That's a different world from India. According to the article, the posh private school is $1,500 per year, and the government only pays $300 for the 25% poor students it has forced the school to take. $1,500 per year is less than 1/10th what some US states pay per pupil.
Doing some poking around, it appears the average primary school teacher salary is about $3,000 per year in India, with a student:teacher ratio of 40:1, meaning the government expenditure on teaching is only $75/year per student. Even so, that is over 10% of the total government budget, and the expenditure on education as a percentage of GDP is about the same as South Korea. (4.1% vs. 4.2%) Other poor countries such as Yemen, Cuba and Morocco spend much more in GDP terms than India, though, and India could afford to reduce the student-teacher ratio to 30:1 while at the same time increasing the average teacher salary to $4,500. (Thus doubling teacher salary expenditures to $150 per student per year.) The effective cost per teacher hour would rise less than 20% if absenteeism were reduced from 25% to 5%. A less ridiculously low salary would attract more and better teachers and a smaller class size would reduce burnout and absenteeism.
In the US, on the other hand, most states are well into the region of diminishing returns from paying higher salaries, although reducing class size could still have benefits. The main opportunities in the US for improving education are in eliminating the expensive and counterproductive bureaucracy and the ridiculously over-specified curricula and methods that it mandates.
I agree with your first paragraph and the overall point of your whole post, but you say:
"Secondly it is in your interest to subsidise poorer children. You want the best and brightest children to power the economy or become your doctor. "
Yes, but ability is highly correlated with parental wealth. This is because intelligence and other success related qualities of the parents are correlated with earning power, and these abilities are highly heritable, both genetically and indirectly through parent-child interaction and home environment. Statistically, students from poor households have lower mean abilities than those from wealthier families. To get the gifted from poor households into better schools, one must test them, not do what India is doing, forcing those who pay for private schools to compete for entry while the poor are waved in without any selection for ability. That is a recipe for failure for the poor who do get in, and for making sure that most of the most able poor students are left behind in the lousy public schools.
The incarcerated population in the US is about 2.3 million - the other 4.7 million are on probation or parole. The school-age population does not count preschool, early kindergarten or any students over age 17, so it should be ~12% higher. The ratio of populations is about 24.
Samsung has had 32GB/40nm sticks out for over a year, 16GB out for 2 years, now they are about to ship the 30nm 16/32GB modules with lower power consumption. Price per GB is bad. Memory4less has: Samsung 16GB PC3-8500 DDR3-1066MHz ECC Registered CL7 240-Pin DIMM ~$950, other speeds for more $ 32GB sticks run ~$2,150 and up = about 6 to 7 times the cost/GB of vanilla 4GB sticks on newegg, or about 3 to 4 times the cost/GB of 8GB ECC sticks.
"...the falling tree would have to avoid hitting anything at all (even a single photon hitting it would end it's superposition)" No, , (assuming the timing of the fall of the tree has some uncertainty, which it must) the tree would just entangle whatever it interacted with into that superposition between fallen and not-fallen and as those things in turn interacted with other things there would be an outward spreading wave of entanglement that would quickly become effectively irreversible (decoherent). In one set of paths the environment (including humans) would be indirectly entangled with a standing tree, in the other with a fallen tree. Decoherence is a matter of degree - for slight degrees it can be reversed, the "observation" undone - see the "quantum eraser" experiments.
I think you have that backwards. QM places strict limits on the information obtainable from individual measurements, but much less strict limits on measurements of ensembles. Any individual interaction can only yield so much information, many interactions can yield more information - but each interaction is separate and doesn't technically say anything about any of the other individual interactions, but rather about the process producing the interactions. The Heisenberg uncertainty principle (specifically the time-energy form) applies even to classical waves - I've seen it myself in testing tunable-Q filters - no need for particles at all. Also, ensembles are not needed for a rigorous statistical theory - the Bayesians deride those who think that ensembles are the basis of statistics as "frequentists", and instead say that probability is really a measure of observers' information, and has different values depending on the degree of information each observer has. The "statistical distribution" of the frequentists is not an objective thing, but depends on the degree of information the observer has, and changes as more data comes in. The Bayesian approach meshes much better with information theory and QM and easily explains apparent paradoxes such as the Monty Hall problem.
That's much better than the original explanation. To boil it down even further, quanta are waves when they are going somewhere (propagating) and particles when they get there (interacting). Each photon does actually go through both slits, which isn't a problem because it's a wave. When it hits the screen, it interacts in an all-or nothing, localized fashion, which gives the appearance of a particle.
The interesting thing about this experiment is that it further demonstrates that there is a continuum between particle and wave, interaction and propagation, but that this can only be shown as a statistical effect using many observations.
No, GPU rendering should achieve at least 10x and perhaps much more speedup. The Fusion architecture will allow using main RAM for the GPU with little penalty. Allowing a major project like Blender to get locked into your arch-nemesis' language standard is also a huge opportunity cost for ATI.
According to the Martin Jetpack website, 5 gal gives 30 minutes, which is 10 gal./hr. I'm not sure where the 45 minutes came from - a larger tank on the test craft or just an optimistic early estimate would be my guess, but it's possible that the efficiency was better than they hoped and they just haven't updated their site.
Yeah, the methodology on this sucks. They're counting everybody up to 65 years old, and those over 45-50 went to school when it was cheap, easily available side jobs would pay your tuition, and you could have your student loans (if any) discharged in bankruptcy. The relevant measure today is "will I be able to make my loan payments throughout my first decade out of school?", "how fucked will I be if it turns out I can't?", "how long will it take me to pay off those loans while still eating everyday and sleeping indoors?", "what is the net present value of all the payments I will make over that period?" and "does the data show that this particular course of study at this particular school is a better investment of time and money than other opportunities?".
I think looking at those questions, for many of those currently contemplating college it really isn't a good deal on the terms offered today. The education bubble is going to burst someday - it isn't affordable, the schools and student-loan pushers are bilking the students as hard as they can, and one can get a better education by reading and doing, and more prestige by teaming up to start a company. Why take a mortgage out on your brain so you can beg employers for the opportunity to be treated like a Dilbert?
And we're nowhere near that limit. Using Frink: 535lbs gravity 5000ft -> gal gasoline = 0.0259... Even assuming 25% efficiency, it's still about 1/10 of a US gallon of gas. (the 200hp Martin engine burns 10gal/hr, so it could be as much as 38% efficient, but the transmission, fans and est.90% throttle pull that figure down.) The actual gas burned, though, to get to 5000 ft was likely about an order of magnitude greater. (~4000ft/800fpm=5min; 10gph*1/12hr=5/6gal)
The energy mostly goes not to lifting the craft from one height to another but rather to accelerating air simply to maintain altitude.
The jetpack produces ~600 lbs. thrust at 10 gph = 61lbs/hr.. Empty weight is 250lbs, max takeoff weight is about 535lbs.. A 180lb pilot could carry over 15gal of fuel, sufficient for 90 minutes of flight. The excess thrust over takeoff weight ensures a rapid climb and a decent forward velocity.
The test did not start at high altitude, but on the Christchurch plain in South Island, NZ which slopes gently from 30ft to 1900ft over about 20 to 35 miles. The test takeoff was likely at less than 1000ft, and the craft clearly achieved a substantial height above the ground since it nearly disappeared from view.
I missed the 800fpm bit, but according to their site, the design is an part103-compliant ultralight which requires no license (though Martin will require a training program when they start selling them), Fuel capacity is 5gal (by FAA requirement) / 30min / ~30mi range.
There seems to be a huge amount of spam in this thread coming from this long-winded industry apologist. The reasoning is always pathetic yet condescending, and while some may be swayed by this sort of hand-waving, I think more perceptive and discerning readers will come to almost automatically conclude the opposite of any position this manipulator is pushing.
Because they require several different kinds of completely alien thinking; knowing imperative programming is almost a disadvantage to learning functional programming. Also, most functional languages were too purist to allow actually getting much real work done - the libraries were usually very weak so every problem involved reinventing dozens of wheels. Clojure may have solved this problem - it isn't too purist and it can use all the Java libraries. It still requires a lot from the programmer, though perhaps less than the gain in power it provides.
They do owe us. Every bit of their success as a monster advertizing company comes from us using their services (a.k.a. ad bait). To the extent they reduce services, to at least the same extent they should lose revenue. In this case I'd argue that they should lose more than a proportional amount since they have harmed people through encouraging reliance on the service and by shading out and stunting competing translation services. Other free translation services lost a huge amount of traffic because of Google translate and thus couldn't afford to improve their service - some likely went out of business, others became drastically less valuable to the world when they became pay services. Google got users hooked on a service, hurt their competitors with unfair trade practices (abuse of effective monopoly in other areas), and now is cutting the users off and not sharing their translation software with former competitors in translation services or users. This indicates a dog-in-the-manger attitude - it's not really that it's too much trouble to provide the service, as Google is claiming, but rather that they don't want the service to be freely available at all.
I prefer kid (young goat - the word "kid" as signifying "a child" is a figure of speech, when you refer to someone as a "kid", you're calling them a young goat. Apologies to those who already knew this. Some people react badly when you tell them you enjoy eating the flesh of kids. )
That's how it should work, but sometimes there is a premature hardening of consensus and the whole scientific community may believing nonsense for decades on end. Commonly believed theories need to be tested and challenged, too - but funding and social/ peer review concerns often make thatt difficult, particularly in the softer sciences.
It's not in redshift space in the sense of a spatial dimension, the dots are just color coded on a 2-D projection. The plot appears to include all positive redshifts. Most of the redshifts are below 0.02. I don't think they are concentrating on the cosmological redshifts, rather all reasonably bright, low redshift galaxies - the point is to get the observational data, others can crunch and interpret it as desired.
Slashdot Mirror
That's a different world from India. According to the article, the posh private school is $1,500 per year, and the government only pays $300 for the 25% poor students it has forced the school to take. $1,500 per year is less than 1/10th what some US states pay per pupil.
Doing some poking around, it appears the average primary school teacher salary is about $3,000 per year in India, with a student:teacher ratio of 40:1, meaning the government expenditure on teaching is only $75/year per student. Even so, that is over 10% of the total government budget, and the expenditure on education as a percentage of GDP is about the same as South Korea. (4.1% vs. 4.2%) Other poor countries such as Yemen, Cuba and Morocco spend much more in GDP terms than India, though, and India could afford to reduce the student-teacher ratio to 30:1 while at the same time increasing the average teacher salary to $4,500. (Thus doubling teacher salary expenditures to $150 per student per year.) The effective cost per teacher hour would rise less than 20% if absenteeism were reduced from 25% to 5%. A less ridiculously low salary would attract more and better teachers and a smaller class size would reduce burnout and absenteeism.
In the US, on the other hand, most states are well into the region of diminishing returns from paying higher salaries, although reducing class size could still have benefits. The main opportunities in the US for improving education are in eliminating the expensive and counterproductive bureaucracy and the ridiculously over-specified curricula and methods that it mandates.
I agree with your first paragraph and the overall point of your whole post, but you say:
"Secondly it is in your interest to subsidise poorer children. You want the best and brightest children to power the economy or become your doctor. "
Yes, but ability is highly correlated with parental wealth. This is because intelligence and other success related qualities of the parents are correlated with earning power, and these abilities are highly heritable, both genetically and indirectly through parent-child interaction and home environment. Statistically, students from poor households have lower mean abilities than those from wealthier families. To get the gifted from poor households into better schools, one must test them, not do what India is doing, forcing those who pay for private schools to compete for entry while the poor are waved in without any selection for ability. That is a recipe for failure for the poor who do get in, and for making sure that most of the most able poor students are left behind in the lousy public schools.
The incarcerated population in the US is about 2.3 million - the other 4.7 million are on probation or parole. The school-age population does not count preschool, early kindergarten or any students over age 17, so it should be ~12% higher. The ratio of populations is about 24.
see: http://www.physorg.com/news/2011-05-samsung-mass-30nm-class-gigabyte-memory.html
Samsung has had 32GB/40nm sticks out for over a year, 16GB out for 2 years, now they are about to ship the 30nm 16/32GB modules with lower power consumption. Price per GB is bad.
Memory4less has:
Samsung 16GB PC3-8500 DDR3-1066MHz ECC Registered CL7 240-Pin DIMM
~$950, other speeds for more $
32GB sticks run ~$2,150 and up
= about 6 to 7 times the cost/GB of vanilla 4GB sticks on newegg, or about 3 to 4 times the cost/GB of 8GB ECC sticks.
"...the falling tree would have to avoid hitting anything at all (even a single photon hitting it would end it's superposition)"
No, , (assuming the timing of the fall of the tree has some uncertainty, which it must) the tree would just entangle whatever it interacted with into that superposition between fallen and not-fallen and as those things in turn interacted with other things there would be an outward spreading wave of entanglement that would quickly become effectively irreversible (decoherent). In one set of paths the environment (including humans) would be indirectly entangled with a standing tree, in the other with a fallen tree. Decoherence is a matter of degree - for slight degrees it can be reversed, the "observation" undone - see the "quantum eraser" experiments.
I think you have that backwards. QM places strict limits on the information obtainable from individual measurements, but much less strict limits on measurements of ensembles. Any individual interaction can only yield so much information, many interactions can yield more information - but each interaction is separate and doesn't technically say anything about any of the other individual interactions, but rather about the process producing the interactions. The Heisenberg uncertainty principle (specifically the time-energy form) applies even to classical waves - I've seen it myself in testing tunable-Q filters - no need for particles at all. Also, ensembles are not needed for a rigorous statistical theory - the Bayesians deride those who think that ensembles are the basis of statistics as "frequentists", and instead say that probability is really a measure of observers' information, and has different values depending on the degree of information each observer has. The "statistical distribution" of the frequentists is not an objective thing, but depends on the degree of information the observer has, and changes as more data comes in. The Bayesian approach meshes much better with information theory and QM and easily explains apparent paradoxes such as the Monty Hall problem.
You are correct, the OP wasn't. The wave goes through both slits, but the measurement of an individual photon must be discrete and thus particle-like.
That's much better than the original explanation. To boil it down even further, quanta are waves when they are going somewhere (propagating) and particles when they get there (interacting). Each photon does actually go through both slits, which isn't a problem because it's a wave. When it hits the screen, it interacts in an all-or nothing, localized fashion, which gives the appearance of a particle.
The interesting thing about this experiment is that it further demonstrates that there is a continuum between particle and wave, interaction and propagation, but that this can only be shown as a statistical effect using many observations.
No, GPU rendering should achieve at least 10x and perhaps much more speedup. The Fusion architecture will allow using main RAM for the GPU with little penalty. Allowing a major project like Blender to get locked into your arch-nemesis' language standard is also a huge opportunity cost for ATI.
Hmm... the 386-40 with coprocessor wasn't much slower (~20%) than the 486SX-33 and it only got about 15-20 FPS.
Whatever they call it, it'll be programmed in Ali-OOP
Only if we submit to a government not of LOLs, but of men.
According to the Martin Jetpack website, 5 gal gives 30 minutes, which is 10 gal./hr. I'm not sure where the 45 minutes came from - a larger tank on the test craft or just an optimistic early estimate would be my guess, but it's possible that the efficiency was better than they hoped and they just haven't updated their site.
Rumors to the contrary aside, Finnish only causes mild brain damage, no worse than APL, aquavit or Apple.
w^2(t) f ?!
Yeah, the methodology on this sucks. They're counting everybody up to 65 years old, and those over 45-50 went to school when it was cheap, easily available side jobs would pay your tuition, and you could have your student loans (if any) discharged in bankruptcy. The relevant measure today is "will I be able to make my loan payments throughout my first decade out of school?", "how fucked will I be if it turns out I can't?", "how long will it take me to pay off those loans while still eating everyday and sleeping indoors?", "what is the net present value of all the payments I will make over that period?" and "does the data show that this particular course of study at this particular school is a better investment of time and money than other opportunities?".
I think looking at those questions, for many of those currently contemplating college it really isn't a good deal on the terms offered today. The education bubble is going to burst someday - it isn't affordable, the schools and student-loan pushers are bilking the students as hard as they can, and one can get a better education by reading and doing, and more prestige by teaming up to start a company. Why take a mortgage out on your brain so you can beg employers for the opportunity to be treated like a Dilbert?
And we're nowhere near that limit. Using Frink:
535lbs gravity 5000ft -> gal gasoline = 0.0259...
Even assuming 25% efficiency, it's still about 1/10 of a US gallon of gas. (the 200hp Martin engine burns 10gal/hr, so it could be as much as 38% efficient, but the transmission, fans and est.90% throttle pull that figure down.) The actual gas burned, though, to get to 5000 ft was likely about an order of magnitude greater. (~4000ft/800fpm=5min; 10gph*1/12hr=5/6gal)
The energy mostly goes not to lifting the craft from one height to another but rather to accelerating air simply to maintain altitude.
The jetpack produces ~600 lbs. thrust at 10 gph = 61lbs/hr.. Empty weight is 250lbs, max takeoff weight is about 535lbs.. A 180lb pilot could carry over 15gal of fuel, sufficient for 90 minutes of flight. The excess thrust over takeoff weight ensures a rapid climb and a decent forward velocity.
The test did not start at high altitude, but on the Christchurch plain in South Island, NZ which slopes gently from 30ft to 1900ft over about 20 to 35 miles. The test takeoff was likely at less than 1000ft, and the craft clearly achieved a substantial height above the ground since it nearly disappeared from view.
I missed the 800fpm bit, but according to their site, the design is an part103-compliant ultralight which requires no license (though Martin will require a training program when they start selling them), Fuel capacity is 5gal (by FAA requirement) / 30min / ~30mi range.
There seems to be a huge amount of spam in this thread coming from this long-winded industry apologist. The reasoning is always pathetic yet condescending, and while some may be swayed by this sort of hand-waving, I think more perceptive and discerning readers will come to almost automatically conclude the opposite of any position this manipulator is pushing.
Because they require several different kinds of completely alien thinking; knowing imperative programming is almost a disadvantage to learning functional programming. Also, most functional languages were too purist to allow actually getting much real work done - the libraries were usually very weak so every problem involved reinventing dozens of wheels. Clojure may have solved this problem - it isn't too purist and it can use all the Java libraries. It still requires a lot from the programmer, though perhaps less than the gain in power it provides.
They do owe us. Every bit of their success as a monster advertizing company comes from us using their services (a.k.a. ad bait). To the extent they reduce services, to at least the same extent they should lose revenue. In this case I'd argue that they should lose more than a proportional amount since they have harmed people through encouraging reliance on the service and by shading out and stunting competing translation services. Other free translation services lost a huge amount of traffic because of Google translate and thus couldn't afford to improve their service - some likely went out of business, others became drastically less valuable to the world when they became pay services. Google got users hooked on a service, hurt their competitors with unfair trade practices (abuse of effective monopoly in other areas), and now is cutting the users off and not sharing their translation software with former competitors in translation services or users. This indicates a dog-in-the-manger attitude - it's not really that it's too much trouble to provide the service, as Google is claiming, but rather that they don't want the service to be freely available at all.
I prefer kid (young goat - the word "kid" as signifying "a child" is a figure of speech, when you refer to someone as a "kid", you're calling them a young goat. Apologies to those who already knew this. Some people react badly when you tell them you enjoy eating the flesh of kids. )
That's how it should work, but sometimes there is a premature hardening of consensus and the whole scientific community may believing nonsense for decades on end. Commonly believed theories need to be tested and challenged, too - but funding and social/ peer review concerns often make thatt difficult, particularly in the softer sciences.
It's not in redshift space in the sense of a spatial dimension, the dots are just color coded on a 2-D projection. The plot appears to include all positive redshifts. Most of the redshifts are below 0.02. I don't think they are concentrating on the cosmological redshifts, rather all reasonably bright, low redshift galaxies - the point is to get the observational data, others can crunch and interpret it as desired.