But the steel in the chassis is probably the most environmentally friendly part of a server.
Also, unless you are going blade-level nuts about density(which the highly cost-sensitive massive cloud guys don't generally seem to be), it's the part of the server that can be standardized with little more than some basic agreement about screw holes and airflow...
There have been improvements; but a 1995-vintage ATX case can be trivially populated with hardware purchased today(even an AT case could probably be bodged, with a few standoffs and some swearing). Worst case, a part that is essentially 100% steel is something that you won't exactly have to fight to get a recycler to accept.
The ugly trick is interconnect performance, unless you aren't planning to scale up very much at all or have the (atypical) good fortune to be attacking nothing but hugely parallel problems.
It's been a while since the supercomputer crowd found rolling their own esoteric CPUs to be worth it(with POWER the possible exception); but if all the silicon you want to devote to the problem won't fit on a single motherboard, you quickly enter the realm of the rather specialized.
At very least, you are probably looking at doing some networking as or more costly than a 10GbE setup. If you want a single system image, well, call your vendor and get your checkbook warmed up...
For those that were not previously aware, banking via email or smartphone is begging to have your account emptied.
Architecturally, 'smartphone' is pretty much identical to 'pc', possibly more secure in practice. If a phone is 'smart' in any serious sense, it will support the same HTTP+SSL arrangement that you'd use on a computer. SMS, on the other hand, combines the very finest weaknesses of email with those of a direct connection to your local malevolent telco's billing infrastructure...
It could have been the fault of the hacks who hacked together the image; but my one experience with a MIPS based Android tablet was... dire.
I didn't even get far enough to learn what Android market applications didn't work on it, since most of the default Android applications would crash if you tried to do anything serious with them(and by 'serious' I mean "load google.com with the Android browser", that sort of strenuous work).
I have no reason to assume that MIPS is fundamentally incapable of running Android properly(ARM is clearly Android's favorite child; but the x86 builds work fine and the fact that it isn't based on a 'real' JVM means that some ARM-only like Jazelle isn't an issue); but the state of Android-on-MIPS in the wild is pretty unpleasant. Google hasn't targeted the platform, so no love there, none of the MIPS licensees have nearly the sort of war chest(and sense of urgency) that Intel brought to their x86 work, and the "Chinese anony-tablets so cheap that they couldn't afford an Allwinner A10" market does not attract quality firmware engineers...
Why a user would not simply install MS Security Essentials and be done with it?
Among other considerations(like central management), I'm pretty sure that the MSSE license frowns on use in anything larger than a home/home office type environment.
At least in its Windows incarnation, Steam doesn't run with admin privileges. It simply, when necessary, starts whatever the installer is with a request for elevation. You get the pop-up and/or username and password prompt(depending on your system settings) and the privileged process does whatever install needs doing while steam continues to chug along at its usual level. Presumably a linux implementation of the idea would work in roughly the same way: Steam downloads the installer package and, depending on whether it is a binary installer or a.deb/.rpm package, kicks up a sudo prompt to allow you to run the installer or your package manager as root.
That said, the main incentive to have Steam offer GPU driver updates is that the ones provided through Windows update are always ages behind the curve, and having a vendor-specific update-nagging tray object is annoying. On a system with a package manager, there really isn't much point.
Just pretend that your cable mess is life critical and/or supporting something that you'd need to leave the gravity well to fix. Then do it the correct way.
TFA is pretty useless and doesn't indicate what sorts of fiber this works on, or why it is different from other OOFDM-related work; but is there any reason to suspect that a technology that improves fiber transmit rates wouldn't help the CO backbone link speed as well?
Given the, um, vigorous state of competition in the broadband market, it isn't clear that that will matter much; but if they have some new secret sauce that makes transmissions over fiber faster it would, naively, seem to be something that could be added to any part of the network carried over fiber.
The Cadillac target market seems to be eerily similar to the cognac target market: a combination of old, at least vaguely affluent, white guys and young hip-hop aspirants. I don't know how it happened.
1. The fact that certain not-officially-known nation states have pulled off highly sophisticated attacks is a matter of public knowledge.
2. A 'nation-state', as an entity that gets to collect taxes and is charged with assorted non-market processes like 'defense', has much broader ability to do things that make minimal financial sense. If you are worried about spammers, or PIN-skimmers, or whatnot it suffices to be more expensive to attack than your resources are worth. If you are in some clandestine entity's sights, you actually have to be hard, rather than simply uneconomic.
3. Even if (and this is far from obviously true) all state employees are a bunch of drooling idiots herded by ideological Kommisars, they always have the option of contracting the attack out to their private sector superiors. Plenty of contractors who specialize, or have a department specializing in, electronic attack tools and they'll hold your hand every step of the way if you cut them large enough checks.
What is "proprietary" supposed to tell me about hardware?
There is just so much wrong with calling things "proprietary" and thinking it'll make the reader perceive the product as superior.
TFS does a terrible job of, um, summarizing the situation; but it does actually make sense in context:
Intel's initial entry into SSDs(X-25) was based on an in-house controller, which(with the exception of the unpleasant 8MB firmware bug) was generally quite well regarded. Then it stagnated. They did a few tepid bumps and firmware updates; but no successor controller appeared. With SSDs actually able to saturate a 3GB/s SATA bus, the fact that Intel had nothing on the table for 6GB/s SATA began to become an issue.
More recently, Intel began shipping 3rd party controllers (most recently Sandforce, possibly some Marvel at some point) on everything but their enterprise gear.
Now, after the thick end of four years, they've brought out their first new SSD controller architecture. Whether it does, in fact, turn out to be better is not known; but it is news after such a long hiatus.
The 'kinect' sensor package includes pretty robust(for the price) depth detection. There are also a fair number of demo applications with adequate-if-not-exactly-perfect facial expression tracking.
It would probably add some false negatives; but the hardware capabilities are there to reject all 2D fakes, as well as weeding out 3D fakes that are inhumanly static...
Normal technology is designed to fail as robustly as possible. Since video input is non-essential to media playback, graceful degradation and continued operation would be trivial.
DRM isn't normal technology. It's technology that is designed to fail as dramatically as possible. Expect any minor anomalies in the 'trusted' system state to be treated as catastrophic and absolutely incompatible with continued playback.
On the other hand, I don't condone it, and if it became subsidized by government money (at some point in the future), then I would then be forced to have an opinion on it because then I become responsible for paying for it.
Hopefully, someone does the smart thing and leaves the funding for the actual actions to a nice, private charity organization made up of people who agree with it.
Why would you handle the costs of a lethal dose of anesthetics differently than a sub-lethal dose(or, more likely, a substantial sequence of sublethal doses)?
If medical care is state-financed(either in the form of direct aid to the poor or in the form of compulsory insurance risk pooling) the public does have an interest in curtailing treatments viewed as excessively costly for their benefit(because resources are hardly finite); but they really ought to be careful about tampering with the independence of other people's doctor/patient interactions lest other people return the favor. In the case of assisted suicide, it's almost certainly going to be cheaper than the alternative, so somebody who chooses it is hardly picking your pocket(if you are on the hook somehow for his care, he's doing you a favor) so it's hard to object on absolute cost grounds, and do you really want to start tampering with patient autonomy if you are against the option that is cheaper for everybody else?
I'd be less worried about the lithium-water reaction(Li-ion batteries tend to be sealed, if only so the internals don't degrade even faster than usual, they are touchy things) and more worried about a short circuit anywhere near a battery pack punchy enough to run a car. At 330 volts, you don't need an ultra-low resistance path to get some serious current flowing, and serious current is something that large battery packs are more than happy to supply.
Now, once the electrical heating breaches the seals and touches off a merry metal fire, you have additional problems...
I realize that this is a lost cause and all; but why would you endorse a 16:10(at least it's not bloody 16:9...) rather than a 4:3 for a laptop? For a tablet, sure, where you can change the orientation and turn your sprawling rectangle into a nice, readable, page-width reading surface; but a laptop, where the keyboard keeps you from doing that?
If virtually all laptop displays are going to be laid out as though they are used for nothing but watching movies it would be nice if they at least threw in some additional pixels; but do we have to give up the shape that is better for dealing with text in a reasonably sized package? Absurdly wide desktop screens are fine, because you can just make them larger, and treat them as multiple page-sized screens when needed; but laptops have space constraints to deal with...
It's always a dangerous assumption to make; but architecturally the concept of 'partially hacked' isn't terribly nonsensical. Consider the enormous number of web server setups where OS-level credentials and web application authentication are entirely different things. It happens all the time that kiddies will crack the web component and scribble all over your php forum or CMS or whatnot; but without ever gaining access to the OS.
You really don't want to work on the assumption that 'eh, I'm sure we were only partially hacked, no need to reinstall the OS'; but it may well often be true.
You would think that a company playing at something mildly important(like, oh being a CA for the Dutch government...) could, at very least, do basic things like store logs on WORM tape... Yes, those are overpriced compared to the normal ones; but they aren't that expensive.
Posse commitatus' legislative history basically boils down to southern insurgents, magnanimously granted congressional representation amazingly quickly, voting against the continuation of counter-insurgency efforts so that they could continue rebelling with greater convenience...
I wonder how much of that "domestic development" was looking at what foreign companies were making in their country and surrounding areas?
Has anybody even been working on a MIPS processor that they could outsource, for anything larger than little router boxes and things, since DEC died a horrible death ages ago?
Why do people freak out if we send unmanned drones up to surveil things, but if you stick a guy in the plane suddenly it's OK?
Most of the protection of your privacy is economic, rather than legal or technological. A guy in a plane or a helicopter is Not Cheap, per hour, which creates a sort of 'de facto probable cause' requirement, since the cops can only justify sending one up if they think that they'll find something worth finding.
Drones are cheaper(still pretty expensive now, getting less so), which means that the economic disincentives to surveillance fall and people enjoy less actual protection from surveillance(since the strict legal protections are markedly lower than the historical economic ones).
Say farewell forever to even the concept of posse comitatus, limited as it was. Now it is just a Latin phrase you never heard of.
Oh, don't you worry your pretty little head about that. The military won't technically do any law enforcement(though it may prove necessary to engage in certain 'domestic Force Protection' activities in order to safeguard DoD assets and personel...), they'll just fire-sale off military hardware under the Law Enforcement Support Office(unless you trust DoD certs, you'll probably get an SSL warning here) program to various police SWAT teams who will then use it for them.
See, absolutely nothing to worry about. Yes, the police may be logistically indistinguishable from your average upper-developing-world mechanized infantry; but the org chart says they aren't military, so it's all good.
I suspect that we'll just have to kill a lot of fuzzy little animals in order to find out if those binding sites are specific to pathogens or whether they show up elsewhere...
Incidentally, if you want a category of vaccines that seems like it is just begging for dramatic trouble, how about Immunocontraceptives? Already used with success in a variety of nuisance mammals; but uneconomic for use in smaller, more numerous, or harder-to-catch pests(because it has to be injected to work). So, logically enough, work is ongoing to produce virally delivered vaccines that will spread themselves through the target population!
Depends on who 'they' are. If it's a private outfit putting up their money in the hopes of developing a marketable product, no, I'd expect to see it priced at whatever premium over the current annual strain-specific vaccines they think that they can get.
If it's research done by one or more of the assorted state-funded public health medical research institutes or university researchers working under similar grants, then it's already been paid for, and I'd hope to see it being farmed out for production with a more or less exclusive focus on making it cheap.
But the steel in the chassis is probably the most environmentally friendly part of a server.
Also, unless you are going blade-level nuts about density(which the highly cost-sensitive massive cloud guys don't generally seem to be), it's the part of the server that can be standardized with little more than some basic agreement about screw holes and airflow...
There have been improvements; but a 1995-vintage ATX case can be trivially populated with hardware purchased today(even an AT case could probably be bodged, with a few standoffs and some swearing). Worst case, a part that is essentially 100% steel is something that you won't exactly have to fight to get a recycler to accept.
The ugly trick is interconnect performance, unless you aren't planning to scale up very much at all or have the (atypical) good fortune to be attacking nothing but hugely parallel problems.
It's been a while since the supercomputer crowd found rolling their own esoteric CPUs to be worth it(with POWER the possible exception); but if all the silicon you want to devote to the problem won't fit on a single motherboard, you quickly enter the realm of the rather specialized.
At very least, you are probably looking at doing some networking as or more costly than a 10GbE setup. If you want a single system image, well, call your vendor and get your checkbook warmed up...
From the department of No Shit Sherlock!
For those that were not previously aware, banking via email or smartphone is begging to have your account emptied.
Architecturally, 'smartphone' is pretty much identical to 'pc', possibly more secure in practice. If a phone is 'smart' in any serious sense, it will support the same HTTP+SSL arrangement that you'd use on a computer. SMS, on the other hand, combines the very finest weaknesses of email with those of a direct connection to your local malevolent telco's billing infrastructure...
It could have been the fault of the hacks who hacked together the image; but my one experience with a MIPS based Android tablet was... dire.
I didn't even get far enough to learn what Android market applications didn't work on it, since most of the default Android applications would crash if you tried to do anything serious with them(and by 'serious' I mean "load google.com with the Android browser", that sort of strenuous work).
I have no reason to assume that MIPS is fundamentally incapable of running Android properly(ARM is clearly Android's favorite child; but the x86 builds work fine and the fact that it isn't based on a 'real' JVM means that some ARM-only like Jazelle isn't an issue); but the state of Android-on-MIPS in the wild is pretty unpleasant. Google hasn't targeted the platform, so no love there, none of the MIPS licensees have nearly the sort of war chest(and sense of urgency) that Intel brought to their x86 work, and the "Chinese anony-tablets so cheap that they couldn't afford an Allwinner A10" market does not attract quality firmware engineers...
Why a user would not simply install MS Security Essentials and be done with it?
Among other considerations(like central management), I'm pretty sure that the MSSE license frowns on use in anything larger than a home/home office type environment.
At least in its Windows incarnation, Steam doesn't run with admin privileges. It simply, when necessary, starts whatever the installer is with a request for elevation. You get the pop-up and/or username and password prompt(depending on your system settings) and the privileged process does whatever install needs doing while steam continues to chug along at its usual level. Presumably a linux implementation of the idea would work in roughly the same way: Steam downloads the installer package and, depending on whether it is a binary installer or a .deb/.rpm package, kicks up a sudo prompt to allow you to run the installer or your package manager as root.
That said, the main incentive to have Steam offer GPU driver updates is that the ones provided through Windows update are always ages behind the curve, and having a vendor-specific update-nagging tray object is annoying. On a system with a package manager, there really isn't much point.
Just pretend that your cable mess is life critical and/or supporting something that you'd need to leave the gravity well to fix. Then do it the correct way.
The suffering will make you a better person.
TFA is pretty useless and doesn't indicate what sorts of fiber this works on, or why it is different from other OOFDM-related work; but is there any reason to suspect that a technology that improves fiber transmit rates wouldn't help the CO backbone link speed as well?
Given the, um, vigorous state of competition in the broadband market, it isn't clear that that will matter much; but if they have some new secret sauce that makes transmissions over fiber faster it would, naively, seem to be something that could be added to any part of the network carried over fiber.
The Cadillac target market seems to be eerily similar to the cognac target market: a combination of old, at least vaguely affluent, white guys and young hip-hop aspirants. I don't know how it happened.
A few factors come to mind:
1. The fact that certain not-officially-known nation states have pulled off highly sophisticated attacks is a matter of public knowledge.
2. A 'nation-state', as an entity that gets to collect taxes and is charged with assorted non-market processes like 'defense', has much broader ability to do things that make minimal financial sense. If you are worried about spammers, or PIN-skimmers, or whatnot it suffices to be more expensive to attack than your resources are worth. If you are in some clandestine entity's sights, you actually have to be hard, rather than simply uneconomic.
3. Even if (and this is far from obviously true) all state employees are a bunch of drooling idiots herded by ideological Kommisars, they always have the option of contracting the attack out to their private sector superiors. Plenty of contractors who specialize, or have a department specializing in, electronic attack tools and they'll hold your hand every step of the way if you cut them large enough checks.
What is "proprietary" supposed to tell me about hardware?
There is just so much wrong with calling things "proprietary" and thinking it'll make the reader perceive the product as superior.
TFS does a terrible job of, um, summarizing the situation; but it does actually make sense in context:
Intel's initial entry into SSDs(X-25) was based on an in-house controller, which(with the exception of the unpleasant 8MB firmware bug) was generally quite well regarded. Then it stagnated. They did a few tepid bumps and firmware updates; but no successor controller appeared. With SSDs actually able to saturate a 3GB/s SATA bus, the fact that Intel had nothing on the table for 6GB/s SATA began to become an issue.
More recently, Intel began shipping 3rd party controllers (most recently Sandforce, possibly some Marvel at some point) on everything but their enterprise gear.
Now, after the thick end of four years, they've brought out their first new SSD controller architecture. Whether it does, in fact, turn out to be better is not known; but it is news after such a long hiatus.
Can it identify a photograph and a real person ?
If not, it's easily hackable...
http://it.slashdot.org/story/12/06/18/184217/samsung-galaxy-s3-face-unlock-tricked-by-photograph
The 'kinect' sensor package includes pretty robust(for the price) depth detection. There are also a fair number of demo applications with adequate-if-not-exactly-perfect facial expression tracking.
It would probably add some false negatives; but the hardware capabilities are there to reject all 2D fakes, as well as weeding out 3D fakes that are inhumanly static...
Over the camera should solve the problem.
Normal technology is designed to fail as robustly as possible. Since video input is non-essential to media playback, graceful degradation and continued operation would be trivial.
DRM isn't normal technology. It's technology that is designed to fail as dramatically as possible. Expect any minor anomalies in the 'trusted' system state to be treated as catastrophic and absolutely incompatible with continued playback.
On the other hand, I don't condone it, and if it became subsidized by government money (at some point in the future), then I would then be forced to have an opinion on it because then I become responsible for paying for it.
Hopefully, someone does the smart thing and leaves the funding for the actual actions to a nice, private charity organization made up of people who agree with it.
Why would you handle the costs of a lethal dose of anesthetics differently than a sub-lethal dose(or, more likely, a substantial sequence of sublethal doses)?
If medical care is state-financed(either in the form of direct aid to the poor or in the form of compulsory insurance risk pooling) the public does have an interest in curtailing treatments viewed as excessively costly for their benefit(because resources are hardly finite); but they really ought to be careful about tampering with the independence of other people's doctor/patient interactions lest other people return the favor. In the case of assisted suicide, it's almost certainly going to be cheaper than the alternative, so somebody who chooses it is hardly picking your pocket(if you are on the hook somehow for his care, he's doing you a favor) so it's hard to object on absolute cost grounds, and do you really want to start tampering with patient autonomy if you are against the option that is cheaper for everybody else?
I'd be less worried about the lithium-water reaction(Li-ion batteries tend to be sealed, if only so the internals don't degrade even faster than usual, they are touchy things) and more worried about a short circuit anywhere near a battery pack punchy enough to run a car. At 330 volts, you don't need an ultra-low resistance path to get some serious current flowing, and serious current is something that large battery packs are more than happy to supply.
Now, once the electrical heating breaches the seals and touches off a merry metal fire, you have additional problems...
I realize that this is a lost cause and all; but why would you endorse a 16:10(at least it's not bloody 16:9...) rather than a 4:3 for a laptop? For a tablet, sure, where you can change the orientation and turn your sprawling rectangle into a nice, readable, page-width reading surface; but a laptop, where the keyboard keeps you from doing that?
If virtually all laptop displays are going to be laid out as though they are used for nothing but watching movies it would be nice if they at least threw in some additional pixels; but do we have to give up the shape that is better for dealing with text in a reasonably sized package? Absurdly wide desktop screens are fine, because you can just make them larger, and treat them as multiple page-sized screens when needed; but laptops have space constraints to deal with...
It's always a dangerous assumption to make; but architecturally the concept of 'partially hacked' isn't terribly nonsensical. Consider the enormous number of web server setups where OS-level credentials and web application authentication are entirely different things. It happens all the time that kiddies will crack the web component and scribble all over your php forum or CMS or whatnot; but without ever gaining access to the OS.
You really don't want to work on the assumption that 'eh, I'm sure we were only partially hacked, no need to reinstall the OS'; but it may well often be true.
You would think that a company playing at something mildly important(like, oh being a CA for the Dutch government...) could, at very least, do basic things like store logs on WORM tape... Yes, those are overpriced compared to the normal ones; but they aren't that expensive.
Posse commitatus' legislative history basically boils down to southern insurgents, magnanimously granted congressional representation amazingly quickly, voting against the continuation of counter-insurgency efforts so that they could continue rebelling with greater convenience...
A domestically developed MIPS processor?
I wonder how much of that "domestic development" was looking at what foreign companies were making in their country and surrounding areas?
Has anybody even been working on a MIPS processor that they could outsource, for anything larger than little router boxes and things, since DEC died a horrible death ages ago?
Why do people freak out if we send unmanned drones up to surveil things, but if you stick a guy in the plane suddenly it's OK?
Most of the protection of your privacy is economic, rather than legal or technological. A guy in a plane or a helicopter is Not Cheap, per hour, which creates a sort of 'de facto probable cause' requirement, since the cops can only justify sending one up if they think that they'll find something worth finding.
Drones are cheaper(still pretty expensive now, getting less so), which means that the economic disincentives to surveillance fall and people enjoy less actual protection from surveillance(since the strict legal protections are markedly lower than the historical economic ones).
Say farewell forever to even the concept of posse comitatus, limited as it was. Now it is just a Latin phrase you never heard of.
Oh, don't you worry your pretty little head about that. The military won't technically do any law enforcement(though it may prove necessary to engage in certain 'domestic Force Protection' activities in order to safeguard DoD assets and personel...), they'll just fire-sale off military hardware under the Law Enforcement Support Office(unless you trust DoD certs, you'll probably get an SSL warning here) program to various police SWAT teams who will then use it for them.
See, absolutely nothing to worry about. Yes, the police may be logistically indistinguishable from your average upper-developing-world mechanized infantry; but the org chart says they aren't military, so it's all good.
I suspect that we'll just have to kill a lot of fuzzy little animals in order to find out if those binding sites are specific to pathogens or whether they show up elsewhere...
Incidentally, if you want a category of vaccines that seems like it is just begging for dramatic trouble, how about Immunocontraceptives? Already used with success in a variety of nuisance mammals; but uneconomic for use in smaller, more numerous, or harder-to-catch pests(because it has to be injected to work). So, logically enough, work is ongoing to produce virally delivered vaccines that will spread themselves through the target population!
Depends on who 'they' are. If it's a private outfit putting up their money in the hopes of developing a marketable product, no, I'd expect to see it priced at whatever premium over the current annual strain-specific vaccines they think that they can get.
If it's research done by one or more of the assorted state-funded public health medical research institutes or university researchers working under similar grants, then it's already been paid for, and I'd hope to see it being farmed out for production with a more or less exclusive focus on making it cheap.
Ah; but MMR was all part of the Big Pharma/reptoid autism conspiracy, so they were willing to accept lower margins on that one...