Ah, ok. That would explain the slightly unpredictable amount of time, following a 'bus reset' message before ports in that port cluster came back online. I would have expected a more precise period from some sort of fancy power-control IC; but a fully analog component would make perfect sense.
Based on some unpleasant experiences with a USB printer that had a neat internal short, my impression is that a device has to be really nasty to just die when subjected to excessive attempted current draw by a peripheral. You actually do get a little 'host reset' message and no permanent damage, really rather civilized. I don't know whether it's something clever or just a re-settable fuse; but it doesn't seem to result in the hard-kill you'd expect from an ordinary fuse.
For compatibility purposes, though, all kinds of attempted power draws that are off-spec but below whatever the device considers dangerous are generally well tolerated. USB HDDs drawing ~800ma, or connected to the data lines of one port and the power lines of a second port, little fans and LEDs on goosenecks, all kinds of nasty stuff. Especially on desktops, where +5v is available in nigh-unlimited quantities. Laptops and routers and things, with actual power budgets, seem to be a bit pickier.
"Negotiate" is a loose term - really it's just some fixed resistances across the data pins that set USB charging mode. This can be built into the plug without any extra copper in the cable.
Sometimes a device/cable combination can work in either way. They'll have some goofy resistances-between-pins coding scheme for dumb chargers; but they'll also do the official USB SIG power negotiation dance if plugged into an actual USB host. Of course, so long as you only want to charge the device, it just has to work, not necessarily work the best or correct way.
I suspect that if they thought they could get away with that, they would. Taking the 'not our fucking problem' position costs substantially less than hiring a contractor to make it your problem.
These are completely different architectures, and I highly doubt portability will be easier between the two, just because one makes the jump to being native 64bit.
Emulating an ARM core on x86 is totally doable (QEMU, in fact, does it for the Android SDK's test/emulator component. I don't think anybody has an OSS iPhone clone fully working; but that wouldn't take Apple long); what I find baffling is the theory that the application being emulated would need to be 64 bit to 'play nice' with the larger memory space available to the host.
Even if Apple decided to support cross-platform binaries(and that's one hell of an 'if', Apple Loathes half-assed ports, and has so far not embraced touchscreens in desktops and laptops, so cross-platform binary support would be asking for half-assed ports in both directions); allocating one or more 32-bit processes its own happy little address space has been trivial on 64-bit OSes for ages. Hell, even pre-64-bit, PAE systems were happily slicing large quantities of RAM into 32-bit address chunks to support 32-bit processes. Since Apple makes no ARM devices with 4GB+ of RAM(indeed, they tend to ship rather on the low side compared to the pricier Android OEMs), iOS apps aren't exactly going to be starving on the desktop, and a desktop OS has no trouble carving up a larger memory space into smaller chunks.
Oh, noting requires that the RNG be on-die (and, as you say, there are all kinds of options that aren't, the universe is a very noisy place). However, unless your computer is actually configured to use whatever access to the noise of the universe it has (nasty little webcams are another good source), and dump that entropy in whatever pool(s) your software environment specifies, it doesn't help you much.
The on-die ones are valued mostly because they are really, really, fast and a whole lot cheaper than the earlier generations of purpose-built crypto coprocessors, which were very much priced for a niche market.
If you can't trust them, though, you have options.
Are we going to have another decade+ of handwringing about what will happen to the poor, poor, old and/or poor people whose constitutional right to TV will be destroyed if they need to buy an ATSC tuner with newer codec support? Because damn was the NTSC->ATSC transition annoying...
"Yeah, if they got serious they could push for the byte encoding again, or even make it the default output. But then, you wouldn't get much benefit if you had to mix non-.NET code with it."
I have particularly unpleasant memories of dealing with some application whose developers had managed to make it require.NET 4 and some 16-bit version of Installshield or something just to copy it into place, so I don't doubt that there is a lot of 'mostly ready' code that has resulted in products that aren't going anywhere but x86 without one hell of a shove.
I just find it somewhat ironic that Microsoft theoretically has one of the better technological positions to support a multi-architecture binary software ecosystem, and has been working on it for years, and yet managed to fumble so profoundly and make so little use of that.
The whole point of TFA is about a technique for (mostly undetectably) modifying a good hardware RNG and turning it into a really lousy one.
Getting your entropy from multiple places probably helps (if they don't know what 6 RNGs you chose it's harder to dope them all, and even if they do, they still have to slog through the entropy from multiple crippled sources rather than only a single one (and, while it is possible to cripple the RNG entirely, that will show up on tests, so plausible real-world implementations would still provide some entropy, just less than advertised).
Oh, I don't doubt that the TV broadcast standards people had their reasons, they might even have been good ones from the perspective of the awful legacy tech they were dealing with. I just bitterly resent how they managed to smear their shit all over the computer market. Had it stayed confined to broadcast television, they could have done whatever they liked for all I care.
"Or they could have tried to get Visual Studio to leverage LLVM and ship bitcode so things could be ever further future-proofed and extend to more than just 2 architectures."
That's the humorous part of all this: Microsoft started work (more than a decade ago, if I recall) on the 'Common Language Runtime' and the 'Common Language Infrastructure', with the 'Common Intermediate Language' playing the part of architecture-independent bytecode representation. It's ostensibly a standard and whatnot; but basically Microsoft's ".NET" is the serious implementation.
The already have, in house, widely used, supported by their dev tools, an architecture independent mechanism. Loads of ISVs even use it fairly extensively.
Architecturally, they might actually have the best position among any major vendor to make cross-platform binaries happen; but they threw it all away to try to have a mandatory app store. Elegant, really.
Exactly, that's the 'linux' angle on the 'Intel finally delivering chips that should finish nailing the coffin shut on Windows RT' story.
There obviously isn't much stopping Microsoft from having another try at iOS-envy and mandatory app stores on x86 (the implementation on ARM, using UEFI secureboot and a restrictive SRP in the windows image, would be 100% doable on x86); but as previously implemented, that was a major distinction: Wintels would be more or less as they have always been, Win-ARM would be walled garden dystopia. Hence the general enthusiasm for the demise of the same.
Thatcher said she would never privatise the post office.
And Reagan would be derisively referred to as a 'RHINO' if he were to attempt to gain office as a republican today... The acolytes tend to...get out of hand... in their veneration after the venerated has been dead or nonfuctional for a while.
Stop giving up your life to big business, and they'll stop being able to tear you a new one.
This isn't like that other article, where the British government is selling off a natural monopoly so you're forced to use a particular business. This is you thinking that you are entitled to get Zuckerberg to do anything more than widen the smile on his deservedly smug face.
Or, at least, chose a big business that doesn't have so much market power. There are plenty of suppliers who offer, y'know, actually-two-sided contracts, or (this one blows my mind; but it's true) actually run a business where they do better by not screwing over their customers. It's crazy. There are also suppliers who sell commodities, and the worst they can do is give their competitors' salesweasels a good day.
I'm not entirely taken with Nicholas Carr; but he has a useful little coinage to cover this situation: "Digital sharecropper".
It beats real sharecropping (sometimes you get air conditioning, and even paid in real money rather than scrip and debt peonage!); but if your business (or your hobby, though businesses tend to be more financially painful) depends on a third party, with which you have absolutely no leverage other than their power and mere pleasure, (and where your business consists largely of making their business incrementally more successful), you are a sharecropper. And, while the timing of the crackdown is sometimes rather baffling, since it doesn't even seem to be to the landlord's advantage, it is closer to being an inevitability than a mere possibility.
This doesn't mean that you have to do everything 100% alone in order to not be a sharecropper, commodities are safe enough, as are companies so mired in the demands of actually-powerful customers that they will have difficultly cutting the feet out from under you at a greater than glacial pace; but a situation where you are 100% dependent on a single third party who has the right, and the ability, to cut you down just by revoking an API key or deleting a page on their own servers? They own you.
Where is the linux angle? Not even a conversion coupon code?
It isn't mentioned; but 'Windows RT's fairly clear demise, in favor of cheap-ass x86 devices, is almost certainly good news for Linux(Not 'This is the year of Linux on the Desktop!!!' news; but good).
Per Microsoft's secure boot requirements, ARM-based 'Windows RT' hardware Must Not allow (either out of the factory, or by user modification) signing keys for boot payloads other than Microsoft's own and cannot allow disabling 'secure boot', while x86 Win8 devices can.
It remains to be seen how many will actually be purchased for linuxization; but Windows RT devices are (short of breaking TPM-backed UEFI secure boot) 100% useless for Linux, or anything else that isn't blessed by Redmond. Wintel tablets, though, are just funny shaped wintels, and so only the questionable state of Linux touch GUIs stands between you and installation.
I don't know whether Surface RT was a genuine fuckup, or whether it was Microsoft reminding Intel that they've ported the NT kernel before and can do it again in the hopes of spurring them to get their shit together on the low-TDP side; but either way RT looks dead, dead, dead. Under interpretation one, Microsoft gimped it hard enough, either to protect other parts of their business or to push 'winRT'(the runtime not the OS) that it was pretty sick already and Intel just shoved a knife in it's back. Under interpretation two, Intel appears to have risen to the challenge, or at least close enough that full binary compatibility with all things Windows will be worth more (to anybody considering a Microsoft product at all) than an extra sliver of battery life.
What I wouldn't give to be able to travel back in time and prevent 1366x768 or '720p' from being defined as 'HD' resolution. Ideally with some sort of plan that involves more explosions than a braindead summer action movie. What a pox upon the eyes of the world, especially with so many applications making poor use of extra horizontal space (so it's barely better than 1024x768, circa 15 years ago) and 768 pixels being pretty narrow for the 'well, just flip it 90 degrees' strategy that saves other widescreens for non-movie purposes.
"You would think that the private sector could manage to do at least one thing better than the British government, wouldn't you?"
Apparently, they are a great deal better at recognizing which side the bread is buttered on in the 'privatization' deals than the British government, Does that count?
Bah, why 'say' "trigger warning" when we could just have an XML-based, machine readable, semantic tagging mechanism for trigger warnings? Clearly a superior solution.
The vermin only teaze and pinch
Their foes superior by an inch.
So, naturalists observe, a flea
Has smaller fleas that on him prey;
And these have smaller still to bite 'em,
And so proceed ad infinitum.
Slashdot Mirror
Ah, ok. That would explain the slightly unpredictable amount of time, following a 'bus reset' message before ports in that port cluster came back online. I would have expected a more precise period from some sort of fancy power-control IC; but a fully analog component would make perfect sense.
[sigh] So much for "Universal".
Hey, man, not my problem if two passengers don't get along, I just drive the bus.
-USB Implementors' Forum, Inc.
Based on some unpleasant experiences with a USB printer that had a neat internal short, my impression is that a device has to be really nasty to just die when subjected to excessive attempted current draw by a peripheral. You actually do get a little 'host reset' message and no permanent damage, really rather civilized. I don't know whether it's something clever or just a re-settable fuse; but it doesn't seem to result in the hard-kill you'd expect from an ordinary fuse.
For compatibility purposes, though, all kinds of attempted power draws that are off-spec but below whatever the device considers dangerous are generally well tolerated. USB HDDs drawing ~800ma, or connected to the data lines of one port and the power lines of a second port, little fans and LEDs on goosenecks, all kinds of nasty stuff. Especially on desktops, where +5v is available in nigh-unlimited quantities. Laptops and routers and things, with actual power budgets, seem to be a bit pickier.
"Negotiate" is a loose term - really it's just some fixed resistances across the data pins that set USB charging mode. This can be built into the plug without any extra copper in the cable.
Sometimes a device/cable combination can work in either way. They'll have some goofy resistances-between-pins coding scheme for dumb chargers; but they'll also do the official USB SIG power negotiation dance if plugged into an actual USB host. Of course, so long as you only want to charge the device, it just has to work, not necessarily work the best or correct way.
I suspect that if they thought they could get away with that, they would. Taking the 'not our fucking problem' position costs substantially less than hiring a contractor to make it your problem.
These are completely different architectures, and I highly doubt portability will be easier between the two, just because one makes the jump to being native 64bit.
Emulating an ARM core on x86 is totally doable (QEMU, in fact, does it for the Android SDK's test/emulator component. I don't think anybody has an OSS iPhone clone fully working; but that wouldn't take Apple long); what I find baffling is the theory that the application being emulated would need to be 64 bit to 'play nice' with the larger memory space available to the host.
Even if Apple decided to support cross-platform binaries(and that's one hell of an 'if', Apple Loathes half-assed ports, and has so far not embraced touchscreens in desktops and laptops, so cross-platform binary support would be asking for half-assed ports in both directions); allocating one or more 32-bit processes its own happy little address space has been trivial on 64-bit OSes for ages. Hell, even pre-64-bit, PAE systems were happily slicing large quantities of RAM into 32-bit address chunks to support 32-bit processes. Since Apple makes no ARM devices with 4GB+ of RAM(indeed, they tend to ship rather on the low side compared to the pricier Android OEMs), iOS apps aren't exactly going to be starving on the desktop, and a desktop OS has no trouble carving up a larger memory space into smaller chunks.
Oh, noting requires that the RNG be on-die (and, as you say, there are all kinds of options that aren't, the universe is a very noisy place). However, unless your computer is actually configured to use whatever access to the noise of the universe it has (nasty little webcams are another good source), and dump that entropy in whatever pool(s) your software environment specifies, it doesn't help you much.
The on-die ones are valued mostly because they are really, really, fast and a whole lot cheaper than the earlier generations of purpose-built crypto coprocessors, which were very much priced for a niche market.
If you can't trust them, though, you have options.
Are we going to have another decade+ of handwringing about what will happen to the poor, poor, old and/or poor people whose constitutional right to TV will be destroyed if they need to buy an ATSC tuner with newer codec support? Because damn was the NTSC->ATSC transition annoying...
"Yeah, if they got serious they could push for the byte encoding again, or even make it the default output. But then, you wouldn't get much benefit if you had to mix non-.NET code with it."
.NET 4 and some 16-bit version of Installshield or something just to copy it into place, so I don't doubt that there is a lot of 'mostly ready' code that has resulted in products that aren't going anywhere but x86 without one hell of a shove.
I have particularly unpleasant memories of dealing with some application whose developers had managed to make it require
I just find it somewhat ironic that Microsoft theoretically has one of the better technological positions to support a multi-architecture binary software ecosystem, and has been working on it for years, and yet managed to fumble so profoundly and make so little use of that.
"Heck what about random generator devices?"
The whole point of TFA is about a technique for (mostly undetectably) modifying a good hardware RNG and turning it into a really lousy one.
Getting your entropy from multiple places probably helps (if they don't know what 6 RNGs you chose it's harder to dope them all, and even if they do, they still have to slog through the entropy from multiple crippled sources rather than only a single one (and, while it is possible to cripple the RNG entirely, that will show up on tests, so plausible real-world implementations would still provide some entropy, just less than advertised).
Pedestrian's unwillingness to voluntarily surrender the contents of their pockets is the primary reason for so many of today's muggings.
Oh, I don't doubt that the TV broadcast standards people had their reasons, they might even have been good ones from the perspective of the awful legacy tech they were dealing with. I just bitterly resent how they managed to smear their shit all over the computer market. Had it stayed confined to broadcast television, they could have done whatever they liked for all I care.
'Wanna try some snow crash?'
"Or they could have tried to get Visual Studio to leverage LLVM and ship bitcode so things could be ever further future-proofed and extend to more than just 2 architectures."
That's the humorous part of all this: Microsoft started work (more than a decade ago, if I recall) on the 'Common Language Runtime' and the 'Common Language Infrastructure', with the 'Common Intermediate Language' playing the part of architecture-independent bytecode representation. It's ostensibly a standard and whatnot; but basically Microsoft's ".NET" is the serious implementation.
The already have, in house, widely used, supported by their dev tools, an architecture independent mechanism. Loads of ISVs even use it fairly extensively.
Architecturally, they might actually have the best position among any major vendor to make cross-platform binaries happen; but they threw it all away to try to have a mandatory app store. Elegant, really.
Exactly, that's the 'linux' angle on the 'Intel finally delivering chips that should finish nailing the coffin shut on Windows RT' story.
There obviously isn't much stopping Microsoft from having another try at iOS-envy and mandatory app stores on x86 (the implementation on ARM, using UEFI secureboot and a restrictive SRP in the windows image, would be 100% doable on x86); but as previously implemented, that was a major distinction: Wintels would be more or less as they have always been, Win-ARM would be walled garden dystopia. Hence the general enthusiasm for the demise of the same.
Thatcher said she would never privatise the post office.
And Reagan would be derisively referred to as a 'RHINO' if he were to attempt to gain office as a republican today... The acolytes tend to...get out of hand... in their veneration after the venerated has been dead or nonfuctional for a while.
Stop giving up your life to big business, and they'll stop being able to tear you a new one.
This isn't like that other article, where the British government is selling off a natural monopoly so you're forced to use a particular business. This is you thinking that you are entitled to get Zuckerberg to do anything more than widen the smile on his deservedly smug face.
Or, at least, chose a big business that doesn't have so much market power. There are plenty of suppliers who offer, y'know, actually-two-sided contracts, or (this one blows my mind; but it's true) actually run a business where they do better by not screwing over their customers. It's crazy. There are also suppliers who sell commodities, and the worst they can do is give their competitors' salesweasels a good day.
I'm not entirely taken with Nicholas Carr; but he has a useful little coinage to cover this situation: "Digital sharecropper".
It beats real sharecropping (sometimes you get air conditioning, and even paid in real money rather than scrip and debt peonage!); but if your business (or your hobby, though businesses tend to be more financially painful) depends on a third party, with which you have absolutely no leverage other than their power and mere pleasure, (and where your business consists largely of making their business incrementally more successful), you are a sharecropper. And, while the timing of the crackdown is sometimes rather baffling, since it doesn't even seem to be to the landlord's advantage, it is closer to being an inevitability than a mere possibility.
This doesn't mean that you have to do everything 100% alone in order to not be a sharecropper, commodities are safe enough, as are companies so mired in the demands of actually-powerful customers that they will have difficultly cutting the feet out from under you at a greater than glacial pace; but a situation where you are 100% dependent on a single third party who has the right, and the ability, to cut you down just by revoking an API key or deleting a page on their own servers? They own you.
Where is the linux angle? Not even a conversion coupon code?
It isn't mentioned; but 'Windows RT's fairly clear demise, in favor of cheap-ass x86 devices, is almost certainly good news for Linux(Not 'This is the year of Linux on the Desktop!!!' news; but good).
Per Microsoft's secure boot requirements, ARM-based 'Windows RT' hardware Must Not allow (either out of the factory, or by user modification) signing keys for boot payloads other than Microsoft's own and cannot allow disabling 'secure boot', while x86 Win8 devices can.
It remains to be seen how many will actually be purchased for linuxization; but Windows RT devices are (short of breaking TPM-backed UEFI secure boot) 100% useless for Linux, or anything else that isn't blessed by Redmond. Wintel tablets, though, are just funny shaped wintels, and so only the questionable state of Linux touch GUIs stands between you and installation.
I don't know whether Surface RT was a genuine fuckup, or whether it was Microsoft reminding Intel that they've ported the NT kernel before and can do it again in the hopes of spurring them to get their shit together on the low-TDP side; but either way RT looks dead, dead, dead. Under interpretation one, Microsoft gimped it hard enough, either to protect other parts of their business or to push 'winRT'(the runtime not the OS) that it was pretty sick already and Intel just shoved a knife in it's back. Under interpretation two, Intel appears to have risen to the challenge, or at least close enough that full binary compatibility with all things Windows will be worth more (to anybody considering a Microsoft product at all) than an extra sliver of battery life.
What I wouldn't give to be able to travel back in time and prevent 1366x768 or '720p' from being defined as 'HD' resolution. Ideally with some sort of plan that involves more explosions than a braindead summer action movie. What a pox upon the eyes of the world, especially with so many applications making poor use of extra horizontal space (so it's barely better than 1024x768, circa 15 years ago) and 768 pixels being pretty narrow for the 'well, just flip it 90 degrees' strategy that saves other widescreens for non-movie purposes.
You'd think the British people would have noticed by now.
St. Thatcher demands sacrifice and promises her faithful that the pain means it's working.
"You would think that the private sector could manage to do at least one thing better than the British government, wouldn't you?"
Apparently, they are a great deal better at recognizing which side the bread is buttered on in the 'privatization' deals than the British government, Does that count?
Bah, why 'say' "trigger warning" when we could just have an XML-based, machine readable, semantic tagging mechanism for trigger warnings? Clearly a superior solution.
The vermin only teaze and pinch
Their foes superior by an inch.
So, naturalists observe, a flea
Has smaller fleas that on him prey;
And these have smaller still to bite 'em,
And so proceed ad infinitum.