This is yet another piece of software which the user needs to download, enable installation of third-party apps, and install. Or the user might've installed it from a dodgy app store (in which case their device is likely already a teeming mess of malware).
Either way, the user needs to do something we've spent the last umpteen years trying to indoctrinate people against.
Wake me up when someone starts injecting this stuff through advertisements in web pages.
What if Google also opened a subsidiary in a country that banned garb like that required by Saudi Arabian law? How do they follow both laws at once across their entire organization?
Generally speaking, if the corporation can demonstrate that it's not possible to implement the law of all countries it operates in, and within any one country it *does* follow the laws of that country, that's good enough.
The problem is, if you actually read this particular ruling, Google didn't actually demonstrate that. Their argument was, in a nutshell, "we're a global operation, the database is hosted outside your country, so your order doesn't apply to anything except.ca".
Where the database is hosted doesn't matter because it's pretty much a given that google.ca and google.com and google.* are essentially the same place. If google.ca can have a result blocked, that same flag can be applied globally. So that argument doesn't really pass the giggle test.
The fact that Google (or Twitter, or Facebook, or...) normally only blocks certain content in the country that ordered the block is a corporate policy decision, not a technical or legal limitation. Child porn, for example, is something they can and will block globally. But corporate policy is not the law in any particular country.
So, that leaves the "we're a global operation" argument. And the judges response to that argument was, pretty much, "Good. Then you can block it globally, too."
In other words, someone made an assumption that all you have to do to stop a court ruling at the border is to point out that there's a border. The judge didn't buy it. Better luck in appeals court.
In other words, in this judge's opinion, since Google works on a global scale, they should be subject to the laws of all nations at once.
I think it's more that Google has a physical and legal presence in Canada (and many other countries to some degree or another), and hence a Canadian court order can apply to their entire business. The consequence of not following that order is that there are Google people in Canada who can be jailed for not following that order.
Companies who are only physically present in one country but have a web presence wouldn't be affected by court orders outside of their own country since that leverage doesn't exist, and many (if not most) foreign courts would likely toss a case due to lack of jurisdiction. In countries where a multi-national might have a small and/or expendable presence (i.e. a local shell company staffed by contractors), they could get away with ignoring a local court order with minimal risk (i.e. they might lose their.xx TLD).
No doubt Google Canada is set up as a separate subsidiary corporation and they think that there should be some sort of legal firewall at the political boundaries, but the funny thing about multi-national corporations is that these firewalls are always quite permeable when it comes to money flowing in a particular direction. So, why can't court orders follow the money?
And that do you really want to put your information into a service that may well go away, rather than trying to work out some way to use it locally.
I can live with the risk; I don't keep data online that I don't have local copies of or would be bothered if it went away.
I'd estimate that at least 50% of the services I use daily now won't be around in another 5-10 years, and 90% of the devices I use now won't be usable in another 5 years. That's the nature of modern technology, and it's not specific to Google.
Is there any way to retrieve my Google Health data from Google?
For approximately a 6 month period prior to the deletion date, people could download their data in a whole whack of formats. See the original announcement. Say what you will about Google, but they are good about giving lead time and tools to extract your information prior to a service shutdown.
That being said, being able to download your information and being able to do anything with it is a whole other thing. In the case of Google Reader, at least the output format was a standard format (OPML), so migrating elsewhere was relatively trivial (and Feedly pretty much turned it into a one-click experience). I'm not sure any competitors bothered to handle the Google Health formats.
Is this the kind of problem that cyanogenmod addresses?
With limits, yes.
CM's privacy guard allows you to block apps from getting at your address book or SMS and such. It also allows you to control things like camera/microphone access. And you can even disable background apps and notifications (for example, I have Facebook pretty much tuned so it can't do anything more than it can in a web browser).
One notable thing CM doesn't do is allow you to prevent Internet access for apps. I read that this is to prevent someone from downloading an add-supported app and then cutting it off from its ad networks. I order to do that sort of thing, you usually need to root and install a firewall or some other ad blocker.
Quite frankly, if you've got a phone that's out of warranty or no longer getting vendor updates, installing CM is worth looking into. It's a bit of a pain in the ass the first time (at least it was for my devices), but after that it's pretty smooth sailing.
Nothing. At this point it's got nothing to do with the capabilities of the aircraft.
I'm guessing that the people who originally pushed for the F-35 still haven't retired/been voted out, and hence any choice except the F-35 would mean they'd look like idiots.
When I worked in one inner suburb of a medium-sized city, and lived in another, I commuted about 50km each way, 100km in total, and hence 3000km over the course of a little over a month.
It's an add-on to a pure-electric car to extend the range. The Nissan Leaf, for example, is rated at at least 120km/charge. So, in theory you'd never actually draw on this magic battery for your daily driving. It'd only be if you had longer trips or weren't able to plug in one night, etc.
The average commute in North America is well within the range of a plug-in electric vehicle, and this thing is just icing/insurance. There's going to be outliers, but if we routinely killed ideas because they didn't work for 100% of possible scenarios, we'd still be shivering naked in caves (fur being too darned hot for those in warmer climates...)
You see, you're assuming that your usabilty concerns matter. It makes far more sense when you realize that it was all designed for Steve Jobs.
just TRY connecting the damn thing to a projector or second display in a conference room and making it behave in a rational manner.
Steve didn't need to do that stuff. He had a technical team to handle projection and other A/V needs.
Or try taking a screenshot... what was that obnoxious key combo again? That's right... it makes no sense and can't be remembered by a mere mortal.
Don't be silly. Screenshots are handled by the marketing department.
Let's jump to the beginning of a line with the Home key, or the end of the line with the End key... oh wait, it doesn't have one. They conveniently replaced those with more key combinations that can't be remembered by us mortals. Apparently text entry isn't an important usability case for Apple.
Well, it turns out that the dead don't need to do much text entry...
The rule is, you have the contract done and signed before the previous one expires.
I expect that there's also a very applicable rule about how any contract negotiation after getting caught in anti-trust collusion against one of the largest retailers of your goods will probably involve you making more concessions that you might like.
Any wise old vets want to chime in about when to use "goto" and when not to?
Not claiming to be a wise old vet, but as a general rule of thumb, the ideal way to use it is to emulate control structures from even higher level languages. For example, if in C++ you'd write an error handling behavior in a function as an exception, in C you might goto a common area to cleanup transient memory and gracefully return an error code. Which is exactly what that snippet you've shown is doing.
So from the other side, if an Afghani intelligence agency was recording every call in America, that's OK too because it's their job?
Under Afghani law, probably.
Granted, Afghan law has perhaps recently had a lot of outside fingers in it, so that might actually be illegal.
I don't doubt that it's legal for the NSA to be doing this under American law, seeing how foreign signals interception is largely their main function. With American troops in a foreign country with a history of militant extremist activity, it's pretty much a given that there's going to be signals interception, and with America being heavily involved in stuff like infrastructure rebuilding, there would have been plenty of opportunity to build in interception capabilities. I'd also assume Iraq is in a similar state.
Sure, all MHz are not created equal, but still, there's plenty of Windows software that was written in the 200MHz days that's still in use that I suspect would run just fine on my phone.
True, but you get into the x86-emulation-on-ARM problem. I'm sure you can build such an emulator, but I'm not sure what kind of speed.
That and I suspect a lot of the old software from that era that someone might want to run might be tied to hardware. Games targeting specific graphics cards, industrial software using parallel/serial ports, etc.
shrug I imagine it could work, but I don't see it driving enough sales to make it worth Microsoft's time.
It still amazes me that you can't just run normal Windows on the ARM-based surface.
I suspect it's doable, but performance wouldn't be good.
There's a definite horsepower difference between ARM and x86, and if the "normal Windows" config is tuned for x86, it'll (slowly) suck balls on ARM. They'd have to strip down services to make it work. Plus, people are going to expect x86 Windows apps to run (hence the marketing debacle of Windows RT), which isn't feasible on ARM.
It's not exactly scientific, but for comparison try out Android-x86 on older hardware. I have an old Acer AAO ZG5, and every time I've run Android on it, running off a USB key, the first thing I've noticed is that it smokes any ARM Android device I've ever tried. Keeping in mind that many of the apps I've been testing are native ARM builds being emulated.
Besides the privacy and safety concerns of these things, I was under the impression that a major flaw is that it's a bit too easy to sneak things through them.
Is it really a smart idea to move these things from a place where security is theatre to a place where the targets actually *are* sneaking weapons through security and using those to actually kill other people?
Anyone at NSA who is participating in this is committing an act of war against a sovereign nation without any declaration of war.
Actually, since it's interception of communications of a foreign country, this may be one of the few recently reported instances of the NSA actually doing their job.
... but that doesn't mean their lives should be ruined simply for some relatively minor past mistakes.
I agree, and that's essentially part of my point, but is this really the fault of the search engines, or is it the "no forgiveness" attitude of employers (and a whole host of others that are in a position to make major decisions that massively affect someone's life) that's the fundamental problem?
It's not just a criminal background problem, either. People are looking at stuff like Facebook photos of someone partying, or political affiliations, etc, and making decisions about employability.
This isn't anything really new, it's just easier to dig up dirt on someone, it's possible to do it on a wider scale, and the results tend to be more authoritative. But there's always been the concept of a "permanent record". That's where expressions like "reference checks", "background checks", "not burning your bridges", "skeletons in the closet", etc came from.
Having Google take down information isn't going to make that information go away; if anything it's creating a business case for a "reputation search engine" to set up shop in a jurisdiction which wouldn't allow this sort of thing.
Credit reference agencies have millions of people on their files and a legal obligation to make sure that the information is accurate, and to purge it when it is no longer legally relevant.
Credit reference agencies have meta-data about when they were given the information, who they got it from, and specific dollar amounts, and there's a whole legal framework about how reports are made.
And they *still* routinely screw it up.
Google has an indexing date, a URL, and a blob of content. There's a lot of other stuff they can extract/infer from the content and response headers and they can perform an estimate of relevance (PageRank), but accuracy is obviously inconsistent even without things like SEO's messing with results, and the correctness of any content is governed by stuff like libel and copyright laws in balance with civil rights like freedom of speech.
There's a reason Google's search results have been, in court of law, identified as "opinion" rather than "fact".
Isn't the general principle that once you've done the time, you've paid for your crime?
That's a lovely general principle. That's why we, as a society, are perfectly okay with hiring convicted pedophiles to teach kids, or having convicted armed robbers manage the cash in our ATMs?
I'm generally against using a "one-off" crime to punish someone in perpetuity and deny them a decent quality of life, particularly if it was of the "I was young, stupid and desperate" sort of thing. Shit happens. But a criminal history with high recidivism rates or for particularly heinous crimes can arguably have some predictive aspects.
Why would a pardon be any more relevant of a criteria than the successful completion of a prison sentence?
Because a pardon is, getting back on topic, a formal analysis of whether a criminal conviction is still relevant and/or outdated.
In a magical world where completion of a prison sentence implied rehabilitation, then yes, it would be roughly equivalent to a pardon. In the world where we live, successful completion of a prison sentence just means someone hasn't fucked up so badly that they're still in prison.
Of course, even pardons aren't perfect. Depending on jurisdiction, they may be nothing more than a receipt for a large bribe. But that's still a step up from "got out of prison".
Slashdot Mirror
Good point. Mind you, that kinda makes anyone who installs it even dumber than I would have thought.
Easy: Don't. Fucking. Install. It.
This is yet another piece of software which the user needs to download, enable installation of third-party apps, and install. Or the user might've installed it from a dodgy app store (in which case their device is likely already a teeming mess of malware).
Either way, the user needs to do something we've spent the last umpteen years trying to indoctrinate people against.
Wake me up when someone starts injecting this stuff through advertisements in web pages.
ARM emulation under x86 is feasible. Google "libhoudini". Whether it's simple is another question, but the pieces are there.
Generally speaking, if the corporation can demonstrate that it's not possible to implement the law of all countries it operates in, and within any one country it *does* follow the laws of that country, that's good enough.
The problem is, if you actually read this particular ruling, Google didn't actually demonstrate that. Their argument was, in a nutshell, "we're a global operation, the database is hosted outside your country, so your order doesn't apply to anything except .ca".
Where the database is hosted doesn't matter because it's pretty much a given that google.ca and google.com and google.* are essentially the same place. If google.ca can have a result blocked, that same flag can be applied globally. So that argument doesn't really pass the giggle test.
The fact that Google (or Twitter, or Facebook, or...) normally only blocks certain content in the country that ordered the block is a corporate policy decision, not a technical or legal limitation. Child porn, for example, is something they can and will block globally. But corporate policy is not the law in any particular country.
So, that leaves the "we're a global operation" argument. And the judges response to that argument was, pretty much, "Good. Then you can block it globally, too."
In other words, someone made an assumption that all you have to do to stop a court ruling at the border is to point out that there's a border. The judge didn't buy it. Better luck in appeals court.
I think it's more that Google has a physical and legal presence in Canada (and many other countries to some degree or another), and hence a Canadian court order can apply to their entire business. The consequence of not following that order is that there are Google people in Canada who can be jailed for not following that order.
Companies who are only physically present in one country but have a web presence wouldn't be affected by court orders outside of their own country since that leverage doesn't exist, and many (if not most) foreign courts would likely toss a case due to lack of jurisdiction. In countries where a multi-national might have a small and/or expendable presence (i.e. a local shell company staffed by contractors), they could get away with ignoring a local court order with minimal risk (i.e. they might lose their .xx TLD).
No doubt Google Canada is set up as a separate subsidiary corporation and they think that there should be some sort of legal firewall at the political boundaries, but the funny thing about multi-national corporations is that these firewalls are always quite permeable when it comes to money flowing in a particular direction. So, why can't court orders follow the money?
I can live with the risk; I don't keep data online that I don't have local copies of or would be bothered if it went away.
I'd estimate that at least 50% of the services I use daily now won't be around in another 5-10 years, and 90% of the devices I use now won't be usable in another 5 years. That's the nature of modern technology, and it's not specific to Google.
For approximately a 6 month period prior to the deletion date, people could download their data in a whole whack of formats. See the original announcement. Say what you will about Google, but they are good about giving lead time and tools to extract your information prior to a service shutdown.
That being said, being able to download your information and being able to do anything with it is a whole other thing. In the case of Google Reader, at least the output format was a standard format (OPML), so migrating elsewhere was relatively trivial (and Feedly pretty much turned it into a one-click experience). I'm not sure any competitors bothered to handle the Google Health formats.
With limits, yes.
CM's privacy guard allows you to block apps from getting at your address book or SMS and such. It also allows you to control things like camera/microphone access. And you can even disable background apps and notifications (for example, I have Facebook pretty much tuned so it can't do anything more than it can in a web browser).
One notable thing CM doesn't do is allow you to prevent Internet access for apps. I read that this is to prevent someone from downloading an add-supported app and then cutting it off from its ad networks. I order to do that sort of thing, you usually need to root and install a firewall or some other ad blocker.
Quite frankly, if you've got a phone that's out of warranty or no longer getting vendor updates, installing CM is worth looking into. It's a bit of a pain in the ass the first time (at least it was for my devices), but after that it's pretty smooth sailing.
Yes, but Congressman Mike Rogers said he was a Russian spy. Surely a US Congressman wouldn't lie about important stuff like that?
Nothing. At this point it's got nothing to do with the capabilities of the aircraft.
I'm guessing that the people who originally pushed for the F-35 still haven't retired/been voted out, and hence any choice except the F-35 would mean they'd look like idiots.
It's an add-on to a pure-electric car to extend the range. The Nissan Leaf, for example, is rated at at least 120km/charge. So, in theory you'd never actually draw on this magic battery for your daily driving. It'd only be if you had longer trips or weren't able to plug in one night, etc.
The average commute in North America is well within the range of a plug-in electric vehicle, and this thing is just icing/insurance. There's going to be outliers, but if we routinely killed ideas because they didn't work for 100% of possible scenarios, we'd still be shivering naked in caves (fur being too darned hot for those in warmer climates...)
You see, you're assuming that your usabilty concerns matter. It makes far more sense when you realize that it was all designed for Steve Jobs.
Steve didn't need to do that stuff. He had a technical team to handle projection and other A/V needs.
Don't be silly. Screenshots are handled by the marketing department.
Well, it turns out that the dead don't need to do much text entry...
When you're talking about packet protocols, where things get lost, duplicated, or reordered, "previous" is an incredibly imprecise word.
I expect that there's also a very applicable rule about how any contract negotiation after getting caught in anti-trust collusion against one of the largest retailers of your goods will probably involve you making more concessions that you might like.
Not claiming to be a wise old vet, but as a general rule of thumb, the ideal way to use it is to emulate control structures from even higher level languages. For example, if in C++ you'd write an error handling behavior in a function as an exception, in C you might goto a common area to cleanup transient memory and gracefully return an error code. Which is exactly what that snippet you've shown is doing.
Let's not push our luck.
Besides, I was thinking Ford should tag along with Rodman on his next jaunt to North Korea...
As a Canadian, I second this proposal.
Under Afghani law, probably.
Granted, Afghan law has perhaps recently had a lot of outside fingers in it, so that might actually be illegal.
I don't doubt that it's legal for the NSA to be doing this under American law, seeing how foreign signals interception is largely their main function. With American troops in a foreign country with a history of militant extremist activity, it's pretty much a given that there's going to be signals interception, and with America being heavily involved in stuff like infrastructure rebuilding, there would have been plenty of opportunity to build in interception capabilities. I'd also assume Iraq is in a similar state.
True, but you get into the x86-emulation-on-ARM problem. I'm sure you can build such an emulator, but I'm not sure what kind of speed.
That and I suspect a lot of the old software from that era that someone might want to run might be tied to hardware. Games targeting specific graphics cards, industrial software using parallel/serial ports, etc.
shrug I imagine it could work, but I don't see it driving enough sales to make it worth Microsoft's time.
I suspect it's doable, but performance wouldn't be good.
There's a definite horsepower difference between ARM and x86, and if the "normal Windows" config is tuned for x86, it'll (slowly) suck balls on ARM. They'd have to strip down services to make it work. Plus, people are going to expect x86 Windows apps to run (hence the marketing debacle of Windows RT), which isn't feasible on ARM.
It's not exactly scientific, but for comparison try out Android-x86 on older hardware. I have an old Acer AAO ZG5, and every time I've run Android on it, running off a USB key, the first thing I've noticed is that it smokes any ARM Android device I've ever tried. Keeping in mind that many of the apps I've been testing are native ARM builds being emulated.
Besides the privacy and safety concerns of these things, I was under the impression that a major flaw is that it's a bit too easy to sneak things through them.
Is it really a smart idea to move these things from a place where security is theatre to a place where the targets actually *are* sneaking weapons through security and using those to actually kill other people?
Actually, since it's interception of communications of a foreign country, this may be one of the few recently reported instances of the NSA actually doing their job.
I agree, and that's essentially part of my point, but is this really the fault of the search engines, or is it the "no forgiveness" attitude of employers (and a whole host of others that are in a position to make major decisions that massively affect someone's life) that's the fundamental problem?
It's not just a criminal background problem, either. People are looking at stuff like Facebook photos of someone partying, or political affiliations, etc, and making decisions about employability.
This isn't anything really new, it's just easier to dig up dirt on someone, it's possible to do it on a wider scale, and the results tend to be more authoritative. But there's always been the concept of a "permanent record". That's where expressions like "reference checks", "background checks", "not burning your bridges", "skeletons in the closet", etc came from.
Having Google take down information isn't going to make that information go away; if anything it's creating a business case for a "reputation search engine" to set up shop in a jurisdiction which wouldn't allow this sort of thing.
It's a huge can of worms.
Credit reference agencies have meta-data about when they were given the information, who they got it from, and specific dollar amounts, and there's a whole legal framework about how reports are made.
And they *still* routinely screw it up.
Google has an indexing date, a URL, and a blob of content. There's a lot of other stuff they can extract/infer from the content and response headers and they can perform an estimate of relevance (PageRank), but accuracy is obviously inconsistent even without things like SEO's messing with results, and the correctness of any content is governed by stuff like libel and copyright laws in balance with civil rights like freedom of speech.
There's a reason Google's search results have been, in court of law, identified as "opinion" rather than "fact".
That's a lovely general principle. That's why we, as a society, are perfectly okay with hiring convicted pedophiles to teach kids, or having convicted armed robbers manage the cash in our ATMs?
I'm generally against using a "one-off" crime to punish someone in perpetuity and deny them a decent quality of life, particularly if it was of the "I was young, stupid and desperate" sort of thing. Shit happens. But a criminal history with high recidivism rates or for particularly heinous crimes can arguably have some predictive aspects.
Because a pardon is, getting back on topic, a formal analysis of whether a criminal conviction is still relevant and/or outdated.
In a magical world where completion of a prison sentence implied rehabilitation, then yes, it would be roughly equivalent to a pardon. In the world where we live, successful completion of a prison sentence just means someone hasn't fucked up so badly that they're still in prison.
Of course, even pardons aren't perfect. Depending on jurisdiction, they may be nothing more than a receipt for a large bribe. But that's still a step up from "got out of prison".