Racial profiling (non-jews and especially arabs are singled for more thorough examination). This is considered perfectly normal in Israel, but is not legal in US or EU.
Probably not terribly effective either. As far as I can tell the only reason it works at all is not because the Jewish population doesn't contain terrorists but because that they're reasonably sure Israeli Jews will only attack politically acceptable targets. Even that only works due to an atmosphere of imminent danger fed by the media that aims lone-wolf nutters safely at Arabs and Arab-sympathisers.
If companies don't need to hire humans, then where are they spending all of they money in production?
Raw materials, capital (equipment costs, buildings and depreciation), and land. If companies don't need humans anymore, those are the remaining scarce resources, so it's fairly obvious that's where the money will go. The trouble is that pretty much all the sources of raw materials, the majority of the land, and the money required for initial capital costs is owned and controlled by a small proportion of super-rich individuals, so they're the only ones that actually make money.
Ironically, capitalism has the same flaw that did communism in: it gives too much power to the individuals that control the allocation of resources.
If their production costs are very low, then their unit prices (will eventually) be very low. That's a good thing!
Low, low prices still don't help if your income is a big fat zero. Their production costs won't be that low either, because there will still be competition for the resources that are still scarce and in demand - it's just the vast majority of the population won't benefit from this.
They've had a btrfsck for a long time, it's just not actually capable of detecting or fixing most forms of filesystem corruption yet. (I believe the original version literally did nothing.)
As opposed to saying something trite and misleading, like you did? Microsoft wasn't just "using XML to do something XML was designed to do". The patent is fairly specific: it doesn't just require the use of XML, to infringe you have to have one file or piece of data containing sections of text, then a second document file that includes pieces of text from the first file identified by their location in said first file and also applies formatting to them. The reason Microsoft got caught out is that their code was an intentional clone of an i4i product; everyone else tends to use different methods to achieve the same aims that happen to have lots of prior art.
Now, if you want actual patent abuse, take a look at Microsoft's attempt to patent all use of XML files to describe office documents a while ago. (Thankfully that was restricted to only cover their OOXML format by the time it was granted. ODF is actually surprisingly patent-proof thanks to the amount of really old SGML stuff that's described in published papers.)
No inventing necessary. The Microsoft exec in charge of this scheme was practically gloating in internal e-mails about how integrating the functionality of i4i's main product into Word would drive i4i out of business. Not to mention that Microsoft have a long history of doing exactly that...
Well, i4i does at least make something. They now sell a utility that undoes the damage caused to documents by Microsoft's forced removal of their custom XML feature.
Yes, they sell a utility providing very useful (and patented) functionality to Word users - just like they did when Microsoft blatently cloned an earlier version of said utility, called it custom XML and integrated it into Microsoft Office. The court case even turned up internal Microsoft memos gloating about this and about how they would drive i4i oout of business by integrating the functionality of their main product into Office.
It should be fairly easy to tell that this has happened, because (a) the amount of data after decompression will equal the length given in the header and (b) you can detect end-of-file fairly robustly just using the format of the gzipped data anyway.
OK that's exactly what I'm looking for. So its no worse than dropping back to "one factor".
Unfortunately, RSA have tended to encourage customers to use a 4 digit numeric password with SecurID in the past because the security of the dongle supposedly made this safe, and this is the single factor that their level of security has now been reduced too. Hence why they were so keen to encourage the use of strong passwords by their customers when they first discovered they'd been hacked.
That's not terribly difficult to manage if you manage to obtain your fob's seed somehow; the algorithm has been public for a while. Of course, for security reasons they're designed to prevent you from doing that.
With RSA doing the keyfill at point of manufacture, the customer just needs to load the seed file for the entire batch onto their authentication server and then hand out the tokens, which are glued shut with considerable enthusiasm, and have no externally accessible electrical connections of any sort.
They don't do that though. As far as anyone outside RSA can tell, the keyfill port is a row of 7 PCB contacts hidden behind a little rectangular stick-on plastic cover on the back of the device. The cover doesn't even seem to be tamper-evident let alone tamper-resistant - you just press it back down again after you're done gawking and it sticks right back into place.
Off hand, I seem to recall this is true of the equivalent iPhone feature (apparently it's a bit hairy to do in practice too), their new "faster shutdown" feature on Mac OS X (requires apps to do the hard work of tracking whether they can be killed safely without losing data, including within any libraries they use), Grand Central Dispatch (mostly just a fairly standard thread pool, and yet Apple get all the credit for faster third-party app responsiveness when the app developers did all the hard work), and probably more stuff I haven't looked into.
You do realize that Apple released the code for FaceTime for others to build it into their clients.
They said they were going to and got a lot of good PR about it, then did absolutely nothing. As far as anyone can tell it's not even really practical to reverse engineer...
Errrrm... probably on one of the sites that saw the announcement at WWDC at a guess. For example, maybe Engadget (if you scroll down a bit on their liveblog to around 1:37PM, you can even see the big shiny "Only in the App Store" slide).
You'd think so, but Apple really don't seem to care too much about the corporate market. The OS X Lion Server add-on is apparently only available from the App Store too for what it's worth.
But I think what you say about Lion is incorrect. "Mac OS X Lion's new Resume feature lets users get back to where they left off after a shutdown or restart"
And based on how Apple's previous features have played out, I suspect they'll probably push all the hard work of actually implementing this onto the application developers...
That's not what Engadget seem to think. Of course, when Apple originally announced OS X Lion was going to be available from the App Store they said it was in addition to existing retail channels, but you know what Apple are like for changing their mind...
Hard questions like "is what the US government's doing to Bradley Manning torture", or "hard" questions like "is Bradley Manning a queer pinko faggot who obviously just leaked the info because he's nuts"? I'm guessing the latter... it's what most of the press has been doing.
Really, more systems should make use of the various encrypted key exchange schemes. They fairly strongly guarantee that you can only get one guess at the password per attempted login, even if you manage to intercept the communication beween server and client. (Obviously there's not a lot they can do about brute force attacks if you manage to acquire the information the server uses to verify the passwords, though.)
Not only that, but this is actually saying that the US embassy's contact didn't see any bloodshed in the square at the time, not that there wasn't any. All of the leaked cables have similar caveats attached.
AMD/ATI support is fine now, but with NVIDIA you're basically guaranteed to stay supported for a long time, since there's quite a large community programming with CUDA on Linux.
That depends - do you mean support for new cards as they're released, or continued support for existing cards? NVidia are very good at supporting newer cards, but some of their older cards aren't really usable under Linux anymore, whereas it looks like older ATI cards will remain supported for a long time...
Well sure, if by 'eating the poor kids' you mean my proposal of having the market forces directing them to various apprenticeships and whatever levels of education they could pay for, though this would require dramatic reduction in the costs
There's pretty much no such thing as apprenticeships anymore - the free market decided it was cheaper to employ a worker that someone else trained than to train them up yourself, and everyone stopped offering them. (Besides, most of the jobs that had apprenticeships aren't in the US anymore.) I somehow doubt there'd be a dramatic reduction in the cost of education either... education providers seem to charge as much as the market can bear.
I hope you're just trolling. When an experiment's results fail to achieve statistical significance, this provides no evidence of anything whatsoever, and most emphatically does not in any way disprove the statement the scientists were originally trying to test. In fact, if you test even the most solid and reliable scientific finding with too little data, you're almost certain to fail to obtain statistical significance in your results.
I think, unless I'm missing something, that paper is arguing through a rather misleading analogy. Suppose I know nothing about the probability of a Congressman being American or a random person being a Congressman, but do know that the probability of a randomly selected American being a Congressman is very low. Further suppose that I pick a random person and find out that they're a Congressman. This new piece of information gives me knowledge about the previously unknown probability of being a Congressman, and from this and Bayes' theorem I can assign a low probability to a Congressman being American. (Think about it this way: if a large proportion of Congressmen are American, then the number of Congressmen can't be much larger than the number of American Congressmen and so I had a low probability of picking one).
Of course, this only works if I don't already have a better way of determining the probability of a Congressman being American (such as, for example, knowing that all Congressmen are American), in which case I shouldn't discard that probability in favour of the new estimate. Bayesian probability is a bit weird...
(Note that probability is not my strong point, so this may not be 100% correct.)
Just because Apple has a mutant version doesn't mean you couldn't build plain GCC and an assload of free software.
You can build plain GCC, but I'm not sure how well linking against Apple's libraries would work with it... they're targetted at different Objective-C ABIs and runtimes.
Slashdot Mirror
Judging from the list, many users probably didn't sign up with real info, or even with e-mail addresses they actually controlled...
Racial profiling (non-jews and especially arabs are singled for more thorough examination). This is considered perfectly normal in Israel, but is not legal in US or EU.
Probably not terribly effective either. As far as I can tell the only reason it works at all is not because the Jewish population doesn't contain terrorists but because that they're reasonably sure Israeli Jews will only attack politically acceptable targets. Even that only works due to an atmosphere of imminent danger fed by the media that aims lone-wolf nutters safely at Arabs and Arab-sympathisers.
If companies don't need to hire humans, then where are they spending all of they money in production?
Raw materials, capital (equipment costs, buildings and depreciation), and land. If companies don't need humans anymore, those are the remaining scarce resources, so it's fairly obvious that's where the money will go. The trouble is that pretty much all the sources of raw materials, the majority of the land, and the money required for initial capital costs is owned and controlled by a small proportion of super-rich individuals, so they're the only ones that actually make money.
Ironically, capitalism has the same flaw that did communism in: it gives too much power to the individuals that control the allocation of resources.
If their production costs are very low, then their unit prices (will eventually) be very low. That's a good thing!
Low, low prices still don't help if your income is a big fat zero. Their production costs won't be that low either, because there will still be competition for the resources that are still scarce and in demand - it's just the vast majority of the population won't benefit from this.
They've had a btrfsck for a long time, it's just not actually capable of detecting or fixing most forms of filesystem corruption yet. (I believe the original version literally did nothing.)
As opposed to saying something trite and misleading, like you did? Microsoft wasn't just "using XML to do something XML was designed to do". The patent is fairly specific: it doesn't just require the use of XML, to infringe you have to have one file or piece of data containing sections of text, then a second document file that includes pieces of text from the first file identified by their location in said first file and also applies formatting to them. The reason Microsoft got caught out is that their code was an intentional clone of an i4i product; everyone else tends to use different methods to achieve the same aims that happen to have lots of prior art.
Now, if you want actual patent abuse, take a look at Microsoft's attempt to patent all use of XML files to describe office documents a while ago. (Thankfully that was restricted to only cover their OOXML format by the time it was granted. ODF is actually surprisingly patent-proof thanks to the amount of really old SGML stuff that's described in published papers.)
No inventing necessary. The Microsoft exec in charge of this scheme was practically gloating in internal e-mails about how integrating the functionality of i4i's main product into Word would drive i4i out of business. Not to mention that Microsoft have a long history of doing exactly that...
Well, i4i does at least make something. They now sell a utility that undoes the damage caused to documents by Microsoft's forced removal of their custom XML feature.
Yes, they sell a utility providing very useful (and patented) functionality to Word users - just like they did when Microsoft blatently cloned an earlier version of said utility, called it custom XML and integrated it into Microsoft Office. The court case even turned up internal Microsoft memos gloating about this and about how they would drive i4i oout of business by integrating the functionality of their main product into Office.
It should be fairly easy to tell that this has happened, because (a) the amount of data after decompression will equal the length given in the header and (b) you can detect end-of-file fairly robustly just using the format of the gzipped data anyway.
OK that's exactly what I'm looking for. So its no worse than dropping back to "one factor".
Unfortunately, RSA have tended to encourage customers to use a 4 digit numeric password with SecurID in the past because the security of the dongle supposedly made this safe, and this is the single factor that their level of security has now been reduced too. Hence why they were so keen to encourage the use of strong passwords by their customers when they first discovered they'd been hacked.
That's not terribly difficult to manage if you manage to obtain your fob's seed somehow; the algorithm has been public for a while. Of course, for security reasons they're designed to prevent you from doing that.
With RSA doing the keyfill at point of manufacture, the customer just needs to load the seed file for the entire batch onto their authentication server and then hand out the tokens, which are glued shut with considerable enthusiasm, and have no externally accessible electrical connections of any sort.
They don't do that though. As far as anyone outside RSA can tell, the keyfill port is a row of 7 PCB contacts hidden behind a little rectangular stick-on plastic cover on the back of the device. The cover doesn't even seem to be tamper-evident let alone tamper-resistant - you just press it back down again after you're done gawking and it sticks right back into place.
Examples?
Off hand, I seem to recall this is true of the equivalent iPhone feature (apparently it's a bit hairy to do in practice too), their new "faster shutdown" feature on Mac OS X (requires apps to do the hard work of tracking whether they can be killed safely without losing data, including within any libraries they use), Grand Central Dispatch (mostly just a fairly standard thread pool, and yet Apple get all the credit for faster third-party app responsiveness when the app developers did all the hard work), and probably more stuff I haven't looked into.
You do realize that Apple released the code for FaceTime for others to build it into their clients.
They said they were going to and got a lot of good PR about it, then did absolutely nothing. As far as anyone can tell it's not even really practical to reverse engineer...
Errrrm... probably on one of the sites that saw the announcement at WWDC at a guess. For example, maybe Engadget (if you scroll down a bit on their liveblog to around 1:37PM, you can even see the big shiny "Only in the App Store" slide).
You'd think so, but Apple really don't seem to care too much about the corporate market. The OS X Lion Server add-on is apparently only available from the App Store too for what it's worth.
But I think what you say about Lion is incorrect. "Mac OS X Lion's new Resume feature lets users get back to where they left off after a shutdown or restart"
And based on how Apple's previous features have played out, I suspect they'll probably push all the hard work of actually implementing this onto the application developers...
That's not what Engadget seem to think. Of course, when Apple originally announced OS X Lion was going to be available from the App Store they said it was in addition to existing retail channels, but you know what Apple are like for changing their mind...
Hard questions like "is what the US government's doing to Bradley Manning torture", or "hard" questions like "is Bradley Manning a queer pinko faggot who obviously just leaked the info because he's nuts"? I'm guessing the latter... it's what most of the press has been doing.
Really, more systems should make use of the various encrypted key exchange schemes. They fairly strongly guarantee that you can only get one guess at the password per attempted login, even if you manage to intercept the communication beween server and client. (Obviously there's not a lot they can do about brute force attacks if you manage to acquire the information the server uses to verify the passwords, though.)
Not only that, but this is actually saying that the US embassy's contact didn't see any bloodshed in the square at the time, not that there wasn't any. All of the leaked cables have similar caveats attached.
AMD/ATI support is fine now, but with NVIDIA you're basically guaranteed to stay supported for a long time, since there's quite a large community programming with CUDA on Linux.
That depends - do you mean support for new cards as they're released, or continued support for existing cards? NVidia are very good at supporting newer cards, but some of their older cards aren't really usable under Linux anymore, whereas it looks like older ATI cards will remain supported for a long time...
Well sure, if by 'eating the poor kids' you mean my proposal of having the market forces directing them to various apprenticeships and whatever levels of education they could pay for, though this would require dramatic reduction in the costs
There's pretty much no such thing as apprenticeships anymore - the free market decided it was cheaper to employ a worker that someone else trained than to train them up yourself, and everyone stopped offering them. (Besides, most of the jobs that had apprenticeships aren't in the US anymore.) I somehow doubt there'd be a dramatic reduction in the cost of education either... education providers seem to charge as much as the market can bear.
I hope you're just trolling. When an experiment's results fail to achieve statistical significance, this provides no evidence of anything whatsoever, and most emphatically does not in any way disprove the statement the scientists were originally trying to test. In fact, if you test even the most solid and reliable scientific finding with too little data, you're almost certain to fail to obtain statistical significance in your results.
I think, unless I'm missing something, that paper is arguing through a rather misleading analogy. Suppose I know nothing about the probability of a Congressman being American or a random person being a Congressman, but do know that the probability of a randomly selected American being a Congressman is very low. Further suppose that I pick a random person and find out that they're a Congressman. This new piece of information gives me knowledge about the previously unknown probability of being a Congressman, and from this and Bayes' theorem I can assign a low probability to a Congressman being American. (Think about it this way: if a large proportion of Congressmen are American, then the number of Congressmen can't be much larger than the number of American Congressmen and so I had a low probability of picking one).
Of course, this only works if I don't already have a better way of determining the probability of a Congressman being American (such as, for example, knowing that all Congressmen are American), in which case I shouldn't discard that probability in favour of the new estimate. Bayesian probability is a bit weird...
(Note that probability is not my strong point, so this may not be 100% correct.)
Just because Apple has a mutant version doesn't mean you couldn't build plain GCC and an assload of free software.
You can build plain GCC, but I'm not sure how well linking against Apple's libraries would work with it... they're targetted at different Objective-C ABIs and runtimes.