent (the emotional pain is, among other things, highly distracting), lack of sleep, and malnutrition. Yes, staring at a screen all day and experiencing nothing but rapid, small rewards can cause an otherwise healthy kid to find other things less rewarding. But I've read a lot about ADHD, and I've never seen anything conclusive that says such things can give a child the actual disorder.
Each of our states is as big as a country. How many Europeans could name the states surrounding the one I live in? Not many.
In this country, a Texan is as close to a foreigner as most people are likely to meet.
When I lived in Scotland, I was asked at least twice a month whether I knew Pamela Anderson. (No, by the way.) The kids there had no idea just how big this place is. Friggin' huge, is what it is. Like Sputnik.
Don't forget that we don't just use our space for candy. There's also Unicode, which is often done with 32-bit characters, so 4x on your strings right there. Along with that is internationalization, so your strings get duplicated 10x or whatever. Further, every 64-bit pointer takes 4x the space of a 16-bit pointer. We also use a LOT more pointers, because we have this crazy, correct idea that fixed-length data structures are evil in multiple ways. We also tend to write in imperative object-oriented or quasi-pure functional style, which are both a bit bloaty, because with those styles it's a lot easier to write extensible code. (With OO, it's easier to extend the kinds of objects on the system; with functional, easier to extend the kinds of operations.)
So yeah, it's not just candy or foregoing optimization. We've been moving toward making it easy for every programmer, regardless of skill, to write code that, if not correct, will at least not blow up your computer, and do a lot of really cool stuff.
Hanging it under is far more practical. You can tear if it off with one hand very easily without having the paper unspool 7 yards of itself onto the floor.
For those of us fit enough to breed, there's another great reason to hang it under. A toddler will often unspool toilet paper with an overhand whack, but almost never underhand.
Yes, and you'll actually get his own account of the main thrust of his speech. Turns out it's not as silly and demanding as the reports of it make it out to be.
He's suggesting business and open-source learn each other's way of doing things and meet in the middle. There are competing interests, yes. There always will be.
He's telling us what difficulties business has with open-source and vice-versa from the perspective of a previously all-closed business that wants badly for everyone to work together. This is valuable information, whether we agree with him or not.
but we are not yet ready to play by the rules; but this needs to work the other way round too
So you're not yet ready to play by our rules, but you want us to play by your rules so that you have an opportunity to take advantage of the work we produce and provide to you for free (beer/speech); when the only stipulation we have is that you provide it back for free? No, not just provide it back for free. This is the man behind the maemo platform, the as-open-source-as-we-could-get-it platform for Nokia's internet tablets. By his estimation the default image is over 2/3's free software, and it's not just stuff they grabbed online and plonked into it gratis. Nokia hired open-source developers actively contribute to open-source projects. (This was before they acquired Trolltech.) They opened as much code as they could on the maemo platform without running into hardware IP issues and other things that made Jaaksi's higher-ups queasy.
Am I saying it's perfect right now? No. But cut the guy some slack.
Also, both reports of his talk I've seen have misrepresented what he said. He also talked about businesses needing to learn how to do things the open-source way:
Companies like Nokia need to learn the open source way of working. This means not only fulfilling the letter of GPL, LGPL etc. but also the spirit. In my mind this means integrating the corporate work with the open source community, participating, contributing back the code, building the code in open projects and not only releasing it when mandatory, not forking, etc. Open source is a very effective way to create software together with others; together with other individuals and other companies. This is something that the corporate must learn to really benefit from open source.
It's not just lip service. I've read elsewhere (can't remember where) that he genuinely regards the open-source development model as generally superior.
A fairer article would have titled the article "Nokia: Business and Open-Source Should Work to Understand Each Other and Compromise". But that doesn't generate traffic, now, does it?
Another good recent read was "God Is Not Great" by Christopher Hitchens.
Unfortunately, way too many of the facts he presents in his book, well, aren't. A very, very fair view from a man with a PhD in New Testament scholarship:
There's no reason to believe that he was any more careful with his facts about any other religions. This isn't to say his philosophical points aren't worth knowing, but if you're looking for something truthful about religion, you may want to look elsewhere.
Cantaloupe is harder to tell by scent, but can be done that way as well. If you have a ripe and unripe one to compare against each other, you can tell pretty easily.
With cantaloupe you can also tell by color. Look inside the gaps of the reticulated skin: green are unripe and light brown are ripe. It should also be fairly firm, but a tiny bit of softness doesn't hurt. It should be softer at the stem end.
I've also read that it's easier to pick up the scent at the stem end.
I'm writing this on my N800 right now, using the on-screen fat-finger keyboard (I can go about 40 WPM on it) posting using a Mozilla-based browser. (It renders Slashdot perfectly, too.) Runs an apt-based Linux, has SSH client and server, 800x480 screen, and is currently $220 at Amazon. It does remote X (with some hacks using xvkbd), rdesktop and VNC. Its only disqualifying attribute is that it's not a phone.
I can answer these questions on the average for Adderall and Dexedrine (dextroamphetamines) and Ritalin (methylphenidate).
What happens to your cognition once you stop taking it, after you've gotten used to taking it? Do you get a tolerance, so you not only need higher doses for a smarts boost, but you also just return to your base performance after getting used to it?
Tolerance is rarely an issue with the low doses given to treat ADHD. A couple of back-of-the-envelope calculations suggest that an average dose of Adderall (20mg) is about 1/10 an average "first hit" of meth or cocaine. Prescription medications are also designed to metabolize much more slowly than recreational drugs.
Tolerance mainly results from neurons being overexcited and altering receptor sites in response. (This is in fact how caffeine tolerance develops.) People who take these medications under a doctor's care are generally not overstimulated. In fact, with ADHD, because medication corrects understimulation it's usually not an issue at all.
I would be more worried about tolerance if the general population started on them, though.
What's the withdrawal like?
Usually a little mentally fuzzier than before medication and maybe a bit crankier. It lasts about half an hour to an hour. People report that Ritalin and Dexedrine have "rougher edges" than Adderall, which makes sense since Adderall is a mixture of amphetamine salts that metabolize at different rates.
I suspect that maybe the many kids given Ritalin while growing up learn to depend on it for their baseline. When they outgrow their "hyperactivity" (AKA "childhood"), they quit the drugs, and sink into an unfamiliar dullness in which they can't think at their previous baseline without the artificial stimulation.
If they don't outgrow ADHD and they need medication to function, they shouldn't stop.
However, often the medication does have a lasting effect, though not one that people with "OMG DRUGGIES!!!" in mind would predict. It can train your mind to mimic the patterns it gets used to while on medication. People will often lower their dosage over time, and some quit altogether. I'm not aware of anyone needing more until they're a prescription crack-head. Both anecdotal evidence and the literature (peer-reviewed studies) support this.
It also tends to train behavior. While on medication, functional behavior is much easier, and people who learn to function effectively while on medication have an easier time off of it than they did before medication.
Again, I wouldn't apply this to the general population, just to people who use medication to treat neurological problems.
And how much do they just get burned out from the steady drugging?
They only do if the dose is too high. The beautiful thing about stimulants at these dosages is that their cognitive effects don't last into the next day, except for the gradual effects I mentioned.
They're not perfect, and won't catch everything, but they do work.
More like can't catch everything. Or they'll "catch" perfectly fine things. Anybody remember Rice's Theorem from computability theory?
(What? You didn't think you'd ever need computability theory again?)
It's a generalization of the Halting Problem. The basic idea is that there is no algorithm for any non-trivial property ("does this program halt on every input?" "does this array get written out-of-bounds?" "does this value overflow?") that can verify every program. All such properties are undecidable. This means that either your static checker 1) must be wrong sometimes, or 2) doesn't always halt. Not halting is generally seen as bad, so strong typing systems halt with false positives and other static checkers tend to halt with false positives and negatives.
As such, they're good for advice only.
Catching errors dynamically is also incomplete, but for more obvious reasons.
For $40 more, you can snag yourself the Nokia N800 Iternet Tablet. The N810 drove its price way down, and the only differences are a built-in GPS, slide-out keyboard, and a 2GB SD card. The rest of the hardware is identical, and you can flash the latest N810 OS on the N800. The thing is highly hackable, with as much open-source software as Ari Jaaksi, Nokia's open source director, could get them to embrace (about 2/3 of the base system). With a very bright 800x480 display, Firefox and mplayer, it renders everything almost perfectly. It's got a thriving open-source community behind it with a bunch of apt repositories and ports. It's also the nicest e-book reader I've ever used.
I don't work for Nokia. I just love mine.:)
Consumer-oriented reviews tend to emphasize its lack of pre-installed PIM apps and synchronization, but that's not a problem for your average technophile.
To give you an idea of how hackable it is, I hacked the init scripts to set up swap and mount my home directory from an SD card's ext2 partition. I SSH into it when I want to do this kind of stuff.
Parts of the hardware (and thus some of the drivers) aren't open. If you're a purist, this might put you off. Which brings us back on topic: the Chumby is completely open. Maybe this'll push Nokia to open more. Ari Jaaksi has even said that the open source software on the N800 is of far better quality than the in-house stuff - it's just convincing the suits that embracing it is a good idea that's difficult.
And here's another two sales lost on the DM of the Rings guy. (Plus a hilarious comic.) Is there some public list we could all sign stating our refusal to buy these games?
Too bad I lost my mod points yesterday. This is the kind of thing people actually come to Slashdot for. I'll just have to try to contribute instead.
Here's some further detail for those interested: the |1> and |0> qubits are actually vectors of probabilities. (Well, probability "amplitudes". More on that later.) The |0> bit means [1 0] and the |1> bit means [0 1]. The "|.>" notation is a bit of convenient shorthand.
If you have two qubits, you'd represent them as |00>, meaning [1 0 0 0]. (That's four possibilities for the qubits, and all the probability mass on the first: both off.) |01> means [0 1 0 0], |000> means [1 0 0 0 0 0 0 0], and so on. Note the exponential growth.
A quantum gate is nothing more than an operator of the same type that governs all discrete quantum system evolution: a unitary matrix. Think of a rotation matrix of rank 2**(number-of-bits), but in complex space. It's got to be some kind of rotation - it must preserve length - to preserve the property that the qubit states and combined qubit states are probability (amplitude) distributions.
A "square root of NOT", IIRC, is an operator (rotation) that turns [1 0] (or |0>) into [sqrt(1/2) -sqrt(1/2)]. Do it again, and you get [0 1]. Again, and you get [-sqrt(1/2) sqrt(1/2)], and again yields the original [1 0]. (I may have some signs wrong.)
The reason this cycle works at all is that the states aren't probabilities per se, but sort of square roots of probabilities, which allows them to keep extra information. This is called "phase". Much of the exciting weirdness of computing with quantum gates is that phase isn't strictly real, but in general has imaginary components.
The other exciting weirdness is of the massively parallel sort. If I do a computation on [sqrt(1/2) -sqrt(1/2)], it's sort of like doing the same computation on [1 0] and [0 1] in parallel. The tricky part is that measuring the outcome restricts me to just one of the results! One way to express the dilemma is that I can compute an answer for every possible input simultaneously (which would be great for solving NP problems), but that I can't easily select the right answer.
Another way to express it is to say that the cat is in a superposition of dead/alive, which will localize when I observe the poor beast.:)
But I was making a very different point in my OP. The particular baseball stat is a "state of nature". It happened. If we assume that the universe is deterministic, then it's nonsensical to assign a probability other than 1.0 to that event. Any model that produces an answer other than 1.0 for that event is probably replacing a variable that in the real world has a fixed value, with a random variable.
Funny, we just talked about that a few days ago. The crux is that you're not necessarily right: it depends on whether you're approaching the problem with a frequentist ("frequency of occurrence") or Bayesian ("degree of belief") interpretation of probability. The Bayesian view is more widely applicable (for instance, this study) and easier to model in (because you can easily compose distributions that don't integrate nicely).
Stodgily advocating a single definition and purpose of "probability" is cute and all, in a "get off my lawn" sort of way. But it's a little irksome to those of us who use Bayesian models to solve AI / machine learning / computer vision problems - among other folks.
Still, feel free to tell everyone you know that a coin flip that has already happened but that you haven't observed the outcome of has heads probabilities of 0.0 and 1.0, period. That won't help anybody who has to compute an expected value over outcomes in similar situations, though. To do that, you need a framework in which "How likely was that?" is a valid question.
Also, the assumption that the universe is deterministic is fairly iffy, especially given our current understanding of quantum mechanics, wherein the only true state of a small-scale object is actually a probability distribution over states.
I probably sound a lot pissier than I am right now.
Since you sound pretty new to programming in general, I'd spend a few nights a week just messing around....
Exactly. Make it an uber-hobby! The best employers know that their best developers do their work as a hobby. Who would you rather hire? Somebody who knows his stuff because it lets him pay the bills, or somebody who knows his stuff because he loves doing it?
I've got an "Interesting Side Projects" section on my resume that everyone I've interviewed with has been impressed with. (I even get compliments on its existence, not just on its contents. Prospective employers appreciate that kind of thing.) At one place I interviewed at, each interviewer (six in all) picked a different side project to ask me about!
There are few professions for which you can demonstrate willingness and love of learning in such a concrete way. Software development is one of them. Take advantage of it!
My mod points expired recently, so could someone mod this up? I do machine learning and computer vision with Bayesian statistics, and the above poster is spot-on. The GP sounds like a frequentist trying to regain control over statistical vocabulary.
FWIW, the frequentists can keep "confidence interval". We don't want to sully our theoretically sound vocabulary with its filthy connotations.:p But "probability" is something we'll lay uncompromising claim to, however much detractors say that subjective probabilities don't count. If they don't count, how else would anyone model something like "belief" in a well-grounded way?
Slashdot Mirror
It is difficult to explain the fact that the changes emerge around age 13 as enculturation. Do societal expectations change around then, to favor more differences between the sexes?
It's not only not impossible, but it's pretty much always possible. You just have to think like someone who chases funding.
Everyone who reviews proposals knows the future is uncertain, so they don't currently expect a proposal to accurately predict, say, how someone's research would benefit math education. The key is to explain how what you're proposing could plausibly help. Doing it well comes down to having a reasonable story, having good salesmanship, and wordsmithing.
The new requirements seem very broadly applicable. For example, I could twist scientific literacy, promotion of scientific progress, and possibly national defense into justifying the grant proposal I'm currently working on. "Scientific progress" in particular would be very easy. I expect it would be similarly easy for any other academic who expects to publish at leat one paper on research that he or she intends to support by an NSF grant.
So this probably wouldn't change anything, except to require another section in every proposal, which would just waste everyone's time. It would save exacly zero dollars, and cost a few for every proposal just by a naive conversion from time to money. There are also one-time costs. The only possible way this could save money is by slowing down the overall process.
While I'm railing, I should also mention that active researchers review other people's NSF proposals. Adding another requirement takes time they could use to, I dunno, do useful research?
Everyone who chases funding knows how to play the game. Adding rules won't keep them from getting money, and it'll cost time.
That's capitalism at work.
No, it's not. It's Oracle demonstrating that they fail to grasp the idea of a loss leader. They have no idea how to leverage a free product, can't stand the idea of a transaction without dollars changing hands, and as a result, continually piss everyone off.
I've done a lot of machine learning. It's very much mathematical rules. The inferred rules just happen to be so complicated that it's often hard to recognize them as such.
Maybe the submitter meant "hand-crafted decision tree".
(Hint: it is not called lambda CALCULUS because it sounds fancy)
It's because it's a calculus: a method of calculating things. In this case, the things are higher-order functions, and the method is beta-reduction.
Your other points are well-taken, however.
My problem with the question and all the answers so far is that they are focusing on generalities. It's true that most programmers won't need calculus. However, we desperately need the 2% that do, because scientists from other fields don't get taught how to program. Oh, they pick it up on the job. They have to take one programming course from the CS department. But they rarely get rigorous training in software engineering that they need to make reliable stuff.
If there are environmental factors, they're slight. It's as heritable as height.
Plenty of things can cause the same symptoms in the short term without the neurological condition. Examples are the death of a parent (the emotional pain is, among other things, highly distracting), lack of sleep, and malnutrition. Yes, staring at a screen all day and experiencing nothing but rapid, small rewards can cause an otherwise healthy kid to find other things less rewarding. But I've read a lot about ADHD, and I've never seen anything conclusive that says such things can give a child the actual disorder.
Bah. This is human nature, not just Americans.
Each of our states is as big as a country. How many Europeans could name the states surrounding the one I live in? Not many.
In this country, a Texan is as close to a foreigner as most people are likely to meet.
When I lived in Scotland, I was asked at least twice a month whether I knew Pamela Anderson. (No, by the way.) The kids there had no idea just how big this place is. Friggin' huge, is what it is. Like Sputnik.
Don't forget that we don't just use our space for candy. There's also Unicode, which is often done with 32-bit characters, so 4x on your strings right there. Along with that is internationalization, so your strings get duplicated 10x or whatever. Further, every 64-bit pointer takes 4x the space of a 16-bit pointer. We also use a LOT more pointers, because we have this crazy, correct idea that fixed-length data structures are evil in multiple ways. We also tend to write in imperative object-oriented or quasi-pure functional style, which are both a bit bloaty, because with those styles it's a lot easier to write extensible code. (With OO, it's easier to extend the kinds of objects on the system; with functional, easier to extend the kinds of operations.)
So yeah, it's not just candy or foregoing optimization. We've been moving toward making it easy for every programmer, regardless of skill, to write code that, if not correct, will at least not blow up your computer, and do a lot of really cool stuff.
For those of us fit enough to breed, there's another great reason to hang it under. A toddler will often unspool toilet paper with an overhand whack, but almost never underhand.
Yes, and you'll actually get his own account of the main thrust of his speech. Turns out it's not as silly and demanding as the reports of it make it out to be.
He's suggesting business and open-source learn each other's way of doing things and meet in the middle. There are competing interests, yes. There always will be.
He's telling us what difficulties business has with open-source and vice-versa from the perspective of a previously all-closed business that wants badly for everyone to work together. This is valuable information, whether we agree with him or not.
Where you actually there for his speech? Because Jaaksi himself differs with you on the accuracy of the reporter's account.
Am I saying it's perfect right now? No. But cut the guy some slack.
Also, both reports of his talk I've seen have misrepresented what he said. He also talked about businesses needing to learn how to do things the open-source way:
It's not just lip service. I've read elsewhere (can't remember where) that he genuinely regards the open-source development model as generally superior.
A fairer article would have titled the article "Nokia: Business and Open-Source Should Work to Understand Each Other and Compromise". But that doesn't generate traffic, now, does it?
Unfortunately, way too many of the facts he presents in his book, well, aren't. A very, very fair view from a man with a PhD in New Testament scholarship:
Christian review
A slightly snarkier take from an expert on Mormonism (who also happens to have a PhD in Near Eastern Languages and Cultures):
Mormon review
There's no reason to believe that he was any more careful with his facts about any other religions. This isn't to say his philosophical points aren't worth knowing, but if you're looking for something truthful about religion, you may want to look elsewhere.
With cantaloupe you can also tell by color. Look inside the gaps of the reticulated skin: green are unripe and light brown are ripe. It should also be fairly firm, but a tiny bit of softness doesn't hurt. It should be softer at the stem end.
I've also read that it's easier to pick up the scent at the stem end.
I'm writing this on my N800 right now, using the on-screen fat-finger keyboard (I can go about 40 WPM on it) posting using a Mozilla-based browser. (It renders Slashdot perfectly, too.) Runs an apt-based Linux, has SSH client and server, 800x480 screen, and is currently $220 at Amazon. It does remote X (with some hacks using xvkbd), rdesktop and VNC. Its only disqualifying attribute is that it's not a phone.
Tolerance is rarely an issue with the low doses given to treat ADHD. A couple of back-of-the-envelope calculations suggest that an average dose of Adderall (20mg) is about 1/10 an average "first hit" of meth or cocaine. Prescription medications are also designed to metabolize much more slowly than recreational drugs.
Tolerance mainly results from neurons being overexcited and altering receptor sites in response. (This is in fact how caffeine tolerance develops.) People who take these medications under a doctor's care are generally not overstimulated. In fact, with ADHD, because medication corrects understimulation it's usually not an issue at all.
I would be more worried about tolerance if the general population started on them, though.
Usually a little mentally fuzzier than before medication and maybe a bit crankier. It lasts about half an hour to an hour. People report that Ritalin and Dexedrine have "rougher edges" than Adderall, which makes sense since Adderall is a mixture of amphetamine salts that metabolize at different rates.
If they don't outgrow ADHD and they need medication to function, they shouldn't stop.
However, often the medication does have a lasting effect, though not one that people with "OMG DRUGGIES!!!" in mind would predict. It can train your mind to mimic the patterns it gets used to while on medication. People will often lower their dosage over time, and some quit altogether. I'm not aware of anyone needing more until they're a prescription crack-head. Both anecdotal evidence and the literature (peer-reviewed studies) support this.
It also tends to train behavior. While on medication, functional behavior is much easier, and people who learn to function effectively while on medication have an easier time off of it than they did before medication.
Again, I wouldn't apply this to the general population, just to people who use medication to treat neurological problems.
They only do if the dose is too high. The beautiful thing about stimulants at these dosages is that their cognitive effects don't last into the next day, except for the gradual effects I mentioned.
More like can't catch everything. Or they'll "catch" perfectly fine things. Anybody remember Rice's Theorem from computability theory?
(What? You didn't think you'd ever need computability theory again?)
It's a generalization of the Halting Problem. The basic idea is that there is no algorithm for any non-trivial property ("does this program halt on every input?" "does this array get written out-of-bounds?" "does this value overflow?") that can verify every program. All such properties are undecidable. This means that either your static checker 1) must be wrong sometimes, or 2) doesn't always halt. Not halting is generally seen as bad, so strong typing systems halt with false positives and other static checkers tend to halt with false positives and negatives.
As such, they're good for advice only.
Catching errors dynamically is also incomplete, but for more obvious reasons.
Don't forget the awesome Marry Poppins horror ("Scary Mary") trailer.
Hide your children!
For $40 more, you can snag yourself the Nokia N800 Iternet Tablet. The N810 drove its price way down, and the only differences are a built-in GPS, slide-out keyboard, and a 2GB SD card. The rest of the hardware is identical, and you can flash the latest N810 OS on the N800. The thing is highly hackable, with as much open-source software as Ari Jaaksi, Nokia's open source director, could get them to embrace (about 2/3 of the base system). With a very bright 800x480 display, Firefox and mplayer, it renders everything almost perfectly. It's got a thriving open-source community behind it with a bunch of apt repositories and ports. It's also the nicest e-book reader I've ever used.
:)
I don't work for Nokia. I just love mine.
Consumer-oriented reviews tend to emphasize its lack of pre-installed PIM apps and synchronization, but that's not a problem for your average technophile.
To give you an idea of how hackable it is, I hacked the init scripts to set up swap and mount my home directory from an SD card's ext2 partition. I SSH into it when I want to do this kind of stuff.
Parts of the hardware (and thus some of the drivers) aren't open. If you're a purist, this might put you off. Which brings us back on topic: the Chumby is completely open. Maybe this'll push Nokia to open more. Ari Jaaksi has even said that the open source software on the N800 is of far better quality than the in-house stuff - it's just convincing the suits that embracing it is a good idea that's difficult.
And here's another two sales lost on the DM of the Rings guy. (Plus a hilarious comic.) Is there some public list we could all sign stating our refusal to buy these games?
Too bad I lost my mod points yesterday. This is the kind of thing people actually come to Slashdot for. I'll just have to try to contribute instead.
:)
Here's some further detail for those interested: the |1> and |0> qubits are actually vectors of probabilities. (Well, probability "amplitudes". More on that later.) The |0> bit means [1 0] and the |1> bit means [0 1]. The "|.>" notation is a bit of convenient shorthand.
If you have two qubits, you'd represent them as |00>, meaning [1 0 0 0]. (That's four possibilities for the qubits, and all the probability mass on the first: both off.) |01> means [0 1 0 0], |000> means [1 0 0 0 0 0 0 0], and so on. Note the exponential growth.
A quantum gate is nothing more than an operator of the same type that governs all discrete quantum system evolution: a unitary matrix. Think of a rotation matrix of rank 2**(number-of-bits), but in complex space. It's got to be some kind of rotation - it must preserve length - to preserve the property that the qubit states and combined qubit states are probability (amplitude) distributions.
A "square root of NOT", IIRC, is an operator (rotation) that turns [1 0] (or |0>) into [sqrt(1/2) -sqrt(1/2)]. Do it again, and you get [0 1]. Again, and you get [-sqrt(1/2) sqrt(1/2)], and again yields the original [1 0]. (I may have some signs wrong.)
The reason this cycle works at all is that the states aren't probabilities per se, but sort of square roots of probabilities, which allows them to keep extra information. This is called "phase". Much of the exciting weirdness of computing with quantum gates is that phase isn't strictly real, but in general has imaginary components.
The other exciting weirdness is of the massively parallel sort. If I do a computation on [sqrt(1/2) -sqrt(1/2)], it's sort of like doing the same computation on [1 0] and [0 1] in parallel. The tricky part is that measuring the outcome restricts me to just one of the results! One way to express the dilemma is that I can compute an answer for every possible input simultaneously (which would be great for solving NP problems), but that I can't easily select the right answer.
Another way to express it is to say that the cat is in a superposition of dead/alive, which will localize when I observe the poor beast.
Funny, we just talked about that a few days ago. The crux is that you're not necessarily right: it depends on whether you're approaching the problem with a frequentist ("frequency of occurrence") or Bayesian ("degree of belief") interpretation of probability. The Bayesian view is more widely applicable (for instance, this study) and easier to model in (because you can easily compose distributions that don't integrate nicely).
Stodgily advocating a single definition and purpose of "probability" is cute and all, in a "get off my lawn" sort of way. But it's a little irksome to those of us who use Bayesian models to solve AI / machine learning / computer vision problems - among other folks.
Still, feel free to tell everyone you know that a coin flip that has already happened but that you haven't observed the outcome of has heads probabilities of 0.0 and 1.0, period. That won't help anybody who has to compute an expected value over outcomes in similar situations, though. To do that, you need a framework in which "How likely was that?" is a valid question.
Also, the assumption that the universe is deterministic is fairly iffy, especially given our current understanding of quantum mechanics, wherein the only true state of a small-scale object is actually a probability distribution over states.
I probably sound a lot pissier than I am right now.
Exactly. Make it an uber-hobby! The best employers know that their best developers do their work as a hobby. Who would you rather hire? Somebody who knows his stuff because it lets him pay the bills, or somebody who knows his stuff because he loves doing it?
I've got an "Interesting Side Projects" section on my resume that everyone I've interviewed with has been impressed with. (I even get compliments on its existence, not just on its contents. Prospective employers appreciate that kind of thing.) At one place I interviewed at, each interviewer (six in all) picked a different side project to ask me about!
There are few professions for which you can demonstrate willingness and love of learning in such a concrete way. Software development is one of them. Take advantage of it!
My mod points expired recently, so could someone mod this up? I do machine learning and computer vision with Bayesian statistics, and the above poster is spot-on. The GP sounds like a frequentist trying to regain control over statistical vocabulary.
:p But "probability" is something we'll lay uncompromising claim to, however much detractors say that subjective probabilities don't count. If they don't count, how else would anyone model something like "belief" in a well-grounded way?
FWIW, the frequentists can keep "confidence interval". We don't want to sully our theoretically sound vocabulary with its filthy connotations.