It's roughly 50% Windows admins whose GPO-fu is weak bitching because their registry hacks don't work anymore, and 50% Chromium developers telling them that, yes, sorry, Chrome queries group policy state directly, only falling back to the registry under specific conditions (and noting that they make no assurance that that fallback will continue in the future.)
If that isn't enough, try a look at 'policy_loader_win.h'. It's fairly clear about reading the registry, rather than grovelling through the policies directly, is a fallback behavior that occurs only if grovelling through the policies doesn't work out.
Oh, I definitely wouldn't be betting against them on that (and, unfortunately, if I were betting against them, it'd be because I was betting on one of their overtly hard lock-in competitors, like Apple.)
It is nice that Google doesn't really ever stop you (and in our delightful world of DRM and 'licensed-not-sold' many vendors are not so polite); but practically everything they do is a softish sell of every part of their ecosystem that they haven't sucked you into yet.
Perhaps not... Apparently, I'm not too deeply versed in Windows technical arcana, but apparently some applications are configured by group policy indirectly (group policy changes the registry in the appropriate places, the application reads the registry in those places and acts accordingly) and some applications use an OS-provided API to query the applied group policies directly and apply those(exactly where the policies are stored in this instance is not clear to me. The domain controller presumably has them, in a networked environment; but where on the client they persist if it goes off the grid for a bit, or in the case of a locally managed machine, I don't know).
As of version 28, (according to the thread in the above link, because of malware leading Chrome around by the nose through modifications of the registry entries), Chrome switched from the registry-based mechanism to the new mechanism. Modifications to the registry are now ignored, and only policies applied by the OS-supported group policy manipulation mechanisms will apply.
I assume that these policies have to live somewhere, and thus can be edited (even if the OS protects them hard enough that you need a hex editor and a liveCD to do it); but it won't be a simple regedit.exe job.
I suspect that the problem is the (which would have companies w/ serious $ behind it) part.
There is at least one fairly serious Android fork, which doesn't really have a name for marketing purposes; but is what all the Kindle tablet devices run. It's arguably even more dystopian than Google's version, though I don't know how competently it locks down the bootloader and keeps you from fleeing entirely.
In the same general vein, the AOSP 'clean' version of Android isn't an FSF-purist-dream; but only because of all the apache components, it's safe enough. It's also pretty spartan. Most of the good stuff on an Android device either is, or depends on, Google Android apps or APIs they provide. Every time Google bumps a version, it's typical to see another bit or piece of the 'Google Android' build drop an AOSP-provided component in favor of a Google one. The old code still exists, and you can still do whatever you want with it; but that's a good sign that it isn't going to be seeing much engineering time from Google in the future.
Arguably, Google's strategy appears to be more of an attempt at bundling (which, strategically, does require walled-gardening some components of the bundle, since using pricing-based bundling strategies on free stuff doesn't really work) than a pure walled-garden play. Use your strengths (eg. how many companies, aside from perhaps Nokia, who actually has a mapping division; but is a thrall of MS now, would dare launch an 'android' handset without Google maps and Gmail?) to push on areas where you are relatively weak (Against Apple, Google can't afford to have 'android' be a synonym for 'cheap shit packed with more crapware than a stereotypical AOL-user's emachine')
Their strategy of remaining 'open' (ie. Android is open source, Chrome is based on Chromium, Chromebook is just Linux with a few binaries on top, etc.); but ensuring that the only really frictionless way to use their stuff is by playing by their rules, with your Google Account, in the Google Play Store, etc. seems to be the implementation of this.
If that were the motivation, would they not also do this on the Mac?
This is about _fucking_ annoying windows malware repeatedly reinstalling chrome extensions.
The fact that they are not breaking the capability on 'enterprise' policy installs suggests the same.
Incidentally, even if you aren't on a domain, this should mean that it isn't exactly rocket surgery to install the 'blocked' Chrome extensions.
Winkey+r, gpedit.msc, import the chromium policy templates, modify 'ExtensionInstallForcelist' to taste. Game over.
A certified letter from the IRS would be polite. A DEA SWAT team who assumes that your mystery-money is a sign of drug dealing... Less so. Be sure that your dog isn't home at the time and that there are no flammible family members who might experience adverse effects is somebody threw a flashbang too close to them.
Oh, definitely. And that would be exactly the point of the strategy I was hypothesizing about: Had IA64 actually worked, Intel would have needed to keep the FTC off their backs; but would want to scoop up as much cash as possible.
Having AMD 'supporting' systems with large amounts of RAM would be a useful talking point when dealing with the regulators; but the practical suckitude of such support would be a useful selling point when justifying rather alarming sticker prices. Intel would have wanted to avoid a second-source for IA64 parts, even 'almost as good, really, please? Buy some?' ones; but the FTC would have likely been called into action if your only options for more than 4GB of RAM were either Intel, or something even more expensive with mainframeish features from IBM.
All largely irrelevant (except that a lot of common Linux distros are defaulting to PAE kernels, rendering obsolete some otherwise still useful Pentium M based hardware), since IA64 mostly flopped and AMD64 and we-aren't-bitter-at-all 'EMT64' have totally crushed any reason to use weirdo hacks to support extra RAM.
Incidentally, I bet it would be cheaper to buy a law declaring people who sell exploits on the black market to be criminally responsible, as 'conspirators', for any and all subsequent use of them, thus encouraging people to remain in our sharecropper bounty system, than it would be to actually pay the workers more...
Strictly speaking, unless the bounties get substantially bigger than the minimum, and relatively quickly, it's more along the lines of 'If you can't beat them, see if you can provide additional motivation to people already on your side; but perhaps not bothering to focus on the problems you care about."
when HP/UX x86-64 port is out, then Itanium2 will be dead
Which happens first: HP/UX gets ported to x86, or Xeons pull far enough ahead of now-stagnant Itanium2s that somebody can release a (probably very expensive) Itanium VM that runs at adequate speed?
I don't doubt that it'll cost you, plenty, to have Intel not laser off their bits and HP not gimp their half of things in firmware; but my understanding is that Intel has every interest in looting Itanium's corpse for the interesting bits and offering those on (high end) Xeon SKUs.
HP's secret, secondary, mission statement requires them to acquire and kill good technologies. It's the rules. In rare cases of dire necessity, they are allowed to spin them off; but that is frowned upon.
Hard to say: having a nice 64-bit address space beats the hell out of PAE; but '32-bit = 4GB of RAM or less' was always only a Windows desktop thing. Server and workstation had PAE support. Had Intel wanted a gimp around, they might have been able to get away with AMD flogging PAE gear...
Itanium has long been a signal, a signal that says "I really do cut POs that large. Soak me."
Now that it's niche and dying, the fun will really start. Sure, the number of customers will shrink; but the last ones to go will be the ones who will pay almost anything for one last hit before they have to port...
Nobody supports 'Linux on ARM based servers' at the size you appear to be thinking of, because there aren't any, more or less.
Now, at the scale where actually-available-now-at-reasonable-prices-and-with-suitable-peripherals ARM cores are available, everybody supports Linux on ARM based servers, they're just called 'NAS'es, because they are small and feeble and typically just configured for light file-serving duty.
Great idea. We could also replace the Olympic games by much faster/higher/longer/more accurate machines.
No, the Olympics we should keep organic. Not necessarily entirely human; but organic. Just imagine hideous man/tick hybrids sprinting and jumping, ghastly quadruped-thing endurance runners, archers with creepy compound eyes... A glorious celebration of mostly-human athleticism!
The extant populations aren't original; but there must have existed a tribe (we'll call them 'Squish-ee Zero') who came first, and thus squished not, until they were squished. In some cases, there may even have been more than one squishless squish-ee (if disease, starvation, cold, or other natural misadventure cleared the stage of tribe N before tribe N+1 showed up); but that isn't logically necessary, just a contingent possibility.
I'd hope that chess cultivates a certain decorum; but given the... atmosphere... at a variety of other nerd-heavy and largely male events, it wouldn't be a huge surprise if they have some of the same appeal that the Japanese women-only subway cars do...
I can see the emotional appeal; but it's a trifle odd to even bother quibbling about the 'bestness metric' when storage and retrieval of every move, in every (reasonably official) game played in somebody's entire career is not exactly a terrifying challenge at the cutting edge of database design...
Especially with the (relative) standardization of computer-readable move notation, you could probably derive practically any wacky fitness metric you could conceive of, compute it, and rank players according to it, as suits your taste. Obviously, only a relatively small subset of possible fitness metrics would be of human interest; but 'best' seems substantially more reductive than required.
Logic has its uses; but why would you expect performance in a deterministic game of perfect information (and only a moderately complex one, at that: this isn't tic-tac-toe; but even some games that humans actually play for fun, like, Go, are considerably nastier, and never mind the ghastly stuff that mathematicians come up with for fun; by generalizing games into N dimensions or building rulesets for use on boards of peculiar sizes and not-fully-possible-in-three-dimensions structures...) to be particularly relevant to performance in the world, which shows...limited... determinism at useful scales, and certainly isn't a game of perfect information?
Shouldn't we just have the chess-playing computers in a 24/7/365 competition cycle, with mastery changing hands as often as is computationally feasible, and then just use the time-honored traditions to decide who among us shall bear the title of 'Meatsack Prime' within the chess world?
Chess does not need to be a show financed by big money. All sports where money got injected tend to turn into reality shows with media buzzing around searching for dirty stories, with pervasive doping, etc.
In that case, give chess one hell of a cash bolus, stat!
Wouldn't having the same care and attention devoted to developing exotic new ways to juice mental performance that sports doping has to enhancing various aspects of physical performance be an amazing boon to humanity?
Right now, the situation is pretty bad. We have a few stimulants and alertness aids, sometimes enough to get ADD Billy to do his homework; but nothing compared to what a suitably enthusiastic athlete can do to muscular performance... Just imagine: A world where the truism 'You can't fix stupid' has been falsified.
Bring on the sponsorships, sign the high-stakes exclusive television coverage contracts, let's roll!
Indians are the only true Americans. Every one else is an immigrant. And if republicans had their way would be kicked out ex post facto haste.
Archeological for any human inhabitation of the Americas is pretty recent. (Some) indian populations did have the distinction of being the only immigrants who didn't need to squish the locals; but all available evidence suggests close to zero chance that any human population originated here.
Arguably, we have a certain talent for importing talent... Scoring all the Jewish physicists when the Nazis drove them out, in order to build a bomb, and then scoring all the Nazi rocket scientists when the Soviets drove them out, in order to build something to deliver it with...
Slashdot Mirror
Did you read the thread I linked to?
It's roughly 50% Windows admins whose GPO-fu is weak bitching because their registry hacks don't work anymore, and 50% Chromium developers telling them that, yes, sorry, Chrome queries group policy state directly, only falling back to the registry under specific conditions (and noting that they make no assurance that that fallback will continue in the future.)
If that isn't enough, try a look at 'policy_loader_win.h'. It's fairly clear about reading the registry, rather than grovelling through the policies directly, is a fallback behavior that occurs only if grovelling through the policies doesn't work out.
Oh, I definitely wouldn't be betting against them on that (and, unfortunately, if I were betting against them, it'd be because I was betting on one of their overtly hard lock-in competitors, like Apple.)
It is nice that Google doesn't really ever stop you (and in our delightful world of DRM and 'licensed-not-sold' many vendors are not so polite); but practically everything they do is a softish sell of every part of their ecosystem that they haven't sucked you into yet.
Perhaps not... Apparently, I'm not too deeply versed in Windows technical arcana, but apparently some applications are configured by group policy indirectly (group policy changes the registry in the appropriate places, the application reads the registry in those places and acts accordingly) and some applications use an OS-provided API to query the applied group policies directly and apply those(exactly where the policies are stored in this instance is not clear to me. The domain controller presumably has them, in a networked environment; but where on the client they persist if it goes off the grid for a bit, or in the case of a locally managed machine, I don't know).
As of version 28, (according to the thread in the above link, because of malware leading Chrome around by the nose through modifications of the registry entries), Chrome switched from the registry-based mechanism to the new mechanism. Modifications to the registry are now ignored, and only policies applied by the OS-supported group policy manipulation mechanisms will apply.
I assume that these policies have to live somewhere, and thus can be edited (even if the OS protects them hard enough that you need a hex editor and a liveCD to do it); but it won't be a simple regedit.exe job.
I suspect that the problem is the (which would have companies w/ serious $ behind it) part.
There is at least one fairly serious Android fork, which doesn't really have a name for marketing purposes; but is what all the Kindle tablet devices run. It's arguably even more dystopian than Google's version, though I don't know how competently it locks down the bootloader and keeps you from fleeing entirely.
In the same general vein, the AOSP 'clean' version of Android isn't an FSF-purist-dream; but only because of all the apache components, it's safe enough. It's also pretty spartan. Most of the good stuff on an Android device either is, or depends on, Google Android apps or APIs they provide. Every time Google bumps a version, it's typical to see another bit or piece of the 'Google Android' build drop an AOSP-provided component in favor of a Google one. The old code still exists, and you can still do whatever you want with it; but that's a good sign that it isn't going to be seeing much engineering time from Google in the future.
Arguably, Google's strategy appears to be more of an attempt at bundling (which, strategically, does require walled-gardening some components of the bundle, since using pricing-based bundling strategies on free stuff doesn't really work) than a pure walled-garden play. Use your strengths (eg. how many companies, aside from perhaps Nokia, who actually has a mapping division; but is a thrall of MS now, would dare launch an 'android' handset without Google maps and Gmail?) to push on areas where you are relatively weak (Against Apple, Google can't afford to have 'android' be a synonym for 'cheap shit packed with more crapware than a stereotypical AOL-user's emachine')
Their strategy of remaining 'open' (ie. Android is open source, Chrome is based on Chromium, Chromebook is just Linux with a few binaries on top, etc.); but ensuring that the only really frictionless way to use their stuff is by playing by their rules, with your Google Account, in the Google Play Store, etc. seems to be the implementation of this.
If that were the motivation, would they not also do this on the Mac?
This is about _fucking_ annoying windows malware repeatedly reinstalling chrome extensions.
The fact that they are not breaking the capability on 'enterprise' policy installs suggests the same.
Incidentally, even if you aren't on a domain, this should mean that it isn't exactly rocket surgery to install the 'blocked' Chrome extensions. Winkey+r, gpedit.msc, import the chromium policy templates, modify 'ExtensionInstallForcelist' to taste. Game over.
A certified letter from the IRS would be polite. A DEA SWAT team who assumes that your mystery-money is a sign of drug dealing... Less so. Be sure that your dog isn't home at the time and that there are no flammible family members who might experience adverse effects is somebody threw a flashbang too close to them.
Oh, definitely. And that would be exactly the point of the strategy I was hypothesizing about: Had IA64 actually worked, Intel would have needed to keep the FTC off their backs; but would want to scoop up as much cash as possible.
Having AMD 'supporting' systems with large amounts of RAM would be a useful talking point when dealing with the regulators; but the practical suckitude of such support would be a useful selling point when justifying rather alarming sticker prices. Intel would have wanted to avoid a second-source for IA64 parts, even 'almost as good, really, please? Buy some?' ones; but the FTC would have likely been called into action if your only options for more than 4GB of RAM were either Intel, or something even more expensive with mainframeish features from IBM.
All largely irrelevant (except that a lot of common Linux distros are defaulting to PAE kernels, rendering obsolete some otherwise still useful Pentium M based hardware), since IA64 mostly flopped and AMD64 and we-aren't-bitter-at-all 'EMT64' have totally crushed any reason to use weirdo hacks to support extra RAM.
Hmm... You have a point there.
Incidentally, I bet it would be cheaper to buy a law declaring people who sell exploits on the black market to be criminally responsible, as 'conspirators', for any and all subsequent use of them, thus encouraging people to remain in our sharecropper bounty system, than it would be to actually pay the workers more...
.. bribe them.
Strictly speaking, unless the bounties get substantially bigger than the minimum, and relatively quickly, it's more along the lines of 'If you can't beat them, see if you can provide additional motivation to people already on your side; but perhaps not bothering to focus on the problems you care about."
I suspect that the crowd would love it, and the moralists would love scolding it and being horrified, so everybody would be happy!
when HP/UX x86-64 port is out, then Itanium2 will be dead
Which happens first: HP/UX gets ported to x86, or Xeons pull far enough ahead of now-stagnant Itanium2s that somebody can release a (probably very expensive) Itanium VM that runs at adequate speed?
I don't doubt that it'll cost you, plenty, to have Intel not laser off their bits and HP not gimp their half of things in firmware; but my understanding is that Intel has every interest in looting Itanium's corpse for the interesting bits and offering those on (high end) Xeon SKUs.
HP's secret, secondary, mission statement requires them to acquire and kill good technologies. It's the rules. In rare cases of dire necessity, they are allowed to spin them off; but that is frowned upon.
Hard to say: having a nice 64-bit address space beats the hell out of PAE; but '32-bit = 4GB of RAM or less' was always only a Windows desktop thing. Server and workstation had PAE support. Had Intel wanted a gimp around, they might have been able to get away with AMD flogging PAE gear...
Itanium has long been a signal, a signal that says "I really do cut POs that large. Soak me."
Now that it's niche and dying, the fun will really start. Sure, the number of customers will shrink; but the last ones to go will be the ones who will pay almost anything for one last hit before they have to port...
Nobody supports 'Linux on ARM based servers' at the size you appear to be thinking of, because there aren't any, more or less.
Now, at the scale where actually-available-now-at-reasonable-prices-and-with-suitable-peripherals ARM cores are available, everybody supports Linux on ARM based servers, they're just called 'NAS'es, because they are small and feeble and typically just configured for light file-serving duty.
Great idea. We could also replace the Olympic games by much faster/higher/longer/more accurate machines.
No, the Olympics we should keep organic. Not necessarily entirely human; but organic. Just imagine hideous man/tick hybrids sprinting and jumping, ghastly quadruped-thing endurance runners, archers with creepy compound eyes... A glorious celebration of mostly-human athleticism!
The extant populations aren't original; but there must have existed a tribe (we'll call them 'Squish-ee Zero') who came first, and thus squished not, until they were squished. In some cases, there may even have been more than one squishless squish-ee (if disease, starvation, cold, or other natural misadventure cleared the stage of tribe N before tribe N+1 showed up); but that isn't logically necessary, just a contingent possibility.
I'd hope that chess cultivates a certain decorum; but given the... atmosphere... at a variety of other nerd-heavy and largely male events, it wouldn't be a huge surprise if they have some of the same appeal that the Japanese women-only subway cars do...
I can see the emotional appeal; but it's a trifle odd to even bother quibbling about the 'bestness metric' when storage and retrieval of every move, in every (reasonably official) game played in somebody's entire career is not exactly a terrifying challenge at the cutting edge of database design...
Especially with the (relative) standardization of computer-readable move notation, you could probably derive practically any wacky fitness metric you could conceive of, compute it, and rank players according to it, as suits your taste. Obviously, only a relatively small subset of possible fitness metrics would be of human interest; but 'best' seems substantially more reductive than required.
Logic has its uses; but why would you expect performance in a deterministic game of perfect information (and only a moderately complex one, at that: this isn't tic-tac-toe; but even some games that humans actually play for fun, like, Go, are considerably nastier, and never mind the ghastly stuff that mathematicians come up with for fun; by generalizing games into N dimensions or building rulesets for use on boards of peculiar sizes and not-fully-possible-in-three-dimensions structures...) to be particularly relevant to performance in the world, which shows...limited... determinism at useful scales, and certainly isn't a game of perfect information?
Shouldn't we just have the chess-playing computers in a 24/7/365 competition cycle, with mastery changing hands as often as is computationally feasible, and then just use the time-honored traditions to decide who among us shall bear the title of 'Meatsack Prime' within the chess world?
Chess does not need to be a show financed by big money. All sports where money got injected tend to turn into reality shows with media buzzing around searching for dirty stories, with pervasive doping, etc.
In that case, give chess one hell of a cash bolus, stat!
Wouldn't having the same care and attention devoted to developing exotic new ways to juice mental performance that sports doping has to enhancing various aspects of physical performance be an amazing boon to humanity?
Right now, the situation is pretty bad. We have a few stimulants and alertness aids, sometimes enough to get ADD Billy to do his homework; but nothing compared to what a suitably enthusiastic athlete can do to muscular performance... Just imagine: A world where the truism 'You can't fix stupid' has been falsified.
Bring on the sponsorships, sign the high-stakes exclusive television coverage contracts, let's roll!
Indians are the only true Americans. Every one else is an immigrant. And if republicans had their way would be kicked out ex post facto haste.
Archeological for any human inhabitation of the Americas is pretty recent. (Some) indian populations did have the distinction of being the only immigrants who didn't need to squish the locals; but all available evidence suggests close to zero chance that any human population originated here.
we've always brought in big talent from elsewhere
Einstein 'nuff said
Arguably, we have a certain talent for importing talent... Scoring all the Jewish physicists when the Nazis drove them out, in order to build a bomb, and then scoring all the Nazi rocket scientists when the Soviets drove them out, in order to build something to deliver it with...
Playing both ends against everybody, awww yeah...