As I said in a previous post - I use mobile broadband, and have no land line, and it is acceptable. The 15Gb a month I get is more than many landline "fair use" policies. I pay total £30 a month for my phone and broadband.
This is still only a "works for me" solution. The problem behind TFA is getting internet to everybody that wants it (for various reasons - such as encouraging business in rural areas and being able to rely on internet delivery of government services).
Unfortunately, the sort of rural area that can't get ADSL is often the sort of area that has lousy mobile (particularly 3G service). Sounds like you are lucky.
Missing from TFA is any information on what the schools were doing to encourage self-directed learning - by (e.g.) setting interesting homework that could be done on computer, making good use of computers in lessons and possibly (gasp) shifting the curriculum towards understanding subjects rather than memorising bite-size factoids for multiple-choice tests.
When the math curriculum is dominated by learning by rote to perform routine, tedious bits of math gruntwork that anybody not stranded on a desert island would leave to the computer, all a home computer is good for in math is cheating. Throw in a bit of (e.g.) spreadsheet modelling (you'll have to teach them to use formulae - I've even seen *adults* with IT proficiency certificates using a pocket calculator to work out values to type into Excel) and more activities about formulating expressions and equations which can be solved or plotted on computer (rather than laboriously drawing graphs by hand or only ever meeting the tiny subset of equations that can be solved analytically) and maybe things would start to change.
If you just dole out free computers without pro-actively ensuring that they're used for education (just blocking pr0n and Tw@tter doesn't cut it) then the result reported in TFA really is one for the department of Urso-sylvanian scatology.
...cue class-action lawsuit from people who thought they were buying heckabyte drives but found a measly 2^89.692 bytes instead of the 2^90 they were expecting.
If you manage to break any of the Laws of Thermodynamics, you can expect to be lauded, copied, co-author a stream of high-impact papers, get offered some cushy sinecures, and eventually receive a Nobel prize.
...which may be some consolation as your body tempreature spontaneously drops to 100 degrees below absolute zero...
I am not an expert, but the back of the cereal packet not-even-wrong explanation is that neural networks are trained, not programmed: You feed in an input, look at the output, strengthen the links that lead to correct bits of the answer, weaken the ones that don't. The "programming" is the weighting of the links between the nodes in a trained net.
There is no guarantee that you will get the right output for the right input - just a probability, but the upside is that there is no "program" as such so it can be used for producing best-guess solutions to problems for which you dont have an algorithm. Good for image recognition and the like.
Neural net "software" for digital computers is used to simulate and/or train simulated neural nets, its not software "for" neural nets.
One of these days IBM will die and then reincarnate as the new Google.
Please keep up - that happened a while ago when IBM saw the light and came over all open source friendly and heroic slayers of the SCO dragon.
Its like Lord of the Rings 6: The Revenge where the hobbits team up with the Nazgul to defeat a common foe. We're just waiting for the sudden but inevitable betrayal...:-)
this is because we do not understand the algorithms/software running inside the brain.
I thought that enough was known to suggest that the brain works more like a neural net than a digital computer.
i also think that the brain is exactly analogous to a digital computer.
Or is it digitous to an analogue computer?
Seriously, though, Neural Net != Digital (Turing/Von Neumann) computer. No real concept of an algorithm, no real concept of "computability" (only ever delevers a good guess, not an analytic solution, doesn't even require the problem to be formulated).
...and while a computer can simulate a neural net, vice-versa is a bit more tricky...
Even the simple English statement "two plus two is four" relies upon unstated assumptions that one is not working in a Modulo-3 (2+2=1) or Modulo-4 (2+2=0) math system.
I think that neatly illustrates the problem: mathematicians don't understand how language works! English is not an alternative formulation for mathematics, it draws on a whole culture's worth of context and unstated assumptions. The vast majority of the population - mathematicans or not - would automatically assume that your example referred to base 10 arithematic - correctly, unless the author was trying for his "annoying smartarse" badge.
With something like the two boys problem, however, the mathematicians' interpretation of the plain English contradicts the interpretation of the rest of the population.
Your complaints about the "ambiguity in the statement of the puzzle" is reviewed, and rejected.
RTFA - especially the bit where Gardener himself decided the "Two Boys" problem was ambiguous. Also note that TFA justifies the "naive" answer of 1/2 with a minor elaboration of the original question, whereas the defence of the "mathematician's" answer of 1/3 completely re-writes the question. With puppies. So "ambiguous" is a rather charitable way of putting it.
It also makes them no fun. Who would care to wonder about how two men died in a locked cabin in the mountains, when they're told immediately upfront that it is an airplane cabin?
The issue is distinguishing between problems, riddles and just plain obfuscation. Problems should be soluble using the information provided in good faith, plus reasonable assumptions that you can reasonably expected to understand. A riddle is about deliberate obfuscation, wordplay and disinginuity, and involves playing "guess what is in my head" with the questioner.
If consenting mathematicians want to play with riddles in the comfort of their own homes then, as long as they don't frighten the horses, that's fine by me. But Think Of The Children - the danger is that smart-arse math riddles will get passed of as "real world problem solving" and inflicted on kids. Riddles like "Two Boys" could get handed out to schoolkids, who would then fail when they wrote 1/2 (despite the evidence in the TFA that a roomfull of grown mathematicans could argue for days about it). OK, in mitigation, most of them would probably write 1/4...
Well, yes, but the cognitive dissonance in this case is not in understanding the problem, but understanding that all of the conflicting answers (1/2, 1/3 and 13/27) are "right" for a particular interpretation of the question. So the real issue is understanding why different people interpret the question the way they do, and whether these interpretations are defensible.
The fact that TFA had to re-write the original question using puppies to defend the 1/3 answer suggests to me that is the "least right" answer...
The fact that the one child is born on a Tuesday is *only* relevant if you start assuming unmentioned constraints
I think you've almost hit the nail on the head - its the question that's wrong, not the answers - but Tuesday isn't the problem.
The real problem is that the original question - "Suppose that Mr. Smith has two children, at least one of whom is a son. What is the probability both children are boys?" (Ans: 1/2 - you've isolated a single case and identified one male child, so the only relevant thing is the probability of the second child being male) was (mis)interpreted by some as "what proportion of families with two children including at least one boy will have two boys" which is a completely different question (Ans 1/3 - this is the same as the "puppies" example in TFA). This is almost (but not quite) acknowledged in TFA (they had to re-write the question with puppies to defend 1/3).
Throwing in "Tuesday" changes the question to either:
"Suppose that Mr. Smith has two children, at least one of whom is a son. What is the probability both children are boys?" (Ans: you've still isolated a single case and identified one of the male children so its still 1/2. D'oh!)
Or, seen through mathgoggles:
"What proportion of families with two children including at least one boy born on a Tuesday will have two boys?" (Ans: 13/27 - you've constrained the sample and its changed the answer, big surprise)
To summarize the summary: the mathematicians are answering the question they thought they'd asked, not the one that was actually asked. The "Tuesday" problem upsets them because the "right" answer to their question is very close the the "wrong" answer that Stupid People get to the actual question.
Well spotted... strange how so many people seem to think that was just goofy British humour and miss the important point about making sure you know what the question means before trying to calculate an answer.
these sorts of problems are really useful for helping to understand how distributions change shape as fragments of information are supplied.
Fine, but if you are looking for mathematical insight either (a) state the problem in pure mathematical terms or (b) find some authentic context in which an actual human being might want to solve that actual problem - don't dress it up as a nonsensical story with a load of hidden assumptions.
The problem as stated isn't a valid or useful application of mathematics - that doesn't mean the mathematics is wrong or worthless, just that the context is bogus.
Actually, the basic two children problem isn't so bad - or wouldn't be if it stated the assumption of independence and 50% boys) - its the stuff about the day of the week that gets silly.
Because us mathematicians come up with these really simple problems, and when we tell you that your naive answer is wrong...
The naivity is the mathematcian thinking that their re-formulation of the problem in plain English is somehow equivalent to the abstract mathematical formulation in their head, and the inability to conceive that another person might interpret the language differently. There's a huge difference between this sort of pseudo-realistic problem and a bone-fide real-world application of math.
Go re-state that "puzzle" in a way that clearly states what the assumptions are (chance of having a boy = 50% and all births are independent - huge assumptions!) and exactly what probability you want to calculate (the probability of having two boys? the probablility of having two boys, one born on a tuesday? the probability of having two boys, one born on a tuesday, one not?) and then you'll have a simple question.
Oh, and don't further confuse people by "personalising" probability problems: probabilities of magintude 1 don't mean much for an individual event: plan an experiment which will give a measurable result and watch all the unstated assumptions crawl out of the woodwork. If the "correct" solution to a problem has no real-world significance how do you expect people to come up with an intuitive answer or a "correct" interpretation of the question?
That's how the equally famous Monty Hall problem works: it bamboozles people into imagining themselves playing the game and thinking that their personal decision is somehow going to have fractions of cars teleporting between doors. Plan how to test it with 100 contestants and you'll see its a stupid question with no physical significance (you either have two games with different rules and. d'oh, different outcomes or need to factor in the probability of Monty persuading a contestant not to switch). If you want a "realistic" problem, advise Monty on how many cars and goats he should budget for this season.
Take an abstract mathematical problem, invent a pseudo-real-world context, rephrase the problem very sloppily and ambiguously in plain English then laugh smugly when people get the wrong answer.
The correct answer to the question, by the way, is "I don't know - you have not given me enough information, and I'd have to go check that the gender of successive offspring from the same couple is actually independent, but its probably gonna be somewhere between 1/3 and 2/3 - and since you'd have to somehow re-formulate it as a viable experiment and run it 100 times to confirm that result, only the Bayesians give a flying fuck what the precise value is".
In other news: you can't actually build a hotel with an infinite number of rooms - you'd run out of bricks - so don't try and engage my interest in all the weird thing that would happen if you did something impossible. And stop hiding goats behind my door!
Do they test skills like being able to effectively research topics on the internet, write extended reports that make good use of various media, use spreadsheets or algorithms to model and investigate math problems?
...because if they don't, and instead focus on rote learning of little atoms of technical information (like being able to solve a quadratic equation that just happens to factor nicely) then what possible combination of misconceptions could lead to the idea that using a computer would improve performance on these tests?
That's not to say that handing out computers will automatically help kids with the first set of skills, either: it just means that if schools changed their curriculum to reflect the late 20th century, teachers could hand out computer-based assignments without agonizing about equity issues (unless Mum's new boyfriend is hogging the computer 24/7 to look at pr0n - but then you can't fix everything).
Shush! Dont you know that Apple are solely responsible for rogue Australian electronics stores, massive Chinese factories with below-average suicide rates, the security of AT&T's website, the state of the 3G network, global warming, exploding lithium batteries, the BP oil spill and the Kennedy Assasination?
In fact, Apple are responsible for original sin: the company name is a dead giveaway.
Shock! horror! Apple are using their own website to push Safari and claim that their own browsers are ahead of the game on standards support? The bastards!!!
In large friendly letters on the page in question (my emphasis):
The demos below show how the latest version of Apple’s Safari web browser, new Macs, and new Apple mobile devices all support the capabilities of HTML5, CSS3, and JavaScript. Not all browsers offer this support. But soon other modern browsers will take advantage of these same web standards — and the amazing things they enable web designers to do.
Note how that doesn't say "Here's a handy resource to allow you to objectively compare different browsers' HTML 5 implementations"? That is because you are looking at an advert for Safari! As is traditional in these "adverts" it is trying to get you to download and try Safari, not find out how close the competition comes. In other news, if you go to a Mercedes dealership they're not going to offer you test drives in a BMW...
Wake me up if anybody smart enough to spoof their browser ID finds out whether Apple's demos use undocumented or non-standard features (rather than ones which don't work in Firefox, yet).
Econophysics is an interdisciplinary research field, applying theories and methods originally developed by physicists in order to solve problems in economics,
Could be interesting - no physicist would believe that something could grow exponentially for ever without hitting some limit or exhausting some resource, yet that seems to be an article of faith for the economy.
OTOH, if they start applying quantum mechanics to the economy, run for the hills... Structured Investments and CDOs were nuffin'!
I'm still confused though, wouldn't Android + Chrome browser work just as well, if not better?
I think the idea is that ChromeOS can be stripped down to the bare minimum of features needed to support the browser, for fast boot, low power, low memory. Not letting apps run locally at all must also make security easier.
You have a point though: the power/hardware requirements for chrome systems had better be vastly lower than for full featured systems or people won't see the point. I think this is what has kept thin client systems in a niche - they're no cheaper than a proper PC.
Slashdot Mirror
As I said in a previous post - I use mobile broadband, and have no land line, and it is acceptable. The 15Gb a month I get is more than many landline "fair use" policies. I pay total £30 a month for my phone and broadband.
This is still only a "works for me" solution. The problem behind TFA is getting internet to everybody that wants it (for various reasons - such as encouraging business in rural areas and being able to rely on internet delivery of government services).
Unfortunately, the sort of rural area that can't get ADSL is often the sort of area that has lousy mobile (particularly 3G service). Sounds like you are lucky.
No you don't, i have a connection through LLU and i'm certainly not paying BT a damn thing.
So your ISP came and laid a new cable to your house? Or are the little fairies carrying the data to and from the exchange?
A monopoly? Interesting.
Virgin only cable where it is profitable - out of those areas its BT or the highway.
Even with "Local Loop Unbundling" you still have to have a BT line.
...a momentary lapse of unreason.
Missing from TFA is any information on what the schools were doing to encourage self-directed learning - by (e.g.) setting interesting homework that could be done on computer, making good use of computers in lessons and possibly (gasp) shifting the curriculum towards understanding subjects rather than memorising bite-size factoids for multiple-choice tests.
When the math curriculum is dominated by learning by rote to perform routine, tedious bits of math gruntwork that anybody not stranded on a desert island would leave to the computer, all a home computer is good for in math is cheating. Throw in a bit of (e.g.) spreadsheet modelling (you'll have to teach them to use formulae - I've even seen *adults* with IT proficiency certificates using a pocket calculator to work out values to type into Excel) and more activities about formulating expressions and equations which can be solved or plotted on computer (rather than laboriously drawing graphs by hand or only ever meeting the tiny subset of equations that can be solved analytically) and maybe things would start to change.
If you just dole out free computers without pro-actively ensuring that they're used for education (just blocking pr0n and Tw@tter doesn't cut it) then the result reported in TFA really is one for the department of Urso-sylvanian scatology.
we may see hellabyte drives in around 70 years!
...cue class-action lawsuit from people who thought they were buying heckabyte drives but found a measly 2^89.692 bytes instead of the 2^90 they were expecting.
If you manage to break any of the Laws of Thermodynamics, you can expect to be lauded, copied, co-author a stream of high-impact papers, get offered some cushy sinecures, and eventually receive a Nobel prize.
...which may be some consolation as your body tempreature spontaneously drops to 100 degrees below absolute zero...
I am not an expert, but the back of the cereal packet not-even-wrong explanation is that neural networks are trained, not programmed: You feed in an input, look at the output, strengthen the links that lead to correct bits of the answer, weaken the ones that don't. The "programming" is the weighting of the links between the nodes in a trained net. There is no guarantee that you will get the right output for the right input - just a probability, but the upside is that there is no "program" as such so it can be used for producing best-guess solutions to problems for which you dont have an algorithm. Good for image recognition and the like. Neural net "software" for digital computers is used to simulate and/or train simulated neural nets, its not software "for" neural nets.
What do you expect? Apple took the wind out of Slashdotters' fantasy of Linux on the desktop supplanting Windows, so there's some bitterness there.
On the other hand, they have rather successfully put Unix on the desktop. That should count for something.
One of these days IBM will die and then reincarnate as the new Google.
Please keep up - that happened a while ago when IBM saw the light and came over all open source friendly and heroic slayers of the SCO dragon.
Its like Lord of the Rings 6: The Revenge where the hobbits team up with the Nazgul to defeat a common foe. We're just waiting for the sudden but inevitable betrayal... :-)
this is because we do not understand the algorithms/software running inside the brain.
I thought that enough was known to suggest that the brain works more like a neural net than a digital computer.
i also think that the brain is exactly analogous to a digital computer.
Or is it digitous to an analogue computer?
Seriously, though, Neural Net != Digital (Turing/Von Neumann) computer. No real concept of an algorithm, no real concept of "computability" (only ever delevers a good guess, not an analytic solution, doesn't even require the problem to be formulated).
...and while a computer can simulate a neural net, vice-versa is a bit more tricky...
Even the simple English statement "two plus two is four" relies upon unstated assumptions that one is not working in a Modulo-3 (2+2=1) or Modulo-4 (2+2=0) math system.
I think that neatly illustrates the problem: mathematicians don't understand how language works! English is not an alternative formulation for mathematics, it draws on a whole culture's worth of context and unstated assumptions. The vast majority of the population - mathematicans or not - would automatically assume that your example referred to base 10 arithematic - correctly, unless the author was trying for his "annoying smartarse" badge.
With something like the two boys problem, however, the mathematicians' interpretation of the plain English contradicts the interpretation of the rest of the population.
Your complaints about the "ambiguity in the statement of the puzzle" is reviewed, and rejected.
RTFA - especially the bit where Gardener himself decided the "Two Boys" problem was ambiguous. Also note that TFA justifies the "naive" answer of 1/2 with a minor elaboration of the original question, whereas the defence of the "mathematician's" answer of 1/3 completely re-writes the question. With puppies. So "ambiguous" is a rather charitable way of putting it.
It also makes them no fun. Who would care to wonder about how two men died in a locked cabin in the mountains, when they're told immediately upfront that it is an airplane cabin?
The issue is distinguishing between problems, riddles and just plain obfuscation. Problems should be soluble using the information provided in good faith, plus reasonable assumptions that you can reasonably expected to understand. A riddle is about deliberate obfuscation, wordplay and disinginuity, and involves playing "guess what is in my head" with the questioner.
If consenting mathematicians want to play with riddles in the comfort of their own homes then, as long as they don't frighten the horses, that's fine by me. But Think Of The Children - the danger is that smart-arse math riddles will get passed of as "real world problem solving" and inflicted on kids. Riddles like "Two Boys" could get handed out to schoolkids, who would then fail when they wrote 1/2 (despite the evidence in the TFA that a roomfull of grown mathematicans could argue for days about it). OK, in mitigation, most of them would probably write 1/4...
It's cognitive dissonance.
Well, yes, but the cognitive dissonance in this case is not in understanding the problem, but understanding that all of the conflicting answers (1/2, 1/3 and 13/27) are "right" for a particular interpretation of the question. So the real issue is understanding why different people interpret the question the way they do, and whether these interpretations are defensible.
The fact that TFA had to re-write the original question using puppies to defend the 1/3 answer suggests to me that is the "least right" answer...
The fact that the one child is born on a Tuesday is *only* relevant if you start assuming unmentioned constraints
I think you've almost hit the nail on the head - its the question that's wrong, not the answers - but Tuesday isn't the problem.
The real problem is that the original question - "Suppose that Mr. Smith has two children, at least one of whom is a son. What is the probability both children are boys?" (Ans: 1/2 - you've isolated a single case and identified one male child, so the only relevant thing is the probability of the second child being male) was (mis)interpreted by some as "what proportion of families with two children including at least one boy will have two boys" which is a completely different question (Ans 1/3 - this is the same as the "puppies" example in TFA). This is almost (but not quite) acknowledged in TFA (they had to re-write the question with puppies to defend 1/3).
Throwing in "Tuesday" changes the question to either:
"Suppose that Mr. Smith has two children, at least one of whom is a son. What is the probability both children are boys?" (Ans: you've still isolated a single case and identified one of the male children so its still 1/2. D'oh!)
Or, seen through mathgoggles:
"What proportion of families with two children including at least one boy born on a Tuesday will have two boys?" (Ans: 13/27 - you've constrained the sample and its changed the answer, big surprise)
To summarize the summary: the mathematicians are answering the question they thought they'd asked, not the one that was actually asked. The "Tuesday" problem upsets them because the "right" answer to their question is very close the the "wrong" answer that Stupid People get to the actual question.
The answer of course, is 42...
Well spotted... strange how so many people seem to think that was just goofy British humour and miss the important point about making sure you know what the question means before trying to calculate an answer.
Actually, the basic two children problem isn't so bad
Sorry, retract that, just read TFA again... That's the one that's ambiguous... My head hurts.
these sorts of problems are really useful for helping to understand how distributions change shape as fragments of information are supplied.
Fine, but if you are looking for mathematical insight either (a) state the problem in pure mathematical terms or (b) find some authentic context in which an actual human being might want to solve that actual problem - don't dress it up as a nonsensical story with a load of hidden assumptions.
The problem as stated isn't a valid or useful application of mathematics - that doesn't mean the mathematics is wrong or worthless, just that the context is bogus.
Actually, the basic two children problem isn't so bad - or wouldn't be if it stated the assumption of independence and 50% boys) - its the stuff about the day of the week that gets silly.
Because us mathematicians come up with these really simple problems, and when we tell you that your naive answer is wrong...
The naivity is the mathematcian thinking that their re-formulation of the problem in plain English is somehow equivalent to the abstract mathematical formulation in their head, and the inability to conceive that another person might interpret the language differently. There's a huge difference between this sort of pseudo-realistic problem and a bone-fide real-world application of math.
Go re-state that "puzzle" in a way that clearly states what the assumptions are (chance of having a boy = 50% and all births are independent - huge assumptions!) and exactly what probability you want to calculate (the probability of having two boys? the probablility of having two boys, one born on a tuesday? the probability of having two boys, one born on a tuesday, one not?) and then you'll have a simple question.
Oh, and don't further confuse people by "personalising" probability problems: probabilities of magintude 1 don't mean much for an individual event: plan an experiment which will give a measurable result and watch all the unstated assumptions crawl out of the woodwork. If the "correct" solution to a problem has no real-world significance how do you expect people to come up with an intuitive answer or a "correct" interpretation of the question?
That's how the equally famous Monty Hall problem works: it bamboozles people into imagining themselves playing the game and thinking that their personal decision is somehow going to have fractions of cars teleporting between doors. Plan how to test it with 100 contestants and you'll see its a stupid question with no physical significance (you either have two games with different rules and. d'oh, different outcomes or need to factor in the probability of Monty persuading a contestant not to switch). If you want a "realistic" problem, advise Monty on how many cars and goats he should budget for this season.
Take an abstract mathematical problem, invent a pseudo-real-world context, rephrase the problem very sloppily and ambiguously in plain English then laugh smugly when people get the wrong answer.
The correct answer to the question, by the way, is "I don't know - you have not given me enough information, and I'd have to go check that the gender of successive offspring from the same couple is actually independent, but its probably gonna be somewhere between 1/3 and 2/3 - and since you'd have to somehow re-formulate it as a viable experiment and run it 100 times to confirm that result, only the Bayesians give a flying fuck what the precise value is".
In other news: you can't actually build a hotel with an infinite number of rooms - you'd run out of bricks - so don't try and engage my interest in all the weird thing that would happen if you did something impossible. And stop hiding goats behind my door!
Do they test skills like being able to effectively research topics on the internet, write extended reports that make good use of various media, use spreadsheets or algorithms to model and investigate math problems?
...because if they don't, and instead focus on rote learning of little atoms of technical information (like being able to solve a quadratic equation that just happens to factor nicely) then what possible combination of misconceptions could lead to the idea that using a computer would improve performance on these tests?
That's not to say that handing out computers will automatically help kids with the first set of skills, either: it just means that if schools changed their curriculum to reflect the late 20th century, teachers could hand out computer-based assignments without agonizing about equity issues (unless Mum's new boyfriend is hogging the computer 24/7 to look at pr0n - but then you can't fix everything).
You forgot to mention the men in black and UFOs
I did - but I was typing on my iPad and so it probably got filtered out.
However, the UFO guys stopped using Apple kit when their "iAnalProbe" App was rejected for using a private API.
Shush! Dont you know that Apple are solely responsible for rogue Australian electronics stores, massive Chinese factories with below-average suicide rates, the security of AT&T's website, the state of the 3G network, global warming, exploding lithium batteries, the BP oil spill and the Kennedy Assasination?
In fact, Apple are responsible for original sin: the company name is a dead giveaway.
Shock! horror! Apple are using their own website to push Safari and claim that their own browsers are ahead of the game on standards support? The bastards!!!
In large friendly letters on the page in question (my emphasis):
The demos below show how the latest version of Apple’s Safari web browser, new Macs, and new Apple mobile devices all support the capabilities of HTML5, CSS3, and JavaScript. Not all browsers offer this support. But soon other modern browsers will take advantage of these same web standards — and the amazing things they enable web designers to do.
Note how that doesn't say "Here's a handy resource to allow you to objectively compare different browsers' HTML 5 implementations"? That is because you are looking at an advert for Safari! As is traditional in these "adverts" it is trying to get you to download and try Safari, not find out how close the competition comes. In other news, if you go to a Mercedes dealership they're not going to offer you test drives in a BMW...
Wake me up if anybody smart enough to spoof their browser ID finds out whether Apple's demos use undocumented or non-standard features (rather than ones which don't work in Firefox, yet).
Econophysics is an interdisciplinary research field, applying theories and methods originally developed by physicists in order to solve problems in economics,
Could be interesting - no physicist would believe that something could grow exponentially for ever without hitting some limit or exhausting some resource, yet that seems to be an article of faith for the economy.
OTOH, if they start applying quantum mechanics to the economy, run for the hills... Structured Investments and CDOs were nuffin'!
I'm still confused though, wouldn't Android + Chrome browser work just as well, if not better?
I think the idea is that ChromeOS can be stripped down to the bare minimum of features needed to support the browser, for fast boot, low power, low memory. Not letting apps run locally at all must also make security easier.
You have a point though: the power/hardware requirements for chrome systems had better be vastly lower than for full featured systems or people won't see the point. I think this is what has kept thin client systems in a niche - they're no cheaper than a proper PC.