The fact that it can muck up cheap-n-cheerful 'zillion sites on one host' arrangements unless everybody's TLS ducks are in a row probably counts as a fairly significant cost, as well. Especially on the low end.
It is fairly hard to make a public-interest argument in favor of using scarce(note to spread spectrum and other tech-trick enthusiasts: yes, how limited it is strongly depends on how smart you are about it; but it is finite) spectrum for high-bandwidth broadcasts.
There are the "public information dissemination"/"cultural goods" arguments, which are reasonably strong; but also amply satisfied by even AM quality voice and smeary NTSC quality video. More video bandwidth is certainly better; but that "better" is simply an aesthetic improvement, not a matter of any significant interest(especially in a world where the bandwidth of blu-ray+USPS is so damn high. There is a value in people being able to get current news/political events/hazard warnings in real time; which is a broadcast specialty; but there is no need to allocate enough spectrum so that they can count the pores on senator scumweasel's nose. For ~$12, you can get 50GB video entertainment chunks mailed to your door, not to mention consoles and HTPCs and all the other non-broadcast uses of HDTVs)
That said, I'm really not an enthusiast of the "sell it all to Verizon, Ma Bell knows best." theory of spectrum allocation. I'd prefer to see a much stronger support of un or minimally licenced data-transmission spectrum, along the lines of wifi; but with spectrum that doesn't totally blow. Even laboring under those restrictions, wifi has been an amazing success, and the possibilities of future minimal-licensing wireless are really much more compelling than "another bunch of TV channels" or "200mb/s TO YOUR CELLPHONE* *Capped at 5GB/month, overage $10/GB, you're damn right other terms and restrictions apply, see 2,000 pages of fine print for details.*"
And hand the spectrum over to the next generation of 802.11b/g/n-esque applications.
Even confined to a couple of really sucky blocks of spectrum, the success of no-license-to-deploy, inexpensive wireless data standards has been extraordinary. Why not murder a few bloated, feckless, incumbents and hand over some proper spectrum for this proven and extremely useful application?
It's more or less undeniable that we are going to run out of low-hanging fruit as time goes on(it is, after all, entirely reasonable to take advantage of the most accessible potential improvements to your technology, which necessarily leaves you with improved technology and a harder set of future improvements). Just imagine how disappointed somebody upgrading from a vacuum tube based system to an IC based one would be with the rate of progress at any point from then on...
The part where I think Kaku goes right off the deep end is where he predicts dire catastrophe because of this. The notion that there may be certain hard limits, based on the increasingly unhelpful properties of matter as you start working with less of it, isn't a secret to anybody, even if we don't know precisely where the hard limit is waiting... Even when we hit it, all our existing chips will still work, and transistors will still be crazy cheap. They just won't keep reliably getting cheaper at high speed.
To go with a historical analogy, Kaku makes me imagine somebody during the industrial revolution, standing on a soapbox and shouting about how "Our steam engines cannot keep improving forever, no matter how clever our machinists and engineers, we will hit the Carnot limit for heat engines, and all progress must cease! Disaster! Calamity!" In terms of the physics, that isn't wrong: thermodynamics does place certain hard limits on the efficiency of ideal heat engines. You can tighten your cylinder tolerances and lubricate away friction and insulate all you like; but that's the limit. However, it doesn't follow that no longer being able to expect next year's heat engines to be more efficient than last year's means doom. It just means that you won't keep getting nicer heat engines automatically...
Among other factors, the Concorde was arguably positioned outside the limits of speed / air-resistance economics... If you look at more recent work in aircraft design, the two big trends are large, high-efficiency, slightly subsonic designs that allow airlines to link major hubs at low cost per unit weight and/or floor space(freeing them to trade off between selling large numbers of cattle-class tickets, or smaller numbers of 'you get your own damn living room' class tickets, or a combination, with minor interior remodeling) and relatively cheap smaller models to cover the peripheral airports that bring travellers to the major hubs at fairly low cost.
The Concorde is a bit of a white elephant because, due to the engineering constraints of supersonic flight, it is actually pretty horrid inside(only four seats wide, limited overhead compartments, etc.) but extremely expensive. Particularly with the advances in mobile computing, cellular/satellite data links, and so forth(and, on the downside, advances in being hassled by rentacops at the airport), it is kind of a hard sell even for the money-is-no-object crowd for which it is the best economic fit. The big subsonic airliners can offer you a fairly roomy and comfortable environment, with electricity and bandwidth, and a reclining chair and such, for less than the Concorde can offer a faster; but rather more spartan, trip. Unless we see a fairly fundamental shakeup in what business travelers want, it is difficult to imagine enough of them choosing fast-and-spartan to bring the economies of scale even remotely within reach of the "it's huge, and fuel efficient, and you can swiftly adjust the ratio between luxury class and cattle class without changing airframes" designs that are preferred today.
While the envelope has continued to be pushed in the less cost-constrained military designs, we've actually seen some orphan high-speed exotics there as well. The SR-71, another sweet looking but hugely expensive high speed design, is out of a job now that a mixture of satellites and long-loiter robotic cheapies are doing surveillance jobs. It would appear that very high speed atmospheric operation, while cool and possible, often isn't worth the trouble.
I agree that the doomtastic prophecies seem rather overblown. If nothing else, the fact that it is (comparatively) easy to make predictions about the endpoint of Moore's law, and the approximate performance thereof, makes being taken by surprise harder. And surprise, not limitations in themselves, is where complex systems really tend to bite it. We know how large atoms are, we have predictions of varying pessimism regarding how many you need to make a transistor that will actually work, and work often enough that you can fabricate chips out of them in some economically viable way. From there, you can calculate the approximate transistor density of the various endpoints. If those crazy kids in engineering happen to out-perform you, great. If not, that sucks; but you still know where you are.
Then there is the fact that only certain applications need ever more power. The scientific computing and simulations people are certainly happy to soak up every cycle you can afford to throw at them; but many other applications reach diminishing returns at much lower levels. The end of Moore's law won't suddenly make those stop working, we just won't be able to shrug and say "eh, it'll be twice as fast and half the cost next year." anymore. If our economy is on such a razor's edge that that is going to throw us into global chaos, we have broader issues...
Arguably, the hardocre apocalyptic gun-toters know that they don't really need survival kits: All they need is a list of nearby people who have survival kits, and their existing supplies of guns and ammunition...
The opportunity cost of the required capital is a "cost of production". It would appear that we differ only in wording in that regard. If capital is fleeing a sector, even if the accountants are putting the sector in the black, it isn't breaking even overall.
Why would a credit card issuer feel threatened by somebody who is trying to increase the convenience and reduce the cost of accepting their credit cards? Some of the bloated incumbent processors are evidently shitting bricks; but why would Visa object to a company whose product offering should markedly increase the number of people accepting credit cards, rather than cash/checks only...
In theory, a competitive market will result in the price of a commodity or service equalling its marginal cost of production.
The fact that the amount the CC guys skim off the top of transactions(in addition to their loan-sharking business) has barely budged despite what are almost certainly declining costs leads one to suspect that the market may not be especially competitive.
The phrase "cosy duopoly", in fact, tends to spring to mind...
Their core tech is actually pretty cool(at least to my layman-when-it-comes-to-music-theory-and-FFTs eyes watching it build a deviation/time graph of the recorded student performance vs. the correct performance is a very neat trick); but pretty much all the 'peripheral' stuff(y'know, unimportant things like 'UI' and 'Not crashing' and 'having a search function that works') sucks pretty badly.
If Ubisoft has an adequately competent core, with the spit and polish given to (most) games that aren't headed right for the bargain bin, it isn't going to be pretty...
I can only assume that a nontrivial chunk of the price tag is for the "expertise required to safely approach a moving satellite and introduce additional fuel, without crashing into it, breaking off any important solar panels/antennas/widgety bits, or otherwise mucking it up.
Sort of a very high end version of the classic techie contractor invoice: "'Typing a one line command, $1' 'Knowing which command to type, $400/hr+travel'"
There is a little more nuance than that: Obviously, the onsite engineers are not "saving face"(if anything, they are running the nontrivial risk of having face slough off most unpleasantly) and the Japanese government also seems to have come round to the severity of the situation(if not quite to the degree that experts from other countries with lots of reactors would like). Every statement from plant management, though, seems to be a variation on "Golly shucks, there are definitely some issues; but it is hard to say exactly how severe they are. We are working on it, check back later." It doesn't give you the warm and fuzzies about how forthright they are being...
There is a certain amount of reason to suspect that they were packing the spent-fuel ponds rather tighter than would really be advisable in an area where damage to the powered systems is known to be plausible...
The reactors themselves, the really complex bits, seem to be doing comparatively adequately; but the condition of the fuel ponds seems pretty dodgy(since, after all, simply spreading the spent assemblies out more, or having bigger ponds, while more costly, would have been a trivial way to increase the safety margin with minimal engineering complexity.
Nuclear reactors are hard. Nuclear fuel ponds are safety-critical applications of swimming pool technology...
I'm not quite sure where you are projecting from; but my comment was a quite narrowly focused observation that childhood bullying was of the same 1st amendment salience as late-night disturbance of the peace: that is, extraordinarily minimal and well below the level of concern.
Now, that said, I have approximately zero confidence that school admin will be able to do much of anything based on this new mandate. In my school experience, adult authority figures were, without exception, useless or worse in dealing with bullies. I doubt that they've improved too much, and now their mandate is supposed to extend to the internet? Good luck with that one, guys.
However, I really must take exception to your tedious argument-from-cliche and your extraordinarily optimistic take on the level of persistence shown by bullies. Again, in my(admittedly anecdotal) experience, such behavior is far from transient and is, in fact, extremely stable over the 4 or so year horizon that a given school has to deal with. Bullies are sadistic animals and they do not respond to being ignored, or appeals to reason. Violence, however, surgically but intensely applied, had a 100% success rate. You have to speak to them in a language that they understand.
If this game works out, these guys are going to be very sad pandas.
They've been active for a few years now, producing computer-aided practice/scoring software for a variety of instruments, and voice. The computer knows what sounds are supposed to be produced, takes MIC input, crunches it into a reasonably meaningful delta(or, rather, series of deltas over time, so that the instructor can see where the student is or isn't having difficulty with a given piece). It is heavily geared toward schools, with lots of "class management" and grading features; but the "cloud" portion of the operation appears to be run by amateurs, and the interface is... utilitarian.
If Ubisoft can do a slick, mass-market version(even if only for a single instrument, at present) they'll likely have something with a lot more polish and shiny bits, as well as an engine that could be adapted to other instruments if the market so demands. That will likely put nontrivial pressure on the SmartMusic guys...
While I have no doubt that some power-tripping admin will manage to fuck it up somewhere, there is the convenient (in this context) fact that most schoolyard harassment is garden-variety apolitical nastiness, the suppression of which imperils the constitution approximately as much as your angry neighbors forcing you to keep your 3am party to below a certain level of noise...
Slashdot Mirror
The fact that it can muck up cheap-n-cheerful 'zillion sites on one host' arrangements unless everybody's TLS ducks are in a row probably counts as a fairly significant cost, as well. Especially on the low end.
It is fairly hard to make a public-interest argument in favor of using scarce(note to spread spectrum and other tech-trick enthusiasts: yes, how limited it is strongly depends on how smart you are about it; but it is finite) spectrum for high-bandwidth broadcasts.
There are the "public information dissemination"/"cultural goods" arguments, which are reasonably strong; but also amply satisfied by even AM quality voice and smeary NTSC quality video. More video bandwidth is certainly better; but that "better" is simply an aesthetic improvement, not a matter of any significant interest(especially in a world where the bandwidth of blu-ray+USPS is so damn high. There is a value in people being able to get current news/political events/hazard warnings in real time; which is a broadcast specialty; but there is no need to allocate enough spectrum so that they can count the pores on senator scumweasel's nose. For ~$12, you can get 50GB video entertainment chunks mailed to your door, not to mention consoles and HTPCs and all the other non-broadcast uses of HDTVs)
That said, I'm really not an enthusiast of the "sell it all to Verizon, Ma Bell knows best." theory of spectrum allocation. I'd prefer to see a much stronger support of un or minimally licenced data-transmission spectrum, along the lines of wifi; but with spectrum that doesn't totally blow. Even laboring under those restrictions, wifi has been an amazing success, and the possibilities of future minimal-licensing wireless are really much more compelling than "another bunch of TV channels" or "200mb/s TO YOUR CELLPHONE* *Capped at 5GB/month, overage $10/GB, you're damn right other terms and restrictions apply, see 2,000 pages of fine print for details.*"
And hand the spectrum over to the next generation of 802.11b/g/n-esque applications.
Even confined to a couple of really sucky blocks of spectrum, the success of no-license-to-deploy, inexpensive wireless data standards has been extraordinary. Why not murder a few bloated, feckless, incumbents and hand over some proper spectrum for this proven and extremely useful application?
I'm pretty sure that Original Sin is basically a shrinkwrap EULA that you "agree" to by being born.
Attempts to explain the concept of "contract of adhesion" to a vengeful iron age deity have, as yet, been unsuccessful....
Or the cost of biofuels... Humans contain enough lipids that converting them into fuel shouldn't be too difficult.
It's more or less undeniable that we are going to run out of low-hanging fruit as time goes on(it is, after all, entirely reasonable to take advantage of the most accessible potential improvements to your technology, which necessarily leaves you with improved technology and a harder set of future improvements). Just imagine how disappointed somebody upgrading from a vacuum tube based system to an IC based one would be with the rate of progress at any point from then on...
The part where I think Kaku goes right off the deep end is where he predicts dire catastrophe because of this. The notion that there may be certain hard limits, based on the increasingly unhelpful properties of matter as you start working with less of it, isn't a secret to anybody, even if we don't know precisely where the hard limit is waiting... Even when we hit it, all our existing chips will still work, and transistors will still be crazy cheap. They just won't keep reliably getting cheaper at high speed.
To go with a historical analogy, Kaku makes me imagine somebody during the industrial revolution, standing on a soapbox and shouting about how "Our steam engines cannot keep improving forever, no matter how clever our machinists and engineers, we will hit the Carnot limit for heat engines, and all progress must cease! Disaster! Calamity!" In terms of the physics, that isn't wrong: thermodynamics does place certain hard limits on the efficiency of ideal heat engines. You can tighten your cylinder tolerances and lubricate away friction and insulate all you like; but that's the limit. However, it doesn't follow that no longer being able to expect next year's heat engines to be more efficient than last year's means doom. It just means that you won't keep getting nicer heat engines automatically...
Among other factors, the Concorde was arguably positioned outside the limits of speed / air-resistance economics... If you look at more recent work in aircraft design, the two big trends are large, high-efficiency, slightly subsonic designs that allow airlines to link major hubs at low cost per unit weight and/or floor space(freeing them to trade off between selling large numbers of cattle-class tickets, or smaller numbers of 'you get your own damn living room' class tickets, or a combination, with minor interior remodeling) and relatively cheap smaller models to cover the peripheral airports that bring travellers to the major hubs at fairly low cost.
The Concorde is a bit of a white elephant because, due to the engineering constraints of supersonic flight, it is actually pretty horrid inside(only four seats wide, limited overhead compartments, etc.) but extremely expensive. Particularly with the advances in mobile computing, cellular/satellite data links, and so forth(and, on the downside, advances in being hassled by rentacops at the airport), it is kind of a hard sell even for the money-is-no-object crowd for which it is the best economic fit. The big subsonic airliners can offer you a fairly roomy and comfortable environment, with electricity and bandwidth, and a reclining chair and such, for less than the Concorde can offer a faster; but rather more spartan, trip. Unless we see a fairly fundamental shakeup in what business travelers want, it is difficult to imagine enough of them choosing fast-and-spartan to bring the economies of scale even remotely within reach of the "it's huge, and fuel efficient, and you can swiftly adjust the ratio between luxury class and cattle class without changing airframes" designs that are preferred today.
While the envelope has continued to be pushed in the less cost-constrained military designs, we've actually seen some orphan high-speed exotics there as well. The SR-71, another sweet looking but hugely expensive high speed design, is out of a job now that a mixture of satellites and long-loiter robotic cheapies are doing surveillance jobs. It would appear that very high speed atmospheric operation, while cool and possible, often isn't worth the trouble.
I agree that the doomtastic prophecies seem rather overblown. If nothing else, the fact that it is (comparatively) easy to make predictions about the endpoint of Moore's law, and the approximate performance thereof, makes being taken by surprise harder. And surprise, not limitations in themselves, is where complex systems really tend to bite it. We know how large atoms are, we have predictions of varying pessimism regarding how many you need to make a transistor that will actually work, and work often enough that you can fabricate chips out of them in some economically viable way. From there, you can calculate the approximate transistor density of the various endpoints. If those crazy kids in engineering happen to out-perform you, great. If not, that sucks; but you still know where you are.
Then there is the fact that only certain applications need ever more power. The scientific computing and simulations people are certainly happy to soak up every cycle you can afford to throw at them; but many other applications reach diminishing returns at much lower levels. The end of Moore's law won't suddenly make those stop working, we just won't be able to shrug and say "eh, it'll be twice as fast and half the cost next year." anymore. If our economy is on such a razor's edge that that is going to throw us into global chaos, we have broader issues...
Arguably, the hardocre apocalyptic gun-toters know that they don't really need survival kits: All they need is a list of nearby people who have survival kits, and their existing supplies of guns and ammunition...
The opportunity cost of the required capital is a "cost of production". It would appear that we differ only in wording in that regard. If capital is fleeing a sector, even if the accountants are putting the sector in the black, it isn't breaking even overall.
Why would a credit card issuer feel threatened by somebody who is trying to increase the convenience and reduce the cost of accepting their credit cards? Some of the bloated incumbent processors are evidently shitting bricks; but why would Visa object to a company whose product offering should markedly increase the number of people accepting credit cards, rather than cash/checks only...
You might want to take a look at the financial data that banks and similar institutions are already required to divulge...
In theory, a competitive market will result in the price of a commodity or service equalling its marginal cost of production.
The fact that the amount the CC guys skim off the top of transactions(in addition to their loan-sharking business) has barely budged despite what are almost certainly declining costs leads one to suspect that the market may not be especially competitive.
The phrase "cosy duopoly", in fact, tends to spring to mind...
Until the classic American Psycho business card scene fades from our cultural memory, business cards will not quite be obsolete...
I just let the invisible hand provide my orbital lift capacity...
Their core tech is actually pretty cool(at least to my layman-when-it-comes-to-music-theory-and-FFTs eyes watching it build a deviation/time graph of the recorded student performance vs. the correct performance is a very neat trick); but pretty much all the 'peripheral' stuff(y'know, unimportant things like 'UI' and 'Not crashing' and 'having a search function that works') sucks pretty badly.
If Ubisoft has an adequately competent core, with the spit and polish given to (most) games that aren't headed right for the bargain bin, it isn't going to be pretty...
I can only assume that a nontrivial chunk of the price tag is for the "expertise required to safely approach a moving satellite and introduce additional fuel, without crashing into it, breaking off any important solar panels/antennas/widgety bits, or otherwise mucking it up.
Sort of a very high end version of the classic techie contractor invoice: "'Typing a one line command, $1' 'Knowing which command to type, $400/hr+travel'"
Clearly the liberals who wouldn't let us drill ANWR are responsible...
There is a little more nuance than that: Obviously, the onsite engineers are not "saving face"(if anything, they are running the nontrivial risk of having face slough off most unpleasantly) and the Japanese government also seems to have come round to the severity of the situation(if not quite to the degree that experts from other countries with lots of reactors would like). Every statement from plant management, though, seems to be a variation on "Golly shucks, there are definitely some issues; but it is hard to say exactly how severe they are. We are working on it, check back later." It doesn't give you the warm and fuzzies about how forthright they are being...
There is a certain amount of reason to suspect that they were packing the spent-fuel ponds rather tighter than would really be advisable in an area where damage to the powered systems is known to be plausible...
The reactors themselves, the really complex bits, seem to be doing comparatively adequately; but the condition of the fuel ponds seems pretty dodgy(since, after all, simply spreading the spent assemblies out more, or having bigger ponds, while more costly, would have been a trivial way to increase the safety margin with minimal engineering complexity.
Nuclear reactors are hard. Nuclear fuel ponds are safety-critical applications of swimming pool technology...
I'm not quite sure where you are projecting from; but my comment was a quite narrowly focused observation that childhood bullying was of the same 1st amendment salience as late-night disturbance of the peace: that is, extraordinarily minimal and well below the level of concern.
Now, that said, I have approximately zero confidence that school admin will be able to do much of anything based on this new mandate. In my school experience, adult authority figures were, without exception, useless or worse in dealing with bullies. I doubt that they've improved too much, and now their mandate is supposed to extend to the internet? Good luck with that one, guys.
However, I really must take exception to your tedious argument-from-cliche and your extraordinarily optimistic take on the level of persistence shown by bullies. Again, in my(admittedly anecdotal) experience, such behavior is far from transient and is, in fact, extremely stable over the 4 or so year horizon that a given school has to deal with. Bullies are sadistic animals and they do not respond to being ignored, or appeals to reason. Violence, however, surgically but intensely applied, had a 100% success rate. You have to speak to them in a language that they understand.
If this game works out, these guys are going to be very sad pandas.
They've been active for a few years now, producing computer-aided practice/scoring software for a variety of instruments, and voice. The computer knows what sounds are supposed to be produced, takes MIC input, crunches it into a reasonably meaningful delta(or, rather, series of deltas over time, so that the instructor can see where the student is or isn't having difficulty with a given piece). It is heavily geared toward schools, with lots of "class management" and grading features; but the "cloud" portion of the operation appears to be run by amateurs, and the interface is... utilitarian.
If Ubisoft can do a slick, mass-market version(even if only for a single instrument, at present) they'll likely have something with a lot more polish and shiny bits, as well as an engine that could be adapted to other instruments if the market so demands. That will likely put nontrivial pressure on the SmartMusic guys...
Are we supposed to be feeling sorry for you, oh noble martyr on the altar of freedom of expression?
While I have no doubt that some power-tripping admin will manage to fuck it up somewhere, there is the convenient (in this context) fact that most schoolyard harassment is garden-variety apolitical nastiness, the suppression of which imperils the constitution approximately as much as your angry neighbors forcing you to keep your 3am party to below a certain level of noise...
There are some very pricey earth-facing CCDs, behind sophisticated optics, in earth orbit. Be a pity if any of them were to catch fire...