This is somewhat orthogonal to your overall point; but the US actually does a substantial percentage of pulp and paper production domestically. We've got plenty of land suitable for growing paper-grade timber, plenty of riverfront space for siting mills, and the economics of shipping low end lumber and mundane paper products long distances aren't all that thrilling compared to the relatively modest premium you pay for domestic employees.
The morbid entertainment value is at least doubled by the fact that the article is about the hypothetical 'chinaman haxxor!!!!!' threat to the American military supply chain, rather than the much less hypothetical 'a mixture of ill-advised outsourcing and blatant regulatory capture has gone a fair portion of the way toward ensuring that the phrase 'military supply chain' refers to the route by which public money makes its way into the coffers of contractors, rather than any mechanism actually designed to improve or maintain American military capability.
With friends like Duke Cunningham and KBR, we don't need enemies...
Light field stuff in general isn't(you are always limited to a substantially lower resolution than your available sensor(s) would suggest, because the technology requires multiple sensor pixels to be mapped to each microlens; but getting larger and higher resolution digital sensors is one of those problems that can be solved by writing sufficiently large checks); but this particular camera is. Presumably, to hit the $500 price point.
You get get to compute your choice of 1 megapixel images at various focuses; but no one output shot is higher than 1024x1024...
Being able to focus after the fact feels like science fiction; but the ability of contemporary cameras to dump several images per second even on the low end, and simply change focus slightly between each one, means that the brute-force-and-ignorance approach nips dangerously close behind for reasonably static subjects.
I'll admit that 'the optics' was colloquial; but the microlens array is the part of the path where the mapping of multiple CCD pixels to a single available output pixel happens.
As you say, the technology only works if your sensor has enough pixels behind each microlens(and, since very high resolution digital sensors and the necessary supporting processor and storage are more expensive than very fine-grained polymer microlens arrays, the economic limits on the sensor they could afford presumably drove the choice of the microlens array they put on top of it), so the lightfield result will necessarily be much lower resolution than the conventional result; but the microlens optics are the operational location where that reduction happens...
The capabilities of light field cameras have that fun 'technology indistinguishable from magic' touch to them that the impressive-but-evolutionary spec bumps of markedly superior conventional digital cameras don't(It's like playing with your favorite eccentric retro computer from before the Great Standardization: at this point, anything that old is a painfully limited toy; but it is different. Your top-of-the-line-screaming-monster of a PC, on the other hand, is brutally capable and impressively cheap; but practically point-for-point familiar to the p90 running Windows95, with all the performance related numbers bumped by a few decimal places).
Unfortunately, though, the move to release it at a (barely) 'consumer toy' price point really led to a product slightly too compromised to be useful: The optics you need for the light field capture eat so much of the sensor's available resolution that the resolution of the images you can get out of the thing is hovering slightly below 1 megapixel. Yes, the ability to spit out that paltry image at all sorts of focuses, after the fact, is damn cool; but for $500, you could get a high end P&S that could iterate through a series of 10MP shots at different focus points, at time of shooting in a few seconds, netting much of the benefit along with resolutions that wouldn't be ashamed to show up on a $20 webcam.
I'd love to see the same technology applied at a price point and form factor where the sheer sacrifice of available pixels wouldn't be so keenly felt.
There is arguably one major gap in the analogy with vaccines(unless your plan includes it but simply didn't state it):
The vaccine injury system is designed to deal, as efficiently as messy compromise allows, with the fact that vaccines(as with other drugs and procedures) tend to have risks that show up at the population level that couldn't have been detected in clinical trials of any feasible size and/or are substantially lower than their benefits. The logic is that these cases have victims deserving of compensation; but arise without culpable negligence or malice.
It doesn't, and isn't intended to, cover other risk/liability issues arising in medicine that incidentally involve vaccines. If, say, your doctor stored a vaccine improperly and administered a contaminated or spoiled dose, that wouldn't be a vaccine injury, that'd be malpractice that happened to involve a vaccine rather than some other drug. In such a case, the damages would be partially to compensate you and partially to punish them; because there are both damages and culpable negligence or malice at play.
In the case of an autonomous car, the 'vaccine analogous' set of risks/compensations would only cover the set of risks inherent to the system's operation(corner cases where physics simply doesn't allow for a safe solution on the navigational system's part, system defects sufficiently rare and esoteric to have escaped reasonable diligence on the manufacturer's part, and so forth). It wouldn't usefully cover negligence on the part of either the manufacturer(in, say, corner-cutting on testing or design of safety critical systems) or the operator(operating a vehicle despite sensor or system faults, defeating safety-critical systems in order to achieve faster trips, etc.)
When dealing with small, essentially unavoidable, risks there is a strong logic in favor of efficient compensation purely on the basis of injury(assuming that those risks carry benefits sufficient to justify their broad imposition...); but one must be careful not to immunize negligence and malice in a system designed to handle mere accident...
I suspect that there will be fewer impaired computers than there will be impaired drivers; but I suspect that operators running cars with the sensor equivalent of shot breaks and dead turn signals will hardly be unknown, and corner cutting by some manufacturer or other is just a matter of time.
Probably depends on what it was being tested for the treatment of.
Risk tolerance certainly drifts around as a function of time, and the FDA is no exception; but there ends up being some vaguely linear relationship between the ghastliness of the condition being approached and how much slack the assorted side effects and risks of the therapy end up being cut.
Given that alcoholism tends to have pretty dramatic long-term effects on people, many of them pretty nasty, and presently has a lousy cure rate, it might actually have a shot. It would certainly have plenty of company among the 'potentially very unpleasant drugs for definitely very dire indeed psych conditions' that are currently legal, approved, and commonly accepted for use. How much flack it would draw from the 'all you need is more willpower!' school of largely ineffective but morally satisfying therapy would be a different question... Its chances for less serious diagnoses would probably be much poorer, and (depending on what classification it hypothetically received) even off-label use might be strongly discouraged by enthusiastic DEA oversight.
I wouldn't necessarily trust the vendor further than I could throw them "Yeah, sorry guys, our testing indicates that our product is broken and we need to replace it under warranty". In the context of criminal justice, where we use the 'adversarial' legal system under the theory that the contesting sides provide the best chance of achieving the correct answer, it would seem more appropriate for the testing and calibration of forensic apparatus and technique should really be the job of an independent entity whose performance is judged on the basis of how effectively they represent the 'accuracy interest' of the apparatus and technique...
I'm guessing(and the fact that nobody noticed that the calibration data were eerily perfect and uniform, because individual units were just getting cut-and-paste numbers re-enforces my suspicion) that there is no professional accolade to be had for being 'that guy who is always pointing out problems with the gear' within the police department. Ideally, you'd want the department of exacting assholes to be in charge of testing the stuff, and distinct from the officers in charge of using it...
I'm pretty sure that merely being fired for tampering with the evidence in criminal cases is absurdly generous...
At least if you make the mistake of being little people, that is Serious Felony For Real territory. Firing would be step one, followed by prosecution and hard time.
My understanding, admittedly layman's, is that a lot(all of?) the early console systems were hardware constrained as hell and certainly didn't have the luxury of a hardware RNG or even many peripheral interfaces with the unpredictable outside world, so their games tended to not only lack randomness; but rely on techniques for faking randomness sufficiently rudimentary that "luck manipulation" becomes viable...
Everybody who doesn't have a personal analytics database under their control and does have a Facebook 'timeline' knows the answer to this question already. And the ones who don't know the answer are the answer...
Nothing in the budget for that; but the wonderful folks in the advertising and intelligence sectors will be glad to offer an array of custom-tailored consumer products and law enforcement solutions...
Thankfully, this area of their licensing isn't my problem; but I'm pretty sure that 'Windows 7 Enterprise' is about as consumer level as 2k8 is. Similarly, the sauce that VMware and friends want to sell you for conveniently shoving individual VMs at users isn't exactly something you pick up at Best Buy...
MS will let you farm out 2k8 RDS sessions, with essentially all the trimmings that Win7 would be capable of over RDP, for the right money; but they simply Will Not Sell an SPLA to perform the (with contemporary virtualization and deduplicated storage backends) virtually identical act of farming out Win7 VMs.
I honestly find it rather puzzling. If they didn't offer 'desktopish' SPLAs at all, that'd be unpleasant of them; but would be a coherent 'no way are we letting thin clients take over' strategy. If they followed a 'we don't care how you do it, we just want to get paid per month, per seat' approach, that'd be similarly coherent.
As it is, though, there just doesn't seem to be a coherent logic behind the licensing terms.
In retrospect, it is darkly humorous that the 'cypherpunks' of the 80's and 90's thought that strong cryptography would be a force in favor of either freedom or privacy...
My point was that, by insisting on DRM(which makes lock-in to a specific 'ecosystem' trivial and legally enforced), the publishers made it possible for somebody to 'corner' the market at all:
Scenario 1: Ebooks are sold without DRM. Bezos purchases books at wholesale price and sells them at wholesale price - x. Everybody buys from Amazon because Amazon is basically paying them $x to do so. Publishers get wholesale price/copy. Now, if Amazon attempts to turn the screws and raise the price, people lose interest and walk away because they can just as easily buy from B&N and read on their Kindle or read their already purchased texts on another device, or whatever.
Scenario 2: Ebooks are sold with DRM. Bezos purchases books at wholesale price and sells them at wholesale price -x. Everybody buys from Amazon, etc. Now if Amazon decides to turn the screws, they have a locked-in audience who cannot legally read their existing texts outside of Amazon's garden, nor can they purchase texts from elsewhere and read them on their existing hardware. Now buying market share and then turning the screws has value...
That's my point: Since the publisher gets wholesale price either way, Amazon's willingness to lose money isn't money out of their pockets(it is, in fact, a subsidy for their product) unless Amazon can erect barriers to entry by losing money initially. With DRM, barriers to entry are easy(indeed, since it doesn't dissuade pirates, it's one of the few things that DRM is good for). Without it, barriers to entry are a great deal harder.
My point was not that Amazon was playing nice. I have no reason to think that well of them. My point was that, in the absence of DRM, the costs of taking your texts and walking away, or staying but buying elsewhere, are very low. Thus, the value of dominant market share is low. In the presence of DRM, the value of dominant market share is high; because the cost of leaving your walled garden once you've entered is high, which allows those who have been brought in to be squeezed.
If anything, while the hardware requirements of being a true nut have climbed a bit(SLI/Crossfire and motherboards with 4 or more PCIex16 slots certainly make it possible to go overboard in fine style, along with your $1k processor and SSD array..), the price premium of adequate 'gamer' hardware vs. 'a boring computer' hardware has been extremely modest of late.
The total cost of the computer is still a bit higher than the cost of the console(though the games are often cheaper, if you don't insist on having them all on day 1); but the delta between the computer you probably need already and the same computer with a modest GPU and maybe some extra RAM is peanuts...
Slashdot Mirror
This is somewhat orthogonal to your overall point; but the US actually does a substantial percentage of pulp and paper production domestically. We've got plenty of land suitable for growing paper-grade timber, plenty of riverfront space for siting mills, and the economics of shipping low end lumber and mundane paper products long distances aren't all that thrilling compared to the relatively modest premium you pay for domestic employees.
The morbid entertainment value is at least doubled by the fact that the article is about the hypothetical 'chinaman haxxor!!!!!' threat to the American military supply chain, rather than the much less hypothetical 'a mixture of ill-advised outsourcing and blatant regulatory capture has gone a fair portion of the way toward ensuring that the phrase 'military supply chain' refers to the route by which public money makes its way into the coffers of contractors, rather than any mechanism actually designed to improve or maintain American military capability.
With friends like Duke Cunningham and KBR, we don't need enemies...
Light field stuff in general isn't(you are always limited to a substantially lower resolution than your available sensor(s) would suggest, because the technology requires multiple sensor pixels to be mapped to each microlens; but getting larger and higher resolution digital sensors is one of those problems that can be solved by writing sufficiently large checks); but this particular camera is. Presumably, to hit the $500 price point.
You get get to compute your choice of 1 megapixel images at various focuses; but no one output shot is higher than 1024x1024...
Being able to focus after the fact feels like science fiction; but the ability of contemporary cameras to dump several images per second even on the low end, and simply change focus slightly between each one, means that the brute-force-and-ignorance approach nips dangerously close behind for reasonably static subjects.
I'll admit that 'the optics' was colloquial; but the microlens array is the part of the path where the mapping of multiple CCD pixels to a single available output pixel happens.
As you say, the technology only works if your sensor has enough pixels behind each microlens(and, since very high resolution digital sensors and the necessary supporting processor and storage are more expensive than very fine-grained polymer microlens arrays, the economic limits on the sensor they could afford presumably drove the choice of the microlens array they put on top of it), so the lightfield result will necessarily be much lower resolution than the conventional result; but the microlens optics are the operational location where that reduction happens...
The capabilities of light field cameras have that fun 'technology indistinguishable from magic' touch to them that the impressive-but-evolutionary spec bumps of markedly superior conventional digital cameras don't(It's like playing with your favorite eccentric retro computer from before the Great Standardization: at this point, anything that old is a painfully limited toy; but it is different. Your top-of-the-line-screaming-monster of a PC, on the other hand, is brutally capable and impressively cheap; but practically point-for-point familiar to the p90 running Windows95, with all the performance related numbers bumped by a few decimal places).
Unfortunately, though, the move to release it at a (barely) 'consumer toy' price point really led to a product slightly too compromised to be useful: The optics you need for the light field capture eat so much of the sensor's available resolution that the resolution of the images you can get out of the thing is hovering slightly below 1 megapixel. Yes, the ability to spit out that paltry image at all sorts of focuses, after the fact, is damn cool; but for $500, you could get a high end P&S that could iterate through a series of 10MP shots at different focus points, at time of shooting in a few seconds, netting much of the benefit along with resolutions that wouldn't be ashamed to show up on a $20 webcam.
I'd love to see the same technology applied at a price point and form factor where the sheer sacrifice of available pixels wouldn't be so keenly felt.
You do realize that accusing somebody of being One Of The Enemy is not actually a form of rebuttal, right?
There is arguably one major gap in the analogy with vaccines(unless your plan includes it but simply didn't state it):
The vaccine injury system is designed to deal, as efficiently as messy compromise allows, with the fact that vaccines(as with other drugs and procedures) tend to have risks that show up at the population level that couldn't have been detected in clinical trials of any feasible size and/or are substantially lower than their benefits. The logic is that these cases have victims deserving of compensation; but arise without culpable negligence or malice.
It doesn't, and isn't intended to, cover other risk/liability issues arising in medicine that incidentally involve vaccines. If, say, your doctor stored a vaccine improperly and administered a contaminated or spoiled dose, that wouldn't be a vaccine injury, that'd be malpractice that happened to involve a vaccine rather than some other drug. In such a case, the damages would be partially to compensate you and partially to punish them; because there are both damages and culpable negligence or malice at play.
In the case of an autonomous car, the 'vaccine analogous' set of risks/compensations would only cover the set of risks inherent to the system's operation(corner cases where physics simply doesn't allow for a safe solution on the navigational system's part, system defects sufficiently rare and esoteric to have escaped reasonable diligence on the manufacturer's part, and so forth). It wouldn't usefully cover negligence on the part of either the manufacturer(in, say, corner-cutting on testing or design of safety critical systems) or the operator(operating a vehicle despite sensor or system faults, defeating safety-critical systems in order to achieve faster trips, etc.)
When dealing with small, essentially unavoidable, risks there is a strong logic in favor of efficient compensation purely on the basis of injury(assuming that those risks carry benefits sufficient to justify their broad imposition...); but one must be careful not to immunize negligence and malice in a system designed to handle mere accident...
I suspect that there will be fewer impaired computers than there will be impaired drivers; but I suspect that operators running cars with the sensor equivalent of shot breaks and dead turn signals will hardly be unknown, and corner cutting by some manufacturer or other is just a matter of time.
It's been tried. I only wish I were joking about that one.
Probably depends on what it was being tested for the treatment of.
Risk tolerance certainly drifts around as a function of time, and the FDA is no exception; but there ends up being some vaguely linear relationship between the ghastliness of the condition being approached and how much slack the assorted side effects and risks of the therapy end up being cut.
Given that alcoholism tends to have pretty dramatic long-term effects on people, many of them pretty nasty, and presently has a lousy cure rate, it might actually have a shot. It would certainly have plenty of company among the 'potentially very unpleasant drugs for definitely very dire indeed psych conditions' that are currently legal, approved, and commonly accepted for use. How much flack it would draw from the 'all you need is more willpower!' school of largely ineffective but morally satisfying therapy would be a different question... Its chances for less serious diagnoses would probably be much poorer, and (depending on what classification it hypothetically received) even off-label use might be strongly discouraged by enthusiastic DEA oversight.
I wouldn't necessarily trust the vendor further than I could throw them "Yeah, sorry guys, our testing indicates that our product is broken and we need to replace it under warranty". In the context of criminal justice, where we use the 'adversarial' legal system under the theory that the contesting sides provide the best chance of achieving the correct answer, it would seem more appropriate for the testing and calibration of forensic apparatus and technique should really be the job of an independent entity whose performance is judged on the basis of how effectively they represent the 'accuracy interest' of the apparatus and technique...
I'm guessing(and the fact that nobody noticed that the calibration data were eerily perfect and uniform, because individual units were just getting cut-and-paste numbers re-enforces my suspicion) that there is no professional accolade to be had for being 'that guy who is always pointing out problems with the gear' within the police department. Ideally, you'd want the department of exacting assholes to be in charge of testing the stuff, and distinct from the officers in charge of using it...
You sound like one of those pointy-headed intellectuals who get all worked up about 'accurately' 'measuring' 'external reality'...
I'm pretty sure that merely being fired for tampering with the evidence in criminal cases is absurdly generous...
At least if you make the mistake of being little people, that is Serious Felony For Real territory. Firing would be step one, followed by prosecution and hard time.
My understanding, admittedly layman's, is that a lot(all of?) the early console systems were hardware constrained as hell and certainly didn't have the luxury of a hardware RNG or even many peripheral interfaces with the unpredictable outside world, so their games tended to not only lack randomness; but rely on techniques for faking randomness sufficiently rudimentary that "luck manipulation" becomes viable...
If you consider the existing body of ad-hoc, largely commercially driven, data collection mechanisms, you don't really need to use the future tense...
Everybody who doesn't have a personal analytics database under their control and does have a Facebook 'timeline' knows the answer to this question already. And the ones who don't know the answer are the answer...
He's sure that he's a cellular automaton, this is all just an elaborate plan to get somebody to work out his rule number for him...
Nothing in the budget for that; but the wonderful folks in the advertising and intelligence sectors will be glad to offer an array of custom-tailored consumer products and law enforcement solutions...
Thankfully, this area of their licensing isn't my problem; but I'm pretty sure that 'Windows 7 Enterprise' is about as consumer level as 2k8 is. Similarly, the sauce that VMware and friends want to sell you for conveniently shoving individual VMs at users isn't exactly something you pick up at Best Buy...
That's what is so weird:
MS will let you farm out 2k8 RDS sessions, with essentially all the trimmings that Win7 would be capable of over RDP, for the right money; but they simply Will Not Sell an SPLA to perform the (with contemporary virtualization and deduplicated storage backends) virtually identical act of farming out Win7 VMs.
I honestly find it rather puzzling. If they didn't offer 'desktopish' SPLAs at all, that'd be unpleasant of them; but would be a coherent 'no way are we letting thin clients take over' strategy. If they followed a 'we don't care how you do it, we just want to get paid per month, per seat' approach, that'd be similarly coherent.
As it is, though, there just doesn't seem to be a coherent logic behind the licensing terms.
Luckily, neither brass nor bronze are of any value for either holding an edge or inflicting blunt trauma...
And no need to even mention ceramics or the fancier plastics, that shit is totally harmless.
You wouldn't want to lose you access, would you?
In retrospect, it is darkly humorous that the 'cypherpunks' of the 80's and 90's thought that strong cryptography would be a force in favor of either freedom or privacy...
My point was that, by insisting on DRM(which makes lock-in to a specific 'ecosystem' trivial and legally enforced), the publishers made it possible for somebody to 'corner' the market at all:
Scenario 1: Ebooks are sold without DRM. Bezos purchases books at wholesale price and sells them at wholesale price - x. Everybody buys from Amazon because Amazon is basically paying them $x to do so. Publishers get wholesale price/copy. Now, if Amazon attempts to turn the screws and raise the price, people lose interest and walk away because they can just as easily buy from B&N and read on their Kindle or read their already purchased texts on another device, or whatever.
Scenario 2: Ebooks are sold with DRM. Bezos purchases books at wholesale price and sells them at wholesale price -x. Everybody buys from Amazon, etc. Now if Amazon decides to turn the screws, they have a locked-in audience who cannot legally read their existing texts outside of Amazon's garden, nor can they purchase texts from elsewhere and read them on their existing hardware. Now buying market share and then turning the screws has value...
That's my point: Since the publisher gets wholesale price either way, Amazon's willingness to lose money isn't money out of their pockets(it is, in fact, a subsidy for their product) unless Amazon can erect barriers to entry by losing money initially. With DRM, barriers to entry are easy(indeed, since it doesn't dissuade pirates, it's one of the few things that DRM is good for). Without it, barriers to entry are a great deal harder.
My point was not that Amazon was playing nice. I have no reason to think that well of them. My point was that, in the absence of DRM, the costs of taking your texts and walking away, or staying but buying elsewhere, are very low. Thus, the value of dominant market share is low. In the presence of DRM, the value of dominant market share is high; because the cost of leaving your walled garden once you've entered is high, which allows those who have been brought in to be squeezed.
Just leave at least one wireless interface active and I'll handle all the updates for you!
Sincerely, B. Hat,
Honest Gentleman
If anything, while the hardware requirements of being a true nut have climbed a bit(SLI/Crossfire and motherboards with 4 or more PCIex16 slots certainly make it possible to go overboard in fine style, along with your $1k processor and SSD array..), the price premium of adequate 'gamer' hardware vs. 'a boring computer' hardware has been extremely modest of late.
The total cost of the computer is still a bit higher than the cost of the console(though the games are often cheaper, if you don't insist on having them all on day 1); but the delta between the computer you probably need already and the same computer with a modest GPU and maybe some extra RAM is peanuts...