I find the claim that they were able to make people unable to tell the truth much more surprising than the one that they were able to make people unable to lie.
While fun and useful, lying is somewhat cognitively demanding: You have to synthesize and deliver a contracfactual statement, you can't just remember it because it didn't happen. There has been some previous speculation that you should be able to detect lying, based on the greater mental effort(and distributed across more brain regions effort) involved, vs. the recall activity required to tell the truth.
That you can knock-out truth-telling(without just inducing aphasia or amnesia temporarily, which is a bit heavy handed) is much more surprising.
It would definitely make T-Mobile even less happy; but bittorrent is actually an excellent choice for coping with every aspect of that except the tiny bandwidth bit...
I suspect that such a simplified benchmark would work just fine; but would be of interest to relatively few people. Most users of graphics cards either don't care at all, and have integrated graphics, don't care at all about theoretical performance; but do care about sniping n00bs in Medal of Halo 3, or are GPU compute users, who have their own quite specific demands.
Even in the realm of game benchmarking, you can see some pretty dramatic differences, between engines, in how Nvidia's approach or ATI's approach stacks up for a given generation of cards. It has been a while since relatively abstract tests of generic capabilities have been able to provide much insight into how a card will stack up in what you want it to do...
According to the writeup on techcrunch the service is quite high level: you SMS them a URL(through the app), they grab the webpage, bundle it up, and MMS it back for display. It isn't clear that that would even handle some of the dynamic/login-required stuff, much less do so in a manner that doesn't involve revealing your credentials to them.
It's a pity, I was hoping for a more elegant hack, something along the lines of a VPN-style tunneling of arbitrary TCP/IP traffic encapsulated(in this case) in SMS messages. Could be that the unpredictable and sometimes extremely high latency of SMS dooms that one, though...
From the perspective of efficiency or architectural sanity, that is about as far from optimal as you could wish to be(short of running the fastest analog modem connection that will survive GSM voice compression to take advantage of your unlimited voice minutes); but the magic of telco nonsense pricing makes it entirely reasonable.
Hopefully getting their control channel hammered with SMS noise will induce them to offer some sort of reasonably priced modest-speed data mechanism that isn't a horrible pile of hack...
Incidentally, of course, does this lovely mechanism make whoever runs "Smozzy" a MiTM even within SSL-wrapped browsing sessions, or does the TCP/IP->SMS insanity just wrap the packets whole and serve as a peculiar sort of link layer?
This will certainly make benchmarking a bit more complex. One hopes that the gamers like going back to stats class.
You'll need the FPS value, as before, (ideally with a worst-case FPS reported); but you'll also want a measure of the deviation of every frame's draw time from the average draw time being reported. And likely a measure of how atypically bad frames are distributed(ie. 5 seconds of super-low framerate during some sort of loading is annoying. 20 25 millisecond frames scattered throught action-heavy areas is really annoying...)
It would also be interesting to see what this does to the (traditionally poor) reputation of the sucker-edition cards that get loaded up with relatively huge amounts of slow memory in order to make them seem like a good deal(ie. if 2GB of GDDR5 is the lunatic fringe, and 512MB of GGDR5 is the solid-value-gamer special, you'll see cards with 1GB of DDR2/3, marketed to the unsuspecting as alternatives to the solid-value line. Their average framerates are usually pretty tepid, because DDR is slow; but they honestly do have a lot of it, so they needn't hit the PCIe bus to load something from system RAM as often...)
You will find that humanity's desire to fill the aching void of its pitiful existence with lolcats and porn is not, in fact, our doing.
Further, because power is one of our major operating costs, you will find that our competitors are unlikely to be able to deliver lolcats and porn appreciably more efficiently than we can.
So, after I'm finished inserting irksome background noise to compensate for any deficiencies an author may have had in terms of showing rather than telling, or deficiencies I may have in reading ability, can I have a smartphone app that detects when I'm in a restaurant and automatically inserts the sound of somebody with an annoying nasal voice having an obnoxious conversation? How about some random honking every time my phone detects that it is going more than 30mph?
Bloody hell, people, if there is one thing that modernity needs like a hole in the head, it is more fucking background noise...
Don't worry: They can just update to the "Heathkit: 21st century skills" collection, where you learn the art and science of hiring Chinese subcontractors to assemble the kit, and the CAD skills necessary to design a case with your branding and logo to contain the finished product.
Based on the interweb's rough figures, the lump-sum cash payout of her 'we-love-you-so-much-we-terminated-you-by-phone' dismissal is equal to only 30 years of the average salary of a Yahoo software engineer, not quite a lifetime. Luckily, there is also a complex mess of stock options(conveniently not subject to those little-people payroll taxes!) that should save her from penury...
Unfortunately, that theory captures an incomplete picture of the founder effect:
Requiring the CEO to buy a chunk of the company can provide them with a greater financial stake in the company's success, or it can just provide them with the incentive to axe the R&D department, pump out a few quarters that Wall Street loves, and give themselves a giant bonus in the form of "shareholder value" before moving on...
If anything, having a CEO without major holdings(and without a "Congratulations, you fucked up!" bonus larger than a peon's lifetime earnings, if that isn't to scary to think about) might actually help ensure that they take the long view; because they don't have the same financial incentive to pump, loot, and leave.
Founders(except of built-for-acquisition jobs) tend to have major holdings and personal emotional investments, and it takes both to make them do what they do...
I wouldn't be surprised if there are some specialty niche application guys who are just drooling at the prospect of vastly increased silicon area without more board space or interconnect hassle; but anybody who is cranking the clock, the power handling, or both, is going to find the utility of the layers at the center a bit dubious.
All you really need to extinguish the warm-and-fuzzies is a stiff dose of the fact that an alarming number of the present and upcoming SoC designs at least optionally include pretty aggressive Tivoization features, opaque black-box functions handled by cryptographically verified and non-replaceable firmware blobs, and not infrequently a driver or two that isn't available in source form and makes keeping the kernel current rather tricky...
You can have all the open source you want; but if you can only run it on x86 whiteboxes and select dev boards, you still have a problem.
Would purchasing one with a printed skin texture based on the appearance of the burning, partially skeletonized, Hindenburg shortly before its fatal plunge be tasteless?
I'm highly doubtful of the efficacy of that. There are numerous business areas, even well-known names, who do just fine despite public opprobrium.
A highschool course in IT taking down Thawte or forcing them to clean up their act would be about as likely as a highschool course in personal finance bringing down Goldman-Sachs or leading to the end of byzantine financial chicanery...
Given that one major thread of Marx's critique of capitalism is an application of textbook free-market economics, it is no more surprising that he managed to predict some of the outcomes of an approximately capitalist economy than it is surprising when any other economist does(ie. it is moderately surprising; but not entirely so).
In a free market, the price of a commodity tends towards that commodity's marginal cost of production. In a capitalist system, labor is a commodity. Ergo, in a free market capitalist environment, the price of labor will tend towards its marginal cost of production, which is to say that laborers will earn bare subsistence(or bare subsistence+cost of education/training if some of that is required).
The stuff about 'alienation' is more debateable, and any of Marx's proposed solutions(which, often, weren't all that clearly formulated because he didn't see a way out, except a quasi-millenial "revolution") are pretty shaky; but Marx as economist was at least as sound as average for the discipline, if not rather sounder...
Child #7,000,000,000 gets the prize of officially being recognized as "Not actually a bundle of joy" and, on average, a harsh subsistence existence. Congratulations!
I was thinking something a bit more directional: ie. same principle as the parachute flare; but with some mechanism for the operator to designate a target or area to be illuminated before launch, so that the blindingly-light-of-a-false-noon would only apply to your opponents(and their now thoroughly freaked-out night sight gear)...
Parachute flares, if memory serves, substantially predate armored vehicles, possibly even internal combustion vehicles of any sort; so I was curious about anything developed in the contemporary 'highly sensitive optical instruments on expensive but extremely dangerous armor' period... With modern vehicles in the multiple millions a pop, I imagine that selective-illumination systems in the hundreds or thousands per shot might be seen as viable, and that kind of budget might give you room for things more interesting than magnesium-on-a-string.
Statistics on cow-fart volume, mass, and composition seem hard to come by; but I'm assuming that they'd show up as a brief, rapidly expanding, cloud of internal-body-temperature. Anything that could pick up a human should pick one up as well.
Considerably smaller than the output of anything much more than a toy engine, though...
Out of curiosity, if turning on your own light is so overtly suicidal, has there been any work on some sort of disposable system(a balloon, a compressed-gas launched parachute-projectile, etc) that would quietly move a moderate and unpredictable distance from the user, and then unleash the actinic glare of whatever chemical light source is currently in vogue in the correct direction?
My (totally uninformed about the actual system; but moderately familiar with TECs) speculation would be that the thermal cameras provide the control system the data it needs to calculate what the tank "should" look like, to blend in with ambient. That(with the option of mixing in a pattern, like "cow" or "car full of innocent children") would be converted to a target temperature value for each of the camouflage hexagon units.
Each hexagon module, in turn, would have a thermal sensor on the outer surface and a TEC layer. If the target temperature is higher than the real temp, fire up the TEC with the exterior as the hot-side until real temp = target temp. If the target temp is lower than real temp, reverse polarity and run the TEC with the outside as the cold side until you hit the target.
TECs can be run to generate energy(if a sufficient thermal delta exists between hot and cold side); but that would only be of use to the modules that are reliably cooled from the outside and reliably heated from the inside, and that don't need to do any active concealing(you can put power in, to maintain a thermal delta, or you can pull power out by breaking down a thermal delta; but you can't have both...) Given the shitty efficiencies of even the nicer commercial units, this system is going to be a net draw on the vehicle's electrical system, and will probably end up producing a thermal exhaust that needs to be obfuscated however engine exhausts and AC system hot sides currently are.
All you have to do is purchase your system with BAE 4-hour gold support(additional charges may apply). Whenever you are hit by small arms fire, explosions, rocks kicked up by the vehicle in front of you, or any of your camoflauge hexagons is exposed to temp>100 degrees C for too long(which causes TECs to start to break down), just whip out your satellite phone and put in a support call! Your rep should be out there in no time, with a supply of replacement panels.
My understanding is that some IEDs have IR triggers(some bodging based on COTS PIR units, I assume). The low-tech countermeasure is to dangle a running toaster from a pole, a safe distance in front of your vehicle. The low tech counter-countermeasure is to move the IR trigger approximately one 'safe distance' from the explosive...
In theory, I don't see why this stuff wouldn't work against those(though not against the other flavors); but I strongly suspect that the sales pitch involves an enemy equipped with sophisticated armor, tank-killing helicopters, or man-portable rockets more sophisticated than RPG retro classics...
TFA uses the phrase "thermo-electric devices". I'm assuming that means some flavor of Peltier(can't be purely resistive; because some of the pixels need to be cooled and some heated). If that is the case, I would be very interested(and probably not cleared) to know how they handle the heat output of the camouflage, along with the engine and other core systems.
Peltiers are really fun devices; because they are all solid state, respond quickly, and can be driven with a simple DC current; but they aren't what you'd call efficient heat pumps. The fact that they work at all feels like magic; but the heat coming out of the hot side is considerably more than the heat being pumped from the cold side... In IC cooling scenarios, a couple square inches of Peltier can easily consume 100watts. I'd assume that this system, with its much greater surface area, and lower deltaTs on average, wouldn't be that bad; but unless BAE has made some real strides with TECs, nontrivial power is going to be involved(amounts varying depending on what is being emulated and how much the trick differs from the vehicle).
How do you dump all the waste heat from such a system? "looks like a cow" is stealthy. "Looks like a cow, with a thermal exhaust plume that suggests it contains a running AGT1500 and a collection of main battle tank support systems" is less stealthy...
On Windows clients, at any rate, Chrome uses the system certificate store, same as IE, so behavior should be identical(barring any ad-hoc bodging in response to particular issues). I don't know what Chrome on other platforms does.
FF does its own thing, independent of the OS/DE provided certificate store, and throws an untrusted certificate warning.
Slashdot Mirror
I find the claim that they were able to make people unable to tell the truth much more surprising than the one that they were able to make people unable to lie.
While fun and useful, lying is somewhat cognitively demanding: You have to synthesize and deliver a contracfactual statement, you can't just remember it because it didn't happen. There has been some previous speculation that you should be able to detect lying, based on the greater mental effort(and distributed across more brain regions effort) involved, vs. the recall activity required to tell the truth.
That you can knock-out truth-telling(without just inducing aphasia or amnesia temporarily, which is a bit heavy handed) is much more surprising.
It would definitely make T-Mobile even less happy; but bittorrent is actually an excellent choice for coping with every aspect of that except the tiny bandwidth bit...
I suspect that such a simplified benchmark would work just fine; but would be of interest to relatively few people. Most users of graphics cards either don't care at all, and have integrated graphics, don't care at all about theoretical performance; but do care about sniping n00bs in Medal of Halo 3, or are GPU compute users, who have their own quite specific demands.
Even in the realm of game benchmarking, you can see some pretty dramatic differences, between engines, in how Nvidia's approach or ATI's approach stacks up for a given generation of cards. It has been a while since relatively abstract tests of generic capabilities have been able to provide much insight into how a card will stack up in what you want it to do...
According to the writeup on techcrunch the service is quite high level: you SMS them a URL(through the app), they grab the webpage, bundle it up, and MMS it back for display. It isn't clear that that would even handle some of the dynamic/login-required stuff, much less do so in a manner that doesn't involve revealing your credentials to them.
It's a pity, I was hoping for a more elegant hack, something along the lines of a VPN-style tunneling of arbitrary TCP/IP traffic encapsulated(in this case) in SMS messages. Could be that the unpredictable and sometimes extremely high latency of SMS dooms that one, though...
A textbook case of perverse incentives...
From the perspective of efficiency or architectural sanity, that is about as far from optimal as you could wish to be(short of running the fastest analog modem connection that will survive GSM voice compression to take advantage of your unlimited voice minutes); but the magic of telco nonsense pricing makes it entirely reasonable.
Hopefully getting their control channel hammered with SMS noise will induce them to offer some sort of reasonably priced modest-speed data mechanism that isn't a horrible pile of hack...
Incidentally, of course, does this lovely mechanism make whoever runs "Smozzy" a MiTM even within SSL-wrapped browsing sessions, or does the TCP/IP->SMS insanity just wrap the packets whole and serve as a peculiar sort of link layer?
This will certainly make benchmarking a bit more complex. One hopes that the gamers like going back to stats class.
You'll need the FPS value, as before, (ideally with a worst-case FPS reported); but you'll also want a measure of the deviation of every frame's draw time from the average draw time being reported. And likely a measure of how atypically bad frames are distributed(ie. 5 seconds of super-low framerate during some sort of loading is annoying. 20 25 millisecond frames scattered throught action-heavy areas is really annoying...)
It would also be interesting to see what this does to the (traditionally poor) reputation of the sucker-edition cards that get loaded up with relatively huge amounts of slow memory in order to make them seem like a good deal(ie. if 2GB of GDDR5 is the lunatic fringe, and 512MB of GGDR5 is the solid-value-gamer special, you'll see cards with 1GB of DDR2/3, marketed to the unsuspecting as alternatives to the solid-value line. Their average framerates are usually pretty tepid, because DDR is slow; but they honestly do have a lot of it, so they needn't hit the PCIe bus to load something from system RAM as often...)
You will find that humanity's desire to fill the aching void of its pitiful existence with lolcats and porn is not, in fact, our doing.
Further, because power is one of our major operating costs, you will find that our competitors are unlikely to be able to deliver lolcats and porn appreciably more efficiently than we can.
Here endeth the justification.
So, after I'm finished inserting irksome background noise to compensate for any deficiencies an author may have had in terms of showing rather than telling, or deficiencies I may have in reading ability, can I have a smartphone app that detects when I'm in a restaurant and automatically inserts the sound of somebody with an annoying nasal voice having an obnoxious conversation? How about some random honking every time my phone detects that it is going more than 30mph?
Bloody hell, people, if there is one thing that modernity needs like a hole in the head, it is more fucking background noise...
Don't worry: They can just update to the "Heathkit: 21st century skills" collection, where you learn the art and science of hiring Chinese subcontractors to assemble the kit, and the CAD skills necessary to design a case with your branding and logo to contain the finished product.
Based on the interweb's rough figures, the lump-sum cash payout of her 'we-love-you-so-much-we-terminated-you-by-phone' dismissal is equal to only 30 years of the average salary of a Yahoo software engineer, not quite a lifetime. Luckily, there is also a complex mess of stock options(conveniently not subject to those little-people payroll taxes!) that should save her from penury...
Unfortunately, that theory captures an incomplete picture of the founder effect:
Requiring the CEO to buy a chunk of the company can provide them with a greater financial stake in the company's success, or it can just provide them with the incentive to axe the R&D department, pump out a few quarters that Wall Street loves, and give themselves a giant bonus in the form of "shareholder value" before moving on...
If anything, having a CEO without major holdings(and without a "Congratulations, you fucked up!" bonus larger than a peon's lifetime earnings, if that isn't to scary to think about) might actually help ensure that they take the long view; because they don't have the same financial incentive to pump, loot, and leave.
Founders(except of built-for-acquisition jobs) tend to have major holdings and personal emotional investments, and it takes both to make them do what they do...
That is going to be fun to cool...
I wouldn't be surprised if there are some specialty niche application guys who are just drooling at the prospect of vastly increased silicon area without more board space or interconnect hassle; but anybody who is cranking the clock, the power handling, or both, is going to find the utility of the layers at the center a bit dubious.
All you really need to extinguish the warm-and-fuzzies is a stiff dose of the fact that an alarming number of the present and upcoming SoC designs at least optionally include pretty aggressive Tivoization features, opaque black-box functions handled by cryptographically verified and non-replaceable firmware blobs, and not infrequently a driver or two that isn't available in source form and makes keeping the kernel current rather tricky...
You can have all the open source you want; but if you can only run it on x86 whiteboxes and select dev boards, you still have a problem.
Would purchasing one with a printed skin texture based on the appearance of the burning, partially skeletonized, Hindenburg shortly before its fatal plunge be tasteless?
Because I am tempted...
I'm highly doubtful of the efficacy of that. There are numerous business areas, even well-known names, who do just fine despite public opprobrium.
A highschool course in IT taking down Thawte or forcing them to clean up their act would be about as likely as a highschool course in personal finance bringing down Goldman-Sachs or leading to the end of byzantine financial chicanery...
Given that one major thread of Marx's critique of capitalism is an application of textbook free-market economics, it is no more surprising that he managed to predict some of the outcomes of an approximately capitalist economy than it is surprising when any other economist does(ie. it is moderately surprising; but not entirely so).
In a free market, the price of a commodity tends towards that commodity's marginal cost of production. In a capitalist system, labor is a commodity. Ergo, in a free market capitalist environment, the price of labor will tend towards its marginal cost of production, which is to say that laborers will earn bare subsistence(or bare subsistence+cost of education/training if some of that is required).
The stuff about 'alienation' is more debateable, and any of Marx's proposed solutions(which, often, weren't all that clearly formulated because he didn't see a way out, except a quasi-millenial "revolution") are pretty shaky; but Marx as economist was at least as sound as average for the discipline, if not rather sounder...
Child #7,000,000,000 gets the prize of officially being recognized as "Not actually a bundle of joy" and, on average, a harsh subsistence existence. Congratulations!
I was thinking something a bit more directional: ie. same principle as the parachute flare; but with some mechanism for the operator to designate a target or area to be illuminated before launch, so that the blindingly-light-of-a-false-noon would only apply to your opponents(and their now thoroughly freaked-out night sight gear)...
Parachute flares, if memory serves, substantially predate armored vehicles, possibly even internal combustion vehicles of any sort; so I was curious about anything developed in the contemporary 'highly sensitive optical instruments on expensive but extremely dangerous armor' period... With modern vehicles in the multiple millions a pop, I imagine that selective-illumination systems in the hundreds or thousands per shot might be seen as viable, and that kind of budget might give you room for things more interesting than magnesium-on-a-string.
Statistics on cow-fart volume, mass, and composition seem hard to come by; but I'm assuming that they'd show up as a brief, rapidly expanding, cloud of internal-body-temperature. Anything that could pick up a human should pick one up as well.
Considerably smaller than the output of anything much more than a toy engine, though...
Out of curiosity, if turning on your own light is so overtly suicidal, has there been any work on some sort of disposable system(a balloon, a compressed-gas launched parachute-projectile, etc) that would quietly move a moderate and unpredictable distance from the user, and then unleash the actinic glare of whatever chemical light source is currently in vogue in the correct direction?
My (totally uninformed about the actual system; but moderately familiar with TECs) speculation would be that the thermal cameras provide the control system the data it needs to calculate what the tank "should" look like, to blend in with ambient. That(with the option of mixing in a pattern, like "cow" or "car full of innocent children") would be converted to a target temperature value for each of the camouflage hexagon units.
Each hexagon module, in turn, would have a thermal sensor on the outer surface and a TEC layer. If the target temperature is higher than the real temp, fire up the TEC with the exterior as the hot-side until real temp = target temp. If the target temp is lower than real temp, reverse polarity and run the TEC with the outside as the cold side until you hit the target.
TECs can be run to generate energy(if a sufficient thermal delta exists between hot and cold side); but that would only be of use to the modules that are reliably cooled from the outside and reliably heated from the inside, and that don't need to do any active concealing(you can put power in, to maintain a thermal delta, or you can pull power out by breaking down a thermal delta; but you can't have both...) Given the shitty efficiencies of even the nicer commercial units, this system is going to be a net draw on the vehicle's electrical system, and will probably end up producing a thermal exhaust that needs to be obfuscated however engine exhausts and AC system hot sides currently are.
Stop putting the "coward" in "anonymous coward".
All you have to do is purchase your system with BAE 4-hour gold support(additional charges may apply). Whenever you are hit by small arms fire, explosions, rocks kicked up by the vehicle in front of you, or any of your camoflauge hexagons is exposed to temp>100 degrees C for too long(which causes TECs to start to break down), just whip out your satellite phone and put in a support call! Your rep should be out there in no time, with a supply of replacement panels.
My understanding is that some IEDs have IR triggers(some bodging based on COTS PIR units, I assume). The low-tech countermeasure is to dangle a running toaster from a pole, a safe distance in front of your vehicle. The low tech counter-countermeasure is to move the IR trigger approximately one 'safe distance' from the explosive...
In theory, I don't see why this stuff wouldn't work against those(though not against the other flavors); but I strongly suspect that the sales pitch involves an enemy equipped with sophisticated armor, tank-killing helicopters, or man-portable rockets more sophisticated than RPG retro classics...
TFA uses the phrase "thermo-electric devices". I'm assuming that means some flavor of Peltier(can't be purely resistive; because some of the pixels need to be cooled and some heated). If that is the case, I would be very interested(and probably not cleared) to know how they handle the heat output of the camouflage, along with the engine and other core systems.
Peltiers are really fun devices; because they are all solid state, respond quickly, and can be driven with a simple DC current; but they aren't what you'd call efficient heat pumps. The fact that they work at all feels like magic; but the heat coming out of the hot side is considerably more than the heat being pumped from the cold side... In IC cooling scenarios, a couple square inches of Peltier can easily consume 100watts. I'd assume that this system, with its much greater surface area, and lower deltaTs on average, wouldn't be that bad; but unless BAE has made some real strides with TECs, nontrivial power is going to be involved(amounts varying depending on what is being emulated and how much the trick differs from the vehicle).
How do you dump all the waste heat from such a system? "looks like a cow" is stealthy. "Looks like a cow, with a thermal exhaust plume that suggests it contains a running AGT1500 and a collection of main battle tank support systems" is less stealthy...
On Windows clients, at any rate, Chrome uses the system certificate store, same as IE, so behavior should be identical(barring any ad-hoc bodging in response to particular issues). I don't know what Chrome on other platforms does.
FF does its own thing, independent of the OS/DE provided certificate store, and throws an untrusted certificate warning.