I have to wonder at their claim that it works well with standard OEM gear. Even most cheap consumer shit monitors the speed of at least the CPU fan and tends to freak out if a fan that is supposed to be there is either absent or performing substantially below expected speed(and, given the relatively high stall current of these fans, burning a trace isn't totally out of the question, if the fan or fan controller isn't smart enough to give up after a short time...) Given that high-density servers and blades will be forced to shut down, or cook, within a short time after fan failure, I'm strongly suspecting that a lot of system management cards and firmwares will flip out at you nonstop if the fans are either removed or acting as mineral oil pumps at a few hundred RPM...
I'd also be interested to see the details of how they handle the rack rails. One major advantage of horizontal mounting is that, assuming the bearings aren't completely shot, a single person can easily and safely pull out even some 8U monstrosity. Vertically, you are working against gravity, and your spindly geek arms wouldn't need to be all that spindly to encounter some hilariously expensive dropping-the-coolant-lubed-hardware-with-an-expensive-crunch accidents. Unless the system is specced exclusively for dinky little blades and half-depth 1Us(in which case those massively built coolant ponds with aisles on all sides aren't actually all that dense...), they would have to have something in place to make working with the bigger stuff doable.
This makes me wonder about what MS has in mind for WinCE, actually. Have they decided that CE just isn't worth it, and it is going to be EOLed and/or consigned to only the lowest-end embedded/realtime stuff, with an NT derivative replacing it on all higher specced embedded devices? Is there a body of software(their own, or 3rd party), that they have commitments to do an architecture port, if win32-on-ARM is available, that they couldn't get commitments to port to the kinda-sorta-a-bit-win32 world of WinCE?
It just seems strange, since NT on ARM is going to break binary(and possibly to some degree source) compatibility with absolutely everything that isn't 100% pure CLR, that they would bother to port NT, unless they had concluded that CE was beyond hope, or that it would be cheaper to maintain a single kernel.
Given NT's multiplatform history, it doesn't surprise me that they could do it, I just don't really understand why they would... With WP7 they already have a CE-based embedded platform for running CLR applications, and kernel-level stuff like drivers isn't going to survive a architecture transition, so I'm genuinely curious what the plan is.
Flip's chances certainly weren't helped by the fact that, on the one hand, point and shoots with substantially more competent optics have been creeping down in price and creeping up in video capability, and on the other, smartphones(while substantially more expensive) are increasingly seen as a default, and so offer almost as good video recording for "free".
However, it really doesn't help that Cisco did surprisingly little with the company after they acquired it, and some of what they did do was questionable. The 'Slide' model was rather pitiful, their experiments in replacing the simple tried and true physical buttons with (lousy) touchscreens were failures, and they stuck with a price tag that was always hovering dangerously close to more capable devices. Other than a few incremental spec bumps there was almost no development of the product line for two years.
Yes and no. Yes in that willingness to pay is the proximate criterion on which prices are set. No in that one of the major determinants of somebody's willingness to pay is what other providers of similar or identical commodities are charging. In reasonably competitive markets, price competition between approximately equivalent providers of a given good or service means that, in the end, willingness to pay is approximately equal to the lowest price, which is based on the cost structure of the outfit providing that price.
In hilariously non-competitive markets, of course, willingness to pay and cost are more or less completely decoupled. The same is true for 'ahead of their time' products(where everbody's cost is much higher than anybody's willingness to pay, so the product stays in the lab). In a competitive market for a mature product, though, willingness to pay and cost are fairly closely related.
I can appreciate that, for the horticulturally interested, seeing how high a Capsaicin concentration you can get out of something that will actually grow might be an interesting hobby/challenge.
Beyond that, though, it seems sort of pointless. We know what the active ingredient is, we have techniques both for extraction and purification and for synthesis. It would be a modest effort in benchtop chemistry to produce Capsiacin at 90+% purity from even the comparatively mundane peppers, and 100% pure provides the clear upper limit on heat possible from that compound. Unless the interest is purely in the pepper growing, it seems like the Extreme Heat Sauce crowd could simply obtain the pure product, endure the pain, and go home happy.
Now, if the college just wants to move warm bodies through the door, they won't put one in place; don't want to discourage the customers until they've finished paying, after all. At the better class of institutions, though, it shouldn't be too hard to arm-twist the department's Serious Theory dude to run a brutal class that will fairly swiftly divide the 'CS' student population into 'actual CS students', 'purely vocational software engineers', and 'just in it for the money, have you considered a business major?'.
The fact that everything from 'intro to MS Office 101' to 'Nightmares at the edges of computability 895, instructor permission required' gets placed under the heading of 'CS' seems to be the real problem. It'd be like having computational fluid dynamicists, plumbers, and civil engineers all being produced by the same department...
Arguably, we should be much more worried about the financial predation of first-world organized criminals: The banks and the credit rating agencies and similar such institutions are the ones who make it trivially easy to act in other people's names, in order to move their product more easily and cheaply, and then attempt to sidestep the losses from fraud by hounding the people whose names were used.
The only predation by third-world organized criminals that occurs directly against the end user consists of 419 scams. The rest of it consists of various sorts of bank fraud that the banks aren't sufficiently motivated to take measures against; but are willing to put those whose names are used through the wringer.
It's very clever, really: "Identity theft" makes it your problem. Admitting that it is "bank fraud" would make it their problem.
I'd be inclined to look more favorably on this case. Airport Express units are still in active production, so supplies are limited only by your budget for them, and the sacrifice of a single one has now allowed a much greater pool of devices to incorporate a useful feature(since every other aspect of airplay has interoperable replacements written, but the lockdown key was missing).
An analogous case would be somebody decapping the chips in a C64 in order to produce a fully accurate emulator: the sacrifice of a single unit would provide the data needed to produce an unlimited number of replicas.
Unless Feist v. Rural has been dramatically overturned recently, I'm fairly sure that neither copyright nor patent are remotely applicable to an RSA key.
Arguably, extracting the private key that serves as the artificial restriction preventing 3rd party software and/or devices from serving as airplay sinks is a much more value-creating activity than just reselling the widget.
The 'space elevator' is in a funny position. On the one hand, it is much further away, technologically, than just about any other space related project within our solar system. On the other hand, the development of high performance structural materials is something with immediate commercial applications in all sorts of fields, so there is plenty of incentive to make incremental advances in that direction, whether or not that project will ever be feasible. Most other projects are easier; but have fewer short-term payoffs distributed along the path to completion.
Don't forget Adobe Reader. I've lost count of the number of Reader security advisories that apply to basically every OS they release binaries for. It isn't often you see news of an exploit vector for Solaris; but Adobe manages it.
The news on computer security is usually relentlessly bad. It is nice to see an instance where the economic realities of non-targeted attacks make the bad guys slightly more vulnerable. Even if our antivirus overlords are pitifully incapable of keeping us from getting 0wn3d, which seems to be the case, they are in a fairly good position to monitor the 'underground' marketplace and reduce the value of compromised PCs. That won't save the strategically valuable targets; but anything that reduces the rental value of Joe Broadband's horribly compromised porn box is good for Joe, and for the internet generally.
This would appear to be operating on the optimistic assumption that hitting "delete" has any effect other than making the message invisible. If you are lucky, cheapness and/or laziness on the part of the operator will mean that they purge and/or just lose everything they can get away with as fast as possible. If, on the other hand, they are doing some sort of data mining for commercial purposes, or complying with some sort of retention request, game over, man.
Big military vessels don't need indirect fire if they have air cover... which they generally do at all times.
Assuming that the local aircraft carrier isn't pursuing an exciting second career as a submarine at the time, which is presumably why the US is investing in stuff like this(along with their continued emphasis on in-air refueling abilities). The US is in the somewhat tricky position that most hypothetical 'force projection' scenarios that don't simply involve asymmetric beating on dusty hellholes full of irregulars(dubiously winnable; but not a blue-water threat) tend to involve having to move very, very expensive carriers within the potential range of hostile land based missiles.
My impression is that the (eventual) use case, aside from giving our valued contractors something to bill for, is intercepting missiles and possibly nearby aircraft.
A fair number of navy vessels, especially the pricey, strategically important ones, do have a nuclear reactor to power it. They are also subject to some concern about the ability of today's minigun-based CIWS defenses to deal with some contemporary and upcoming anti-ship missiles. An anti-boat test is a serious lowball, compared to the eventual task; but I assume somebody had a 'milestone' that needed to be ticked.
For other ships, and coastal targets, the navy has also been showing considerable interest in railguns...
I'm not sure that Jobs could possibly have been induced to take the offer: He already has more money than he could conceivably spend on any hobby that isn't "tell financial minion to write check for value of liquid holdings to somebody"(and, unlike many wealthy CEOs, he doesn't seem to have any wildly expensive hobbies), so he is unlikely to be buyable.
His work with Apple(which obviously gooses the value of his stock holdings; but for which he doesn't get paid nearly what he easy could demand) seems to be entirely about pursuing his perfectionism wherever it leads him, even if that means killing profitable products(hello iPod Mini...), stomping on backwards compatibility in ways that upset important partners(Yo Adobe, 64 bit carbon is dead, we didn't bother to tell you until the last second; because Cocoa is just better.) and trading marketshare for margin whenever necessary(nearly all the Macs, the continued lack of a 1 socket mini-tower type config).
Google, on the other hand, really only does relentless perfectionism on the back end(datacenter efficiency and search algorithms). Most of their user-facing stuff is not bad; but is proudly beta, low margin, and basically about being good enough to serve its strategically vital cash cows.
Unless Jobs suddenly developed an intense hatred of publicity, in which case he might well be a good recruit for some position in Google's back-end operations, a gig with Google would run strongly against his tastes, and he already has enough money, and not enough interest in money, that Google couldn't easily buy him.
I find that the easiest technique is to book them rooms at the Hilbert Hotel and then leave as quickly as possible. It's one of the few places that will take them all...
The only upside to that would be the prior-art retort to it...
Just remind them that whoever first invented a set of axioms sufficiently comprehensive to derive the series of positive integers from indirectly(and in a deeply useless sense) produced all works capable of digital representation. Heck, unlike dev/random, each one of all possible finite files will even be in the same place each time you look. Luxury! Unfortunately, this also means that possession of almost any text dealing with number or set theory indirectly implies possession of kiddie porn. Oops.
(Next up: a book where each page's contents are simply the page number in base2, interpreted as unicode, Easiest way to encapsulate all written human knowledge ever...)
All the files I've ever created, along with all the files anyone else has created, along with all the files of finite length that nobody has ever created, are waiting right there for you in/dev/random.
I haven't seen the principle applied to faking an HDD before; but the same phenomenon crops up fairly frequently with USB flash drives and flash memory cards sourced from suspiciously cheap ebay sellers and similar places.
The cruder examples are simply a low-capacity drive, with a high capacity label, and a specially doctored partition table and fat32 filesystem written to them. Simply reformatting them will reveal their true size and make them safely usable(to the degree that you would trust the quality of such a device...).
The more sophisticated ones have doctored firmware in the chip that handles abstracting the raw flash into a USB mass storage device, and the OS will detect their false size. You can only determine the true size empirically: exactly what behavior the fake blocks will exhibit varies(all zeros, all ones, garbage); but the real blocks will behave normally. If you are a gambling sort, you can put a partition of exactly that size on the drive and hope for the best; but that isn't really advisable...
Every abstraction layer is a potential lie, I suppose.
Slashdot Mirror
I have to wonder at their claim that it works well with standard OEM gear. Even most cheap consumer shit monitors the speed of at least the CPU fan and tends to freak out if a fan that is supposed to be there is either absent or performing substantially below expected speed(and, given the relatively high stall current of these fans, burning a trace isn't totally out of the question, if the fan or fan controller isn't smart enough to give up after a short time...) Given that high-density servers and blades will be forced to shut down, or cook, within a short time after fan failure, I'm strongly suspecting that a lot of system management cards and firmwares will flip out at you nonstop if the fans are either removed or acting as mineral oil pumps at a few hundred RPM...
I'd also be interested to see the details of how they handle the rack rails. One major advantage of horizontal mounting is that, assuming the bearings aren't completely shot, a single person can easily and safely pull out even some 8U monstrosity. Vertically, you are working against gravity, and your spindly geek arms wouldn't need to be all that spindly to encounter some hilariously expensive dropping-the-coolant-lubed-hardware-with-an-expensive-crunch accidents. Unless the system is specced exclusively for dinky little blades and half-depth 1Us(in which case those massively built coolant ponds with aisles on all sides aren't actually all that dense...), they would have to have something in place to make working with the bigger stuff doable.
This makes me wonder about what MS has in mind for WinCE, actually. Have they decided that CE just isn't worth it, and it is going to be EOLed and/or consigned to only the lowest-end embedded/realtime stuff, with an NT derivative replacing it on all higher specced embedded devices? Is there a body of software(their own, or 3rd party), that they have commitments to do an architecture port, if win32-on-ARM is available, that they couldn't get commitments to port to the kinda-sorta-a-bit-win32 world of WinCE?
It just seems strange, since NT on ARM is going to break binary(and possibly to some degree source) compatibility with absolutely everything that isn't 100% pure CLR, that they would bother to port NT, unless they had concluded that CE was beyond hope, or that it would be cheaper to maintain a single kernel.
Given NT's multiplatform history, it doesn't surprise me that they could do it, I just don't really understand why they would... With WP7 they already have a CE-based embedded platform for running CLR applications, and kernel-level stuff like drivers isn't going to survive a architecture transition, so I'm genuinely curious what the plan is.
Flip's chances certainly weren't helped by the fact that, on the one hand, point and shoots with substantially more competent optics have been creeping down in price and creeping up in video capability, and on the other, smartphones(while substantially more expensive) are increasingly seen as a default, and so offer almost as good video recording for "free".
However, it really doesn't help that Cisco did surprisingly little with the company after they acquired it, and some of what they did do was questionable. The 'Slide' model was rather pitiful, their experiments in replacing the simple tried and true physical buttons with (lousy) touchscreens were failures, and they stuck with a price tag that was always hovering dangerously close to more capable devices. Other than a few incremental spec bumps there was almost no development of the product line for two years.
Yes and no. Yes in that willingness to pay is the proximate criterion on which prices are set. No in that one of the major determinants of somebody's willingness to pay is what other providers of similar or identical commodities are charging. In reasonably competitive markets, price competition between approximately equivalent providers of a given good or service means that, in the end, willingness to pay is approximately equal to the lowest price, which is based on the cost structure of the outfit providing that price.
In hilariously non-competitive markets, of course, willingness to pay and cost are more or less completely decoupled. The same is true for 'ahead of their time' products(where everbody's cost is much higher than anybody's willingness to pay, so the product stays in the lab). In a competitive market for a mature product, though, willingness to pay and cost are fairly closely related.
I can appreciate that, for the horticulturally interested, seeing how high a Capsaicin concentration you can get out of something that will actually grow might be an interesting hobby/challenge.
Beyond that, though, it seems sort of pointless. We know what the active ingredient is, we have techniques both for extraction and purification and for synthesis. It would be a modest effort in benchtop chemistry to produce Capsiacin at 90+% purity from even the comparatively mundane peppers, and 100% pure provides the clear upper limit on heat possible from that compound. Unless the interest is purely in the pepper growing, it seems like the Extreme Heat Sauce crowd could simply obtain the pure product, endure the pain, and go home happy.
Two words: "Weeder Course".
Now, if the college just wants to move warm bodies through the door, they won't put one in place; don't want to discourage the customers until they've finished paying, after all. At the better class of institutions, though, it shouldn't be too hard to arm-twist the department's Serious Theory dude to run a brutal class that will fairly swiftly divide the 'CS' student population into 'actual CS students', 'purely vocational software engineers', and 'just in it for the money, have you considered a business major?'.
The fact that everything from 'intro to MS Office 101' to 'Nightmares at the edges of computability 895, instructor permission required' gets placed under the heading of 'CS' seems to be the real problem. It'd be like having computational fluid dynamicists, plumbers, and civil engineers all being produced by the same department...
Arguably, we should be much more worried about the financial predation of first-world organized criminals: The banks and the credit rating agencies and similar such institutions are the ones who make it trivially easy to act in other people's names, in order to move their product more easily and cheaply, and then attempt to sidestep the losses from fraud by hounding the people whose names were used.
The only predation by third-world organized criminals that occurs directly against the end user consists of 419 scams. The rest of it consists of various sorts of bank fraud that the banks aren't sufficiently motivated to take measures against; but are willing to put those whose names are used through the wringer.
It's very clever, really: "Identity theft" makes it your problem. Admitting that it is "bank fraud" would make it their problem.
I'd be inclined to look more favorably on this case. Airport Express units are still in active production, so supplies are limited only by your budget for them, and the sacrifice of a single one has now allowed a much greater pool of devices to incorporate a useful feature(since every other aspect of airplay has interoperable replacements written, but the lockdown key was missing).
An analogous case would be somebody decapping the chips in a C64 in order to produce a fully accurate emulator: the sacrifice of a single unit would provide the data needed to produce an unlimited number of replicas.
Unless Feist v. Rural has been dramatically overturned recently, I'm fairly sure that neither copyright nor patent are remotely applicable to an RSA key.
Arguably, extracting the private key that serves as the artificial restriction preventing 3rd party software and/or devices from serving as airplay sinks is a much more value-creating activity than just reselling the widget.
The 'space elevator' is in a funny position. On the one hand, it is much further away, technologically, than just about any other space related project within our solar system. On the other hand, the development of high performance structural materials is something with immediate commercial applications in all sorts of fields, so there is plenty of incentive to make incremental advances in that direction, whether or not that project will ever be feasible. Most other projects are easier; but have fewer short-term payoffs distributed along the path to completion.
Don't forget Adobe Reader. I've lost count of the number of Reader security advisories that apply to basically every OS they release binaries for. It isn't often you see news of an exploit vector for Solaris; but Adobe manages it.
The news on computer security is usually relentlessly bad. It is nice to see an instance where the economic realities of non-targeted attacks make the bad guys slightly more vulnerable. Even if our antivirus overlords are pitifully incapable of keeping us from getting 0wn3d, which seems to be the case, they are in a fairly good position to monitor the 'underground' marketplace and reduce the value of compromised PCs. That won't save the strategically valuable targets; but anything that reduces the rental value of Joe Broadband's horribly compromised porn box is good for Joe, and for the internet generally.
Yo dog, I herd you like zero-days, so I put a zero day in your box so somebody else can compute while you compute...
This would appear to be operating on the optimistic assumption that hitting "delete" has any effect other than making the message invisible. If you are lucky, cheapness and/or laziness on the part of the operator will mean that they purge and/or just lose everything they can get away with as fast as possible. If, on the other hand, they are doing some sort of data mining for commercial purposes, or complying with some sort of retention request, game over, man.
Be sure to avoid living in Britain, in that case.
Big military vessels don't need indirect fire if they have air cover... which they generally do at all times.
Assuming that the local aircraft carrier isn't pursuing an exciting second career as a submarine at the time, which is presumably why the US is investing in stuff like this(along with their continued emphasis on in-air refueling abilities). The US is in the somewhat tricky position that most hypothetical 'force projection' scenarios that don't simply involve asymmetric beating on dusty hellholes full of irregulars(dubiously winnable; but not a blue-water threat) tend to involve having to move very, very expensive carriers within the potential range of hostile land based missiles.
My impression is that the (eventual) use case, aside from giving our valued contractors something to bill for, is intercepting missiles and possibly nearby aircraft.
A fair number of navy vessels, especially the pricey, strategically important ones, do have a nuclear reactor to power it. They are also subject to some concern about the ability of today's minigun-based CIWS defenses to deal with some contemporary and upcoming anti-ship missiles. An anti-boat test is a serious lowball, compared to the eventual task; but I assume somebody had a 'milestone' that needed to be ticked.
For other ships, and coastal targets, the navy has also been showing considerable interest in railguns...
I'm not sure that Jobs could possibly have been induced to take the offer: He already has more money than he could conceivably spend on any hobby that isn't "tell financial minion to write check for value of liquid holdings to somebody"(and, unlike many wealthy CEOs, he doesn't seem to have any wildly expensive hobbies), so he is unlikely to be buyable.
His work with Apple(which obviously gooses the value of his stock holdings; but for which he doesn't get paid nearly what he easy could demand) seems to be entirely about pursuing his perfectionism wherever it leads him, even if that means killing profitable products(hello iPod Mini...), stomping on backwards compatibility in ways that upset important partners(Yo Adobe, 64 bit carbon is dead, we didn't bother to tell you until the last second; because Cocoa is just better.) and trading marketshare for margin whenever necessary(nearly all the Macs, the continued lack of a 1 socket mini-tower type config).
Google, on the other hand, really only does relentless perfectionism on the back end(datacenter efficiency and search algorithms). Most of their user-facing stuff is not bad; but is proudly beta, low margin, and basically about being good enough to serve its strategically vital cash cows.
Unless Jobs suddenly developed an intense hatred of publicity, in which case he might well be a good recruit for some position in Google's back-end operations, a gig with Google would run strongly against his tastes, and he already has enough money, and not enough interest in money, that Google couldn't easily buy him.
@binladen: Oh no you di-ent! #dirtybomb #WMD
I find that the easiest technique is to book them rooms at the Hilbert Hotel and then leave as quickly as possible. It's one of the few places that will take them all...
The only upside to that would be the prior-art retort to it...
Just remind them that whoever first invented a set of axioms sufficiently comprehensive to derive the series of positive integers from indirectly(and in a deeply useless sense) produced all works capable of digital representation. Heck, unlike dev/random, each one of all possible finite files will even be in the same place each time you look. Luxury! Unfortunately, this also means that possession of almost any text dealing with number or set theory indirectly implies possession of kiddie porn. Oops.
(Next up: a book where each page's contents are simply the page number in base2, interpreted as unicode, Easiest way to encapsulate all written human knowledge ever...)
I find that saving data is a waste of time.
/dev/random.
All the files I've ever created, along with all the files anyone else has created, along with all the files of finite length that nobody has ever created, are waiting right there for you in
Latency is a bit unpredictable, though.
The key innovation in this product is not keeping it real fake; but selectively discarding it real fake...
I haven't seen the principle applied to faking an HDD before; but the same phenomenon crops up fairly frequently with USB flash drives and flash memory cards sourced from suspiciously cheap ebay sellers and similar places.
The cruder examples are simply a low-capacity drive, with a high capacity label, and a specially doctored partition table and fat32 filesystem written to them. Simply reformatting them will reveal their true size and make them safely usable(to the degree that you would trust the quality of such a device...).
The more sophisticated ones have doctored firmware in the chip that handles abstracting the raw flash into a USB mass storage device, and the OS will detect their false size. You can only determine the true size empirically: exactly what behavior the fake blocks will exhibit varies(all zeros, all ones, garbage); but the real blocks will behave normally. If you are a gambling sort, you can put a partition of exactly that size on the drive and hope for the best; but that isn't really advisable...
Every abstraction layer is a potential lie, I suppose.