Actually, don't the cheques start at $2.56, and then shift left by 1 as each error is found, up to a maximum of $327.68? (It's wise of Knuth to put a cap on it.. you might be tempted to cash a cheque worth (164)*$0.01..)
It's not far off reality. There isn't an NSA room in every data centre, but there might as well be, since their placement at major Internet hubs throughout the USA is equivalent. Thestory is quite well known.
Well, he is right about SMTP. The amount of spam tells us it's a bad design, awful in fact, but it can't be changed. Using PGP will secure your message, but that's no good if it gets lost inside a flood of spam, or rejected because someone has blackholed your SMTP server's IP address. If only encryption and digital signatures had been part of the standard to begin with. But they're not, so we're stuck with spam-filled inboxes until email is entirely replaced by Facebook messages and MSN.
That's not how DRM works in 2038. It works in the only way it could: by preventing you passing copies of information across the network, unless you have the rights to distribute those copies. Maybe you try to get around this by telling the network that the pirated files are "original", "freely redistributable" and that you are the creator... but then, you just identified yourself as the pirate. Not what you want in an era where the penalties for copyright infringement are actually enforced.
You can't hack your way around this: if you want to get online, you have to play by the network's rules. Of course you could always make your own network, possibly with blackjack and hookers, but if the usefulness of a network is O(n^2) in the number of users...
I think what you really mean is "DRM does not work, mathematically speaking, on a standalone machine". If the claim was more generally true, then why would anyone care about the possible applications of TCPA?
Ultimately there is too much investment in what some call "imaginary property" to simply drop the whole idea of copy protection. People are going to find ways to make it work. When this happens, we all have to make sure it will work for us. Personally, I want it to work with Linux and other free software, which is not as crazy as it sounds, because copy protection functionality can (and should) be implemented by a hypervisor that runs on top of the existing OS. This is important for security as well as openness, because the hypervisor can be small enough to be bug free.
So confident are you that DRM can always be hacked. I'm surprised that the stack-smashing Twilight Hack was even possible; this sort of attack was common against the original XBox, and you would think that Nintendo would have taken steps to prevent it.
Microsoft, at least, seem to have learned. It doesn't seem to be so easy to do this sort of thing on the XBox 360. It looks like someone discovered a hypervisor bug some time ago, allowing unsigned code to run, but that's blocked in more recent versions. The pirates use firmware hacks for the DVD-ROM to run illegal copies of signed code, which is useless for homebrew.
Ah, you highlighted it yourself! The camera needs to be specially configured for the purpose. You can't just apply the technology to an existing CCTV network.
This technology doesn't work well enough for CCTV. It requires decent lighting and high-resolution images; not so difficult if you're able to tell the subject to stare into the camera, but pretty useless if they're just walking past it.
Automated CCTV observation is interesting, but it's currently much better at following objects than recognising people. You need some other mechanism to link the objects with their identities.
I think you're right, but it will be an x86-compatible PC, because the ability to run legacy x86/Windows apps is a major selling point. (This is, unfortunately, why other CPU architectures are doomed in consumer space.)
Also, the images won't just be drawn flat on the inside of the glasses. They will be drawn as if you were looking at them on a physical surface in front of you. In reality, there is only a mouse and keyboard on your desk, but you can also see two 25" monitors there.
A few UAVs borrowed from the military are already deployed and functioning along the southern border...
That's interesting, but I think their uses would be very limited if they were over NYC, where there is a lot of cover (buildings) and the number of people is very large. Conversely, cellphone-based tracking continues to be useful in any city, regardless of the population size and the number of buildings.
Sorry to say that I just can't be paranoid about "spy satellites" in the cellphone era. Satellite imaging won't give real-time coverage of a large area, so it is useless for mass surveillance.
These days, there's no point inventing new privacy threats to worry about, since you already carry an radio tracking device, and there's already at least one database with a list of your recent movements in it. But by all means, do continue to keep your passport in a tin-foil wrapper...
Does this require each hotspot owner to register the location of his/her hotspot, so that a database can be queried to find the location?
e.g. "I can see access points with MAC addresses 00:60:08:57:3C:D2 and 00:E0:18:77:D6:40 so I know I'm at 37 23.516 -122 02.625.."
How many hotspot owners can be bothered to register their correct location? And re-register it if their IP address changes? How many even change their password?
Oh and why did't netbook manufactures use OMAP3 yet?... Only because it's not x86 and windows will not run on it???
Yes.
There's no reason why Intel can't (eventually) make a low power x86 CPU that competes seriously with these ARM CPUs... Atom might not be it, but the next generation probably will be. And then, these Intel CPUs will have a major advantage over other CPUs for mobile devices. Their ability to run almost any program is what people actually want. x86 has won time and again, not because it is better than its competitors, but because it is so widely used, and compatibility is what counts.
Slashdot Mirror
Actually, don't the cheques start at $2.56, and then shift left by 1 as each error is found, up to a maximum of $327.68? (It's wise of Knuth to put a cap on it.. you might be tempted to cash a cheque worth (164)*$0.01..)
It's not far off reality. There isn't an NSA room in every data centre, but there might as well be, since their placement at major Internet hubs throughout the USA is equivalent. The story is quite well known.
Well, he is right about SMTP. The amount of spam tells us it's a bad design, awful in fact, but it can't be changed. Using PGP will secure your message, but that's no good if it gets lost inside a flood of spam, or rejected because someone has blackholed your SMTP server's IP address. If only encryption and digital signatures had been part of the standard to begin with. But they're not, so we're stuck with spam-filled inboxes until email is entirely replaced by Facebook messages and MSN.
That's not how DRM works in 2038. It works in the only way it could: by preventing you passing copies of information across the network, unless you have the rights to distribute those copies. Maybe you try to get around this by telling the network that the pirated files are "original", "freely redistributable" and that you are the creator... but then, you just identified yourself as the pirate. Not what you want in an era where the penalties for copyright infringement are actually enforced.
You can't hack your way around this: if you want to get online, you have to play by the network's rules. Of course you could always make your own network, possibly with blackjack and hookers, but if the usefulness of a network is O(n^2) in the number of users...
I think what you really mean is "DRM does not work, mathematically speaking, on a standalone machine". If the claim was more generally true, then why would anyone care about the possible applications of TCPA?
Ultimately there is too much investment in what some call "imaginary property" to simply drop the whole idea of copy protection. People are going to find ways to make it work. When this happens, we all have to make sure it will work for us. Personally, I want it to work with Linux and other free software, which is not as crazy as it sounds, because copy protection functionality can (and should) be implemented by a hypervisor that runs on top of the existing OS. This is important for security as well as openness, because the hypervisor can be small enough to be bug free.
So confident are you that DRM can always be hacked. I'm surprised that the stack-smashing Twilight Hack was even possible; this sort of attack was common against the original XBox, and you would think that Nintendo would have taken steps to prevent it.
Microsoft, at least, seem to have learned. It doesn't seem to be so easy to do this sort of thing on the XBox 360. It looks like someone discovered a hypervisor bug some time ago, allowing unsigned code to run, but that's blocked in more recent versions. The pirates use firmware hacks for the DVD-ROM to run illegal copies of signed code, which is useless for homebrew.
Ah, you highlighted it yourself! The camera needs to be specially configured for the purpose. You can't just apply the technology to an existing CCTV network.
This technology doesn't work well enough for CCTV. It requires decent lighting and high-resolution images; not so difficult if you're able to tell the subject to stare into the camera, but pretty useless if they're just walking past it.
Automated CCTV observation is interesting, but it's currently much better at following objects than recognising people. You need some other mechanism to link the objects with their identities.
I think you're right, but it will be an x86-compatible PC, because the ability to run legacy x86/Windows apps is a major selling point. (This is, unfortunately, why other CPU architectures are doomed in consumer space.)
Also, the images won't just be drawn flat on the inside of the glasses. They will be drawn as if you were looking at them on a physical surface in front of you. In reality, there is only a mouse and keyboard on your desk, but you can also see two 25" monitors there.
A few UAVs borrowed from the military are already deployed and functioning along the southern border...
That's interesting, but I think their uses would be very limited if they were over NYC, where there is a lot of cover (buildings) and the number of people is very large. Conversely, cellphone-based tracking continues to be useful in any city, regardless of the population size and the number of buildings.
Sorry to say that I just can't be paranoid about "spy satellites" in the cellphone era. Satellite imaging won't give real-time coverage of a large area, so it is useless for mass surveillance.
These days, there's no point inventing new privacy threats to worry about, since you already carry an radio tracking device, and there's already at least one database with a list of your recent movements in it. But by all means, do continue to keep your passport in a tin-foil wrapper...
Worse, they might try to make friends with us.
Does this require each hotspot owner to register the location of his/her hotspot, so that a database can be queried to find the location?
e.g. "I can see access points with MAC addresses 00:60:08:57:3C:D2 and 00:E0:18:77:D6:40 so I know I'm at 37 23.516 -122 02.625.."
How many hotspot owners can be bothered to register their correct location? And re-register it if their IP address changes? How many even change their password?
Yes.
There's no reason why Intel can't (eventually) make a low power x86 CPU that competes seriously with these ARM CPUs... Atom might not be it, but the next generation probably will be. And then, these Intel CPUs will have a major advantage over other CPUs for mobile devices. Their ability to run almost any program is what people actually want. x86 has won time and again, not because it is better than its competitors, but because it is so widely used, and compatibility is what counts.