I don't think that TFS's answer is necessarily the correct one. I'd really prefer to hold Ma Bell's feet to the fire concerning the fact that bandwidth(even in 'optimal' build-out areas, spare me the excuses about the boonies) has enjoyed a deeply underwhelming track record in terms of improvements in cost and quantity compared to most other aspects of contemporary computing.
How are we going to get another bubble going with naysayers like you around? Don't you understand that 'social' is the new magic word, now that '.com' is for lame old people?
It's hard to imagine that an IPO would ever be a good place for Joe Civilian to stick his toes in(with the exception of the just-regular-code-monkeys who got lucky enough to score some stock options by working for the right startup and are cashing out as fast as tactfully possible)...
You are unlikely to beat the investment bank(s) and/or Venture guys who are there to handle the IPO, nor do you have a good chance of having better information than the company insiders who are either cashing out or picking up shares, and even if the company is on track to be a stable, long-term 'hold', there is likely to be some fairly volatile oscillation shortly after the IPO.
It appears to be an article of faith among the professional chatterers of the market-news media that THE NUMBERS MUST GO UP!!!!. If interrogated directly, of course, they will concede that 'the market' sometimes requires that the numbers go down, as folly and weakness are eliminated; but day-to-day this saddens them.
Just look at the body of media drivel generated by the recent deflation of the American housing bubble: having a place to live became more affordable than it had been in decades and every last talking head and politician available began screaming about the 'housing crisis'...
There probably are genuinely analytical analysts(who know enough to keep their mouths shut and make real money); but the ones bloviating in public appear to be little more than cheerleaders at a sort of stock market pep rally.
It's extra hardware(and not inexpensive, from what I've read); but I would think that the classic 'terminal window on ttyS0' would be an nearideal match for an 80-column refreshable braille display...
I've no doubt that Windows has superior 'kludging some degree of blind usability on top of a GUI' software offerings, because there is some serious cash in the 'corporations that don't like ADA suits' market; but I would(perhaps naively) expect that the unix-style environment(not exclusive to Linux, of course, BSD would do just as well) that had everybody on 80-column serial terminals back before graphics were cheap would be much easier to adapt more directly... A contemporary Window-manager-with-all-bells-and-whistles Linux desktop, not so much, though.
That sounds a lot like a local micro/pico/femto cell base station, which I've read that carriers are commonly installing in their retail locations to avoid embarassing connectivity/throughput issues when a number of customers are hammering away at the demo units.
It would allow HQ to see that IMEI XYZ appears to be accessing NSFW.com within 30 meters of the store at times not-so-coincidentally similar to those times when human-resource-peon Smith is scheduled for work; but I'd be pretty surprised if that were the primary purpose. Installing a zillion weedy little cells is something you do because they only work if they are on site. Surveillance is both cheaper and much more secure against tampering if you do it closer to the center of the network. There are a lot of people who could get a look at the suspicious looking box in the back room of the local Cell Shack. There are a great deal fewer who get to go inside the windowless mystery bunkers where the bigger gear lives...
The 'due to legal issues' does sound an awful lot like the 'friend' wants a look a traffic going across his network, but from devices that are owned by somebody else.
Given the amount of case law that was basically formulated to address the "Employee is dicking around on facebook, doesn't like the fact that Network caught them at the IDS/Firewall/etc." case, it might well be legal to tap somebody under those conditions; but it's hard to make the case that it is ethical to tap somebody on the wire when they are enough of a legal agent that you couldn't tap them at the endpoint... I get this funny feeling that somebody's wife/girlfriend/roommate/other is going to be lawyering up in the not so distant future...
There are a few ghastly legacy applications that oozed out of the 'we'll just build our own damn widget set, because!' school, which obey absolutely no system-wide settings whatsoever, not theme, not font size, screenreaders can forget about it, and so forth; but those are thankfully rare... Even then, you can always just give the contrast, gamma, or color curves a good hard shove at the driver or monitor level.
More importantly, essentially all South Koreans under the age of 40 have the ability to stop a zerg rush encoded into their brainstems through sheer practice...
I don't know how much is purely down to inertia and inefficiency(these certainly cannot be ruled out); but I get the impression that the US postal service has a certain cultural attachment to a slightly retro ideal of universal service going back to their original constitutional mandate. This is of pretty questionable use in their contemporary capacity as high-volume junk mail distributers with a side of certified legal mailings; but my interactions with postal personnel(especially in smaller markets) has always given me the impression that they take a certain pride in the fact that anyone can scrawl a vague reference to somewhere in Podunk on an envelope, slap on a stamp, and have it actually arrive at the correct slice of nowhere, courtesy of the postman who knows that area.
Fedex, on the other hand, you expect the barcodes and the little scanner/PDA widget.
As noted, it isn't obvious that this cultural orientation is a good fit for the position that the service finds itself in; but it has always struck me as an interesting phenomenon...
If nothing else, it's a cost-saving measure. Politically, you aren't allowed to use your nukes, or even test them much anymore; but that doesn't make them any cheaper to maintain, stockpile, keep ready for use, etc. A shiny, bilateral, humanitarian, oh-so-noble, strategic arms reduction treaty serves as a pleasantly face-saving way for both the US and Russia to say 'fuck it, this shit isn't worth the money'.
Nobody wants to be the first to blink; but being a 'responsible' nuclear power is actually pretty lame. You aren't allowed to do crazy tinpot dictator stuff, and your large conventional forces deter smaller competitors even without nukes; but you still have to pay upkeep, fret about loose cannons sneaking off with a warhead when you aren't looking, and so forth.
I'd wager that none of the nuclear powers would want to get rid of all their toys; but that the US and Russia are delighted to have been able to free up some of the cash spent on the ludicrious overkill buildup levels of the cold war and divert that to other uses...
TFA's thesis is that there simply isn't a dividing line between 'tactical' and 'strategic' that makes your sentence meaningful.
Given the mixture of forward bases and in-air refueling that the nuclear powers certainly aren't going to give up(even if their intentions are purely conventional, being able to B-52 bits of the middle east is just too convenient) delivering a 'tactical' warhead(or 10) right down a 'strategic' target's chimney is a matter of little more than swapping some hardpoints, and the upper edge of the dial-a-yield ranges for 'tactical' devices are well into the realm of 'would ruin a population center's day' territory.
There are, certainly, some unambiguously 'strategic' weapons, of the 'bloody huge thermonuclear warhead on an ICBM' school and there are (a dwindling number of) oddball low-yield artillery shells, demolition charges, and other oddities from the heyday of nuclear optimism(for sheer weirdness, I'm fond of the 'Davy Crockett' system. Essentially a 'technical' as much beloved by ragged 3rd world armies; but with nuclear warheads...); but much of the active hardware falls into the awkward middle ground where, in order to be powerful enough to be worth the trouble of being nuclear, it could certainly bestrategic if asked, and purely because of what people expect from contemporary delivery sytems, even for boring chemical explosives, 'strategic' delivery capabilities are widespread.
As best I can tell, MS purchased the kinect technology with the relatively narrow objective of coming up with something that could beat the Wii's motion-gimmick factor without requiring a new generation of console. They didn't even do the traditional vapor-announce-to-discourage-competitors of a Windows SDK until well after they had finished losing their shit at the people who went ahead and made their own. I don't know if they got a good price from the guys they licensed it from, based on some agreement to limit the scope to cheap gaming stuff, or whether they were just feeling amazingly shortsighted...
It is more curious that the company that MS licensed the tech from hadn't been pushing it left, right, and center before the Kinect made headlines. Even if their lower-volume hardware was a few grand a pop, they could still have found plenty of buyers in markets where that is entirely reasonable.
Certainly, if you are willing to use violence you can do whatever you can get away with. However, while Sharia pretty much sucks by humanist standards, it is subject to its own internal mores and isn't arbitrarily flexible. In this case, the limiting factor(beyond the basic 'Does the idea that Mohammed gave a damn what an internet search company in the early 21st century called that body of water even pass the laugh test?' problem) would likely be that this particular question of naming is an internal spat.
The leading contenders are "Persian Gulf", favored by ethnic nationalists in Persian-majority Iran, and "Arabian Gulf", favored by assorted ethnic nationalists in Arab-majority areas of the middle east. That(along with the not-so-minor fact that you also have Shia/Sunni split) would probably get a Fatwa endorsed by some Iranian theologian laughed right out of court among a substantial majority of the world's Muslims, never mind everybody else. This wouldn't be one of those unifying issues, like being against Rushdie or thinking that Israel pretty much sucks.
Some sort of civil-law 'actions calculated to destabilize the state' nonsense would probably be a lot easier to kangaroo-court through; but then Google would need to have some sort of operations within Iranian jurisdiction.
What sort of legal action, exactly, can you take against somebody for making a map that hurts your feelings?
I've never heard of any legal success in the US on those grounds(though PR concerns certainly motivate companies to tread lightly) and under the present US sanctions, Google is unlikely to be officially operating within Iran at all(given the SSL MiTM incident a while back, there are obviously Iranians using Google services; but the sanctions make it rather unlikely that Google would choose to site any official branch offices there), so it hardly matters whether Iran has a rule against hurting their feelings in the process of mapping.
Using masses of GPUs instead of the slow cpu to do decryption work?
GPUs and FPGAs have, in a number of cases, moved attacks on known-vulnerable systems from 'theoretical; but of great concern' to 'desktop, you don't even need 3-phase 220' faster than other technologies would have; but the cryptographic systems that are actually trusted tend to be of the 'barring fundamental breakthroughs in either mathematics or physics, converting all the mass in the solar system into crypto-chips it would merely shave a few zeros off the expected time...' variety.
Considering the (mostly) invincible state of good encryption, this seems unnecessary. Sure, it is a fun idea, but not a practical one.
I suspect that(aside from simply being relatively cheap to implement, and having some expected sales based on the 'cool' factor alone) the real purpose of any system purporting to substitute for, or complement, disk encryption is to deal with weaknesses unrelated to the cryptographic system itself.
As best we know, contemporary crypto systems with keys of reasonable length are not breakable in any useful sense. However, since humans who can store keys of reasonable length are vanishingly rare, most such systems must weaken themselves by storing a good chunk of the key somewhere where it can be recovered(on the device, with a password/passphrase, embedded in a smartcard/fob that is merely expensive and inconvenient to crack and extract, etc.)
If you just hand somebody an AES blob and tell them to come up with the correct 256 bits for themselves, they are SOL. If somebody's HDD is encrypted such that it can be unlocked with a password it is quite likely that dictionary-aided brute force will be enough. If the key is on a smartcard IC it might not be cheap to pull the key; but they don't produce those things in volume for a few bucks a chip, at most, by using all available security features...
After all, they are excellent tools for padeophiles and terrorists. Amirite?
I suspect that it depends how much other evidence they have. Nuking the only copy of something vital might well save you from conviction on more serious charges, at the cost of some sort of obstruction of justice/destruction of evidence charge. If they already have corroborating evidence from other sources, though, the DA will probably just smirk and add another charge, complete with trivially available evidence that makes you look guilty as hell.
"Yes, your honor. We detected child pornography downloads from the IP that the defendant's ISP had assigned to his residence at the time. When we arrived to execute a search warrant, the defendant dove toward his computer and destroyed its hard drive...."
I suspect that there are (at least) two distinct schools of utter fail:
The professionals, with a legacy in CCTV-as-in-actual-closed-circuit-running-on-private-coax, probably have an attitude much as you describe. The classic CCTV systems were dumb as bricks(not that their designers necessarily were, making largely analog, reasonably high bandwidth systems actually work in practice isn't trivial); but that lack of sophistication served as a strong defense against anybody without a physical tap shoved right into the coax. You just don't develop a very strong culture of caring about remote exploits if your engineering history is almost entirely concerned with systems that are incapable of remote anything, whether you like it or not.
Then you have the upstarts(either new companies, or rebadged ODM crap sold by existing ones), who design CCTV systems on the premise that a CCTV camera is basically just an embedded linux board with a camera interface, and a record/playback system is basically just an x86 with some sort of h264 hardware and a lousy frontend. These assumptions are not false, and advances in silicon sensors and cheap embedded computers definitely mean that the price is right; but the standards of security excellence in low-cost embedded gear are absolutely fucking dire... These guys should know better, since their designs are 100% post-ubiquitous-networking in concept; but they just don't get paid enough, or enjoy long enough development cycles, to give a damn.
Nothing recent enough to be without an ethernet port. If they've made the dongles worth, though, then they have the 'plan B' mentioned, and it shouldn't be a big deal.
They used to be able to beat Intel in the Athlon days. Now they are hopelessly far behind and dumping huge hot graphics cores into their chips putting them further and further behind.
Focus on cheap compute with unlocking cores AMD. Not stupid graphics cores which do nothing for the CPU. A 16 core phenom ii at $100 will sell much better than this insane graphics + cpu crap.
That is pretty much the exact opposite of a good plan for AMD(as much as I would like cheap compute...) Since Intel has a process advantage, and presently has a superior x86 compute core architecture, they can almost certainly beat AMD on production cost for chips of a given level of punch. Trying to compete on price with somebody kicking out chips a process node ahead of you just isn't a good plan. Unless they really fuck it up, or their yields tank horribly or similar, they'll be able to beat you on production cost every time. Intel has little to gain by cutting its own margins in order to chase AMD down a hole(since lower margins are bad, and killing AMD would mean becoming antitrust scrutiny case #1 for the indefinite future...); but there isn't any architectural barrier to their doing so.
Since Intel has comparatively worthless GPU designs, tacking GPUs onto CPU dice is a way for AMD to offer something that Intel cannot(and at a price lower than a discrete CPU + discrete GPU without totally cutting their own throat), and also happens to go well with today's enthusiasm for laptops and all-in-ones. They have a second niche, much more directly focused on price, in compute-light, memory-heavy server applications(since you can populate your sockets with AMD CPUs for less and the number of DIMMs you get is roughly proportional to the number of sockets you have active); but competing on price isn't good for your margins.
With an inferior process and a weaker x86 design, gunning directly for the compute performance crown would just be asking for a whupping from Intel.
Even the fancy, we-make-cardiac-stents-out-of-this-stuff, grade is considered to be only moderately biocompatible. Apparently safe enough to beat the alternative; but they aren't pouring work into fancy new biocompatible polymers because steel is good enough...
I'm not so surprised that some rather alarmingly powerful beam sources would be operated quietly by people with atypical sensor needs. I am a bit surprised that 3.5 lbs of highly enriched Uranium would be available to serve as a beam source.
Not telling the neighbors about a scary-sounding piece of industrial/scientific apparatus is one thing, having enough nuclear material to interest a proliferation wonk in your basement, on the other hand, seems like it would raise eyebrows...
What, exactly, do you think companies use computers for, anyhow?
I've never run into an IT setup of any size that didn't consider "install approved image" as step 1 of deploying a new system, or redeploying one previously used elsewhere. The tools for doing this over ethernet (PXE/Netboot, dump image to disk, reboot) are commonly available and mature. Over wifi? Not so much. Over USB ethernet dongles? Surely you jest.
Doesn't matter much for home users, or very small scale outfits where the slower 'boot from external volume, fire up disk utility, copy disk image to internal volume' method is acceptable; but Apple had better have a plan B for everybody who wants to image machines in bulk if they fancy institutional sales...
Slashdot Mirror
I don't think that TFS's answer is necessarily the correct one. I'd really prefer to hold Ma Bell's feet to the fire concerning the fact that bandwidth(even in 'optimal' build-out areas, spare me the excuses about the boonies) has enjoyed a deeply underwhelming track record in terms of improvements in cost and quantity compared to most other aspects of contemporary computing.
How are we going to get another bubble going with naysayers like you around? Don't you understand that 'social' is the new magic word, now that '.com' is for lame old people?
It's hard to imagine that an IPO would ever be a good place for Joe Civilian to stick his toes in(with the exception of the just-regular-code-monkeys who got lucky enough to score some stock options by working for the right startup and are cashing out as fast as tactfully possible)...
You are unlikely to beat the investment bank(s) and/or Venture guys who are there to handle the IPO, nor do you have a good chance of having better information than the company insiders who are either cashing out or picking up shares, and even if the company is on track to be a stable, long-term 'hold', there is likely to be some fairly volatile oscillation shortly after the IPO.
It appears to be an article of faith among the professional chatterers of the market-news media that THE NUMBERS MUST GO UP!!!!. If interrogated directly, of course, they will concede that 'the market' sometimes requires that the numbers go down, as folly and weakness are eliminated; but day-to-day this saddens them.
Just look at the body of media drivel generated by the recent deflation of the American housing bubble: having a place to live became more affordable than it had been in decades and every last talking head and politician available began screaming about the 'housing crisis'...
There probably are genuinely analytical analysts(who know enough to keep their mouths shut and make real money); but the ones bloviating in public appear to be little more than cheerleaders at a sort of stock market pep rally.
It's extra hardware(and not inexpensive, from what I've read); but I would think that the classic 'terminal window on ttyS0' would be an nearideal match for an 80-column refreshable braille display...
I've no doubt that Windows has superior 'kludging some degree of blind usability on top of a GUI' software offerings, because there is some serious cash in the 'corporations that don't like ADA suits' market; but I would(perhaps naively) expect that the unix-style environment(not exclusive to Linux, of course, BSD would do just as well) that had everybody on 80-column serial terminals back before graphics were cheap would be much easier to adapt more directly... A contemporary Window-manager-with-all-bells-and-whistles Linux desktop, not so much, though.
That sounds a lot like a local micro/pico/femto cell base station, which I've read that carriers are commonly installing in their retail locations to avoid embarassing connectivity/throughput issues when a number of customers are hammering away at the demo units.
It would allow HQ to see that IMEI XYZ appears to be accessing NSFW.com within 30 meters of the store at times not-so-coincidentally similar to those times when human-resource-peon Smith is scheduled for work; but I'd be pretty surprised if that were the primary purpose. Installing a zillion weedy little cells is something you do because they only work if they are on site. Surveillance is both cheaper and much more secure against tampering if you do it closer to the center of the network. There are a lot of people who could get a look at the suspicious looking box in the back room of the local Cell Shack. There are a great deal fewer who get to go inside the windowless mystery bunkers where the bigger gear lives...
The 'due to legal issues' does sound an awful lot like the 'friend' wants a look a traffic going across his network, but from devices that are owned by somebody else.
Given the amount of case law that was basically formulated to address the "Employee is dicking around on facebook, doesn't like the fact that Network caught them at the IDS/Firewall/etc." case, it might well be legal to tap somebody under those conditions; but it's hard to make the case that it is ethical to tap somebody on the wire when they are enough of a legal agent that you couldn't tap them at the endpoint... I get this funny feeling that somebody's wife/girlfriend/roommate/other is going to be lawyering up in the not so distant future...
There are a few ghastly legacy applications that oozed out of the 'we'll just build our own damn widget set, because!' school, which obey absolutely no system-wide settings whatsoever, not theme, not font size, screenreaders can forget about it, and so forth; but those are thankfully rare... Even then, you can always just give the contrast, gamma, or color curves a good hard shove at the driver or monitor level.
More importantly, essentially all South Koreans under the age of 40 have the ability to stop a zerg rush encoded into their brainstems through sheer practice...
I don't know how much is purely down to inertia and inefficiency(these certainly cannot be ruled out); but I get the impression that the US postal service has a certain cultural attachment to a slightly retro ideal of universal service going back to their original constitutional mandate. This is of pretty questionable use in their contemporary capacity as high-volume junk mail distributers with a side of certified legal mailings; but my interactions with postal personnel(especially in smaller markets) has always given me the impression that they take a certain pride in the fact that anyone can scrawl a vague reference to somewhere in Podunk on an envelope, slap on a stamp, and have it actually arrive at the correct slice of nowhere, courtesy of the postman who knows that area.
Fedex, on the other hand, you expect the barcodes and the little scanner/PDA widget.
As noted, it isn't obvious that this cultural orientation is a good fit for the position that the service finds itself in; but it has always struck me as an interesting phenomenon...
If nothing else, it's a cost-saving measure. Politically, you aren't allowed to use your nukes, or even test them much anymore; but that doesn't make them any cheaper to maintain, stockpile, keep ready for use, etc. A shiny, bilateral, humanitarian, oh-so-noble, strategic arms reduction treaty serves as a pleasantly face-saving way for both the US and Russia to say 'fuck it, this shit isn't worth the money'.
Nobody wants to be the first to blink; but being a 'responsible' nuclear power is actually pretty lame. You aren't allowed to do crazy tinpot dictator stuff, and your large conventional forces deter smaller competitors even without nukes; but you still have to pay upkeep, fret about loose cannons sneaking off with a warhead when you aren't looking, and so forth.
I'd wager that none of the nuclear powers would want to get rid of all their toys; but that the US and Russia are delighted to have been able to free up some of the cash spent on the ludicrious overkill buildup levels of the cold war and divert that to other uses...
TFA's thesis is that there simply isn't a dividing line between 'tactical' and 'strategic' that makes your sentence meaningful.
Given the mixture of forward bases and in-air refueling that the nuclear powers certainly aren't going to give up(even if their intentions are purely conventional, being able to B-52 bits of the middle east is just too convenient) delivering a 'tactical' warhead(or 10) right down a 'strategic' target's chimney is a matter of little more than swapping some hardpoints, and the upper edge of the dial-a-yield ranges for 'tactical' devices are well into the realm of 'would ruin a population center's day' territory.
There are, certainly, some unambiguously 'strategic' weapons, of the 'bloody huge thermonuclear warhead on an ICBM' school and there are (a dwindling number of) oddball low-yield artillery shells, demolition charges, and other oddities from the heyday of nuclear optimism(for sheer weirdness, I'm fond of the 'Davy Crockett' system. Essentially a 'technical' as much beloved by ragged 3rd world armies; but with nuclear warheads...); but much of the active hardware falls into the awkward middle ground where, in order to be powerful enough to be worth the trouble of being nuclear, it could certainly bestrategic if asked, and purely because of what people expect from contemporary delivery sytems, even for boring chemical explosives, 'strategic' delivery capabilities are widespread.
Fatality!
As best I can tell, MS purchased the kinect technology with the relatively narrow objective of coming up with something that could beat the Wii's motion-gimmick factor without requiring a new generation of console. They didn't even do the traditional vapor-announce-to-discourage-competitors of a Windows SDK until well after they had finished losing their shit at the people who went ahead and made their own. I don't know if they got a good price from the guys they licensed it from, based on some agreement to limit the scope to cheap gaming stuff, or whether they were just feeling amazingly shortsighted...
It is more curious that the company that MS licensed the tech from hadn't been pushing it left, right, and center before the Kinect made headlines. Even if their lower-volume hardware was a few grand a pop, they could still have found plenty of buyers in markets where that is entirely reasonable.
Certainly, if you are willing to use violence you can do whatever you can get away with. However, while Sharia pretty much sucks by humanist standards, it is subject to its own internal mores and isn't arbitrarily flexible. In this case, the limiting factor(beyond the basic 'Does the idea that Mohammed gave a damn what an internet search company in the early 21st century called that body of water even pass the laugh test?' problem) would likely be that this particular question of naming is an internal spat.
The leading contenders are "Persian Gulf", favored by ethnic nationalists in Persian-majority Iran, and "Arabian Gulf", favored by assorted ethnic nationalists in Arab-majority areas of the middle east. That(along with the not-so-minor fact that you also have Shia/Sunni split) would probably get a Fatwa endorsed by some Iranian theologian laughed right out of court among a substantial majority of the world's Muslims, never mind everybody else. This wouldn't be one of those unifying issues, like being against Rushdie or thinking that Israel pretty much sucks.
Some sort of civil-law 'actions calculated to destabilize the state' nonsense would probably be a lot easier to kangaroo-court through; but then Google would need to have some sort of operations within Iranian jurisdiction.
What sort of legal action, exactly, can you take against somebody for making a map that hurts your feelings?
I've never heard of any legal success in the US on those grounds(though PR concerns certainly motivate companies to tread lightly) and under the present US sanctions, Google is unlikely to be officially operating within Iran at all(given the SSL MiTM incident a while back, there are obviously Iranians using Google services; but the sanctions make it rather unlikely that Google would choose to site any official branch offices there), so it hardly matters whether Iran has a rule against hurting their feelings in the process of mapping.
Using masses of GPUs instead of the slow cpu to do decryption work?
GPUs and FPGAs have, in a number of cases, moved attacks on known-vulnerable systems from 'theoretical; but of great concern' to 'desktop, you don't even need 3-phase 220' faster than other technologies would have; but the cryptographic systems that are actually trusted tend to be of the 'barring fundamental breakthroughs in either mathematics or physics, converting all the mass in the solar system into crypto-chips it would merely shave a few zeros off the expected time...' variety.
Considering the (mostly) invincible state of good encryption, this seems unnecessary. Sure, it is a fun idea, but not a practical one.
I suspect that(aside from simply being relatively cheap to implement, and having some expected sales based on the 'cool' factor alone) the real purpose of any system purporting to substitute for, or complement, disk encryption is to deal with weaknesses unrelated to the cryptographic system itself.
As best we know, contemporary crypto systems with keys of reasonable length are not breakable in any useful sense. However, since humans who can store keys of reasonable length are vanishingly rare, most such systems must weaken themselves by storing a good chunk of the key somewhere where it can be recovered(on the device, with a password/passphrase, embedded in a smartcard/fob that is merely expensive and inconvenient to crack and extract, etc.)
If you just hand somebody an AES blob and tell them to come up with the correct 256 bits for themselves, they are SOL. If somebody's HDD is encrypted such that it can be unlocked with a password it is quite likely that dictionary-aided brute force will be enough. If the key is on a smartcard IC it might not be cheap to pull the key; but they don't produce those things in volume for a few bucks a chip, at most, by using all available security features...
After all, they are excellent tools for padeophiles and terrorists. Amirite?
I suspect that it depends how much other evidence they have. Nuking the only copy of something vital might well save you from conviction on more serious charges, at the cost of some sort of obstruction of justice/destruction of evidence charge. If they already have corroborating evidence from other sources, though, the DA will probably just smirk and add another charge, complete with trivially available evidence that makes you look guilty as hell.
"Yes, your honor. We detected child pornography downloads from the IP that the defendant's ISP had assigned to his residence at the time. When we arrived to execute a search warrant, the defendant dove toward his computer and destroyed its hard drive...."
I suspect that there are (at least) two distinct schools of utter fail:
The professionals, with a legacy in CCTV-as-in-actual-closed-circuit-running-on-private-coax, probably have an attitude much as you describe. The classic CCTV systems were dumb as bricks(not that their designers necessarily were, making largely analog, reasonably high bandwidth systems actually work in practice isn't trivial); but that lack of sophistication served as a strong defense against anybody without a physical tap shoved right into the coax. You just don't develop a very strong culture of caring about remote exploits if your engineering history is almost entirely concerned with systems that are incapable of remote anything, whether you like it or not.
Then you have the upstarts(either new companies, or rebadged ODM crap sold by existing ones), who design CCTV systems on the premise that a CCTV camera is basically just an embedded linux board with a camera interface, and a record/playback system is basically just an x86 with some sort of h264 hardware and a lousy frontend. These assumptions are not false, and advances in silicon sensors and cheap embedded computers definitely mean that the price is right; but the standards of security excellence in low-cost embedded gear are absolutely fucking dire... These guys should know better, since their designs are 100% post-ubiquitous-networking in concept; but they just don't get paid enough, or enjoy long enough development cycles, to give a damn.
Nothing recent enough to be without an ethernet port. If they've made the dongles worth, though, then they have the 'plan B' mentioned, and it shouldn't be a big deal.
They used to be able to beat Intel in the Athlon days. Now they are hopelessly far behind and dumping huge hot graphics cores into their chips putting them further and further behind. Focus on cheap compute with unlocking cores AMD. Not stupid graphics cores which do nothing for the CPU. A 16 core phenom ii at $100 will sell much better than this insane graphics + cpu crap.
That is pretty much the exact opposite of a good plan for AMD(as much as I would like cheap compute...) Since Intel has a process advantage, and presently has a superior x86 compute core architecture, they can almost certainly beat AMD on production cost for chips of a given level of punch. Trying to compete on price with somebody kicking out chips a process node ahead of you just isn't a good plan. Unless they really fuck it up, or their yields tank horribly or similar, they'll be able to beat you on production cost every time. Intel has little to gain by cutting its own margins in order to chase AMD down a hole(since lower margins are bad, and killing AMD would mean becoming antitrust scrutiny case #1 for the indefinite future...); but there isn't any architectural barrier to their doing so.
Since Intel has comparatively worthless GPU designs, tacking GPUs onto CPU dice is a way for AMD to offer something that Intel cannot(and at a price lower than a discrete CPU + discrete GPU without totally cutting their own throat), and also happens to go well with today's enthusiasm for laptops and all-in-ones. They have a second niche, much more directly focused on price, in compute-light, memory-heavy server applications(since you can populate your sockets with AMD CPUs for less and the number of DIMMs you get is roughly proportional to the number of sockets you have active); but competing on price isn't good for your margins.
With an inferior process and a weaker x86 design, gunning directly for the compute performance crown would just be asking for a whupping from Intel.
Even the fancy, we-make-cardiac-stents-out-of-this-stuff, grade is considered to be only moderately biocompatible. Apparently safe enough to beat the alternative; but they aren't pouring work into fancy new biocompatible polymers because steel is good enough...
I'm not so surprised that some rather alarmingly powerful beam sources would be operated quietly by people with atypical sensor needs. I am a bit surprised that 3.5 lbs of highly enriched Uranium would be available to serve as a beam source.
Not telling the neighbors about a scary-sounding piece of industrial/scientific apparatus is one thing, having enough nuclear material to interest a proliferation wonk in your basement, on the other hand, seems like it would raise eyebrows...
What, exactly, do you think companies use computers for, anyhow?
I've never run into an IT setup of any size that didn't consider "install approved image" as step 1 of deploying a new system, or redeploying one previously used elsewhere. The tools for doing this over ethernet (PXE/Netboot, dump image to disk, reboot) are commonly available and mature. Over wifi? Not so much. Over USB ethernet dongles? Surely you jest.
Doesn't matter much for home users, or very small scale outfits where the slower 'boot from external volume, fire up disk utility, copy disk image to internal volume' method is acceptable; but Apple had better have a plan B for everybody who wants to image machines in bulk if they fancy institutional sales...