I am aware that there are laws of nature other than math. They are not at issue here.
The case you cited stands for the principle that if the only NOVEL ADDITION to human knowledge a patent would bring is a mathematical formula, then the idea/invention is not patentable since it adds nothing to human knowledge except unpatentable subject matter. In almost all software patents, the only novel addition to human knowledge is a mathematical algorithm, since all software is composed solely of mathematical algorithms. Thus, the argument goes, software running on a computer is not patentable, since "running on a computer" is not novel and the mathematical algorithms composing software are unpatentable by themselves by virtue of being laws of nature.
This is the way the Supreme Court has ruled so far. The Supreme Court has also been very skeptical of method patents in the recent past. It's impossible to predict how the court will rule in Alice Corp. vs. CLS Bank, of course, but, given precedent, there's a good chance we could get software explicitly declared unpatentable subject matter, which would be a big win. Throwing out "thousands of issued patents" didn't stop the court with Bilski.
First, software exists outside the physical world. Every piece of software is an algorithm, and algorithms are pure math, and pure math, by very long-standing precedent, isn't patentable. Hence, the argument goes, software isn't patentable, because doing so is equivalent to patenting the pure math behind it.
Second, the real reason software shouldn't be patentable is because of all the policy reasons given in this discussion. They just cause more trouble than they're worth. The "software is math" argument is just the vehicle we're using to pursue our policy goals. That's not to say it's not a good legal argument -- it is -- but the reason we like to make this legal argument is because software patents are bad policy, and we think it will be easier to get courts to interpret the existing law to exclude software patents than it would be to get Congress to change the law to explicitly outlaw software patents.
Other ways you could defeat this: - Take a picture of the screen with your camera phone. Yes, they don't allow camera phones, but you could probably smuggle one in. This obviously doesn't work with anything that's too long. - Download an executable with the magic number changed to get past the proxy, edit it in notepad to restore the magic number, and then use it to encrypt the media, then upload it to a server under your control, which won't raise red flags because it doesn't see any "bad keywords" because you encrypted it. XOR encryption would probably work here. This requires you to be able to execute programs, yes, but, uh, you could probably use JavaScript in a browser if you really had to. Local site that uuencodes or similar any binary data that you paste in one window and displays it in the other window. Banned local sites, okay, set up a site outside the firewall that does this and gets past the proxy because the proxy doesn't see anything wrong with it because it's not expecting this.
It would really be pretty hard to do this so it can't be compromised by a motivated attacker. Impossible? If you consider that the system can't be so locked-down no one can work on it, perhaps yes.
Other programming possibilities: Macros in Word. Java applets. FLASH (shudder) applets. These would be much easier to get working right than a JavaScript thing. You can't use the Internet without JavaScript. Can you use the Internet without Flash? Maybe. Or maybe it would annoy your employees so much they quit because you're making it too hard to do their jobs.
Really the hardest part of doing this, now that I'm thinking about it, would be the "hex dump" step. Once you've installed the equivalent of a hex dumper -- some way, some how -- you can manipulate that hex however you want to get it past the proxy. After you can dump the secret data to hex, it's game over.
If you think terrorists are scary, you should never drive or get in a car ever again, because doing that is much more likely to get you killed than the big bad oh-so-scary terrorists you're going on about.
Is the world "scary"? Well, everyone dies eventually, and I guess death is scary, so sure. What's scariest about it? Cancer and heart disease. Yup. If you're going to worry about stuff that could kill you, worry about cancer and heart disease. Because it's about 80% likely that that is what will kill you. Terrorists well let's see they're like #2000 on the list of stuff that is likely to get you killed, if that. So, no, it is not we who are not understanding what's happening around us. It is you who needs a crash course in statistics. Badly.
It's simply a wrong comment. The NPAPI version of Flash is _NOT_ unsupported. 11.2 is the last version that will be made available as an NPAPI Linux plugin, but Adobe plans to keep fixing security issues in the 11.2 version plugin indefinitely.
Question for anyone who knows: how the hell did KlearGear report a debt to a credit reporting agency in the first place? The credit reports are indexed by SSN, and they only have other identifiers like credit card numbers to go by if you don't have that. They paid by PayPal. Doesn't PayPal hide your credit card number from the merchant? With just a name, how did they report it? Does anyone know?
The security gain comes from the fact that it is feasible to perform a side-channel attack on RAM but infeasible to perform a side-channel attack on CPU registers. The data to recreate the keys is scrubbed from RAM; the keys never leave RAM. I have done work on a similar project to TRESOR, called Loop-Amnesia, which uses MSRs instead of the debug registers to perform the same task and does not require AES-NI support.
I can't find the section you're referring to, and my research seems to indicate that laws passed by Congress supersede earlier conflicting international treaties. Perhaps you could indicate what you're referring to?
Parent is trolling. I took the bait so I'm giving up karma bonus.
What I'm saying is that Linux on minority architectures that doesn't get much love, such as SPARC, has QA problems, because, no, unlike the assertion I was replying to, "just recompiling" doesn't cut it. Anything written in a C derivative (other than C#, which is really a Java derivative) on Windows will suffer from the same QA concerns that hit Linux on SPARC, which is why some companies don't even port stuff to AMD64. In the proprietary world, you just don't get software on your favorite platform; in the open source world, you get it, but you might not work. I know which I think is better, especially since I'm using much of the same software I ran on Linux/SPARC on Solaris/SPARC now using OpenCSW, and it works better now since more people use Solaris and so it gets more bug reports and therefore more QA.
You want to make an argument that Windows developers -- who typically have no experience with cross-architecture development -- will be better at writing cross-architecture software than Linux/UNIX developers, who typically do, go ahead and make it. I'd love to hear it.
Yeah, man, go see how that works for you. Take it from someone who spent 3 years running Linux on SPARC before giving up in frustration. Even though most things on Linux are MAINTAINED ACTIVELY for basically every architecture thanks to Debian, you'll run into weird little issues like Firefox crashing randomly, X crashing randomly, GhostScript for some reason crashing quite consistently unless you use a crazy old version, and tons of other little annoying problems that eventually force you to switch to Solaris. And this is with people maintaining the code, just not looking at it too much since few people run or report bugs on Linux/SPARC.
"It's not a big deal. You just rebuild your software." What a laugh. News flash: it's very easy to write totally nonportable C if you don't know what you're doing without even realizing it if you're not careful. News flash 2: most Linux/UNIX developers are careful, and they still mess up (per above). News flash 3: most crap Windows developers aren't careful at all.
Microsoft can't even get people onboard for AMD64 yet. And that's an architecture transition that's actually important.
It does seem unfair. For instance, even if the police "know" someone is a drug pusher (like he got convicted of it before), they still need probable cause to search him if he's walking around town minding his own business. The difference here is that there are NO LIMITS to reasonableness of searches conducted by Customs at the border (the real one, not the alarmist ACLU one). They could give every US citizen returning home a body-cavity search if they wanted to. It's up to the people to use the democratic process to decide how much we want to balance the risk of being personally humiliated at the border with the risk that we will tie Customs' hands and prevent them from doing their job. Now, my personal opinion is that transferring digital files across the border is not something Customs should be guarding against, since there's this little thing called the Internet that bad guys can use to transfer digital files across the border just as easily. My vote would go to a politician who supported reducing Customs' role here. But that's just me. I'm one voter out of many.
Huh. According to http://en.wikipedia.org/wiki/United_States_Border_Patrol_Interior_Checkpoints they shouldn't be able to do that unless they have probable cause. Now, looking through the windows probably does not qualify as "search" in that sense, but if they got in the car and opened the glove compartment, that's a search alright. Or, did you consent (which you shouldn't have if you did)?
I mean, it was an accident. 3 years in prison for a poor woman scavenging for metal doesn't seem too fair, at least assuming her goal wasn't to steal copper wire. Hopefully they won't charge her, or will give her a slap on the wrist.
Actually, under the current system, only evidence A STATE ACTOR illegally obtained is inadmissible. Under the current system, the used condom would be admissible evidence. Your sister's rapist would most likely be convicted. You could theoretically be charged with trespassing, but I doubt there'd be a DA willing to take that case -- especially considering DAs are usually elected officials.
The last I heard, Motorola was still fighting them. Good on them, even though they make crap phones and DRMed the Droid 2 to hell.
I'm actually not too worried about the Android patent battles. Phone companies are big, have a lot of patents, and are used to suing the crap out of each other. That's probably why the whole industry was locked-down and stagnant until Google finally liberated it, but that's beside the point. The point is that phone manufacturers are used to patent wars and can afford to fight them. It's like how the Greeks never built an empire because they were too busy beating each other up but nevertheless managed to stop the Persians because their experiences beating each other up actually made them pretty good at fighting.
I doubt Microsoft will be able to stop Android with lawsuits. Microsoft will have to compete on the merits against Android, which has the functionality advantage, and Apple, which has the fashion advantage. It's in pretty serious trouble.
Slashdot Mirror
Also, have a look at Gottschalk v. Benson. It's an oldie but a goodie.
I am aware that there are laws of nature other than math. They are not at issue here.
The case you cited stands for the principle that if the only NOVEL ADDITION to human knowledge a patent would bring is a mathematical formula, then the idea/invention is not patentable since it adds nothing to human knowledge except unpatentable subject matter. In almost all software patents, the only novel addition to human knowledge is a mathematical algorithm, since all software is composed solely of mathematical algorithms. Thus, the argument goes, software running on a computer is not patentable, since "running on a computer" is not novel and the mathematical algorithms composing software are unpatentable by themselves by virtue of being laws of nature.
This is the way the Supreme Court has ruled so far. The Supreme Court has also been very skeptical of method patents in the recent past. It's impossible to predict how the court will rule in Alice Corp. vs. CLS Bank, of course, but, given precedent, there's a good chance we could get software explicitly declared unpatentable subject matter, which would be a big win. Throwing out "thousands of issued patents" didn't stop the court with Bilski.
Two answers:
First, software exists outside the physical world. Every piece of software is an algorithm, and algorithms are pure math, and pure math, by very long-standing precedent, isn't patentable. Hence, the argument goes, software isn't patentable, because doing so is equivalent to patenting the pure math behind it.
Second, the real reason software shouldn't be patentable is because of all the policy reasons given in this discussion. They just cause more trouble than they're worth. The "software is math" argument is just the vehicle we're using to pursue our policy goals. That's not to say it's not a good legal argument -- it is -- but the reason we like to make this legal argument is because software patents are bad policy, and we think it will be easier to get courts to interpret the existing law to exclude software patents than it would be to get Congress to change the law to explicitly outlaw software patents.
Yeah, and the body of someone who never exercised a day in his life is likely to look better than an Olympic athlete's...
Other ways you could defeat this:
- Take a picture of the screen with your camera phone. Yes, they don't allow camera phones, but you could probably smuggle one in. This obviously doesn't work with anything that's too long.
- Download an executable with the magic number changed to get past the proxy, edit it in notepad to restore the magic number, and then use it to encrypt the media, then upload it to a server under your control, which won't raise red flags because it doesn't see any "bad keywords" because you encrypted it. XOR encryption would probably work here. This requires you to be able to execute programs, yes, but, uh, you could probably use JavaScript in a browser if you really had to. Local site that uuencodes or similar any binary data that you paste in one window and displays it in the other window. Banned local sites, okay, set up a site outside the firewall that does this and gets past the proxy because the proxy doesn't see anything wrong with it because it's not expecting this.
It would really be pretty hard to do this so it can't be compromised by a motivated attacker. Impossible? If you consider that the system can't be so locked-down no one can work on it, perhaps yes.
Other programming possibilities: Macros in Word. Java applets. FLASH (shudder) applets. These would be much easier to get working right than a JavaScript thing. You can't use the Internet without JavaScript. Can you use the Internet without Flash? Maybe. Or maybe it would annoy your employees so much they quit because you're making it too hard to do their jobs.
Really the hardest part of doing this, now that I'm thinking about it, would be the "hex dump" step. Once you've installed the equivalent of a hex dumper -- some way, some how -- you can manipulate that hex however you want to get it past the proxy. After you can dump the secret data to hex, it's game over.
Oh, yeah, I almost forgot:
FUCK BETA!
---linuxrocks123
If you think terrorists are scary, you should never drive or get in a car ever again, because doing that is much more likely to get you killed than the big bad oh-so-scary terrorists you're going on about.
Is the world "scary"? Well, everyone dies eventually, and I guess death is scary, so sure. What's scariest about it? Cancer and heart disease. Yup. If you're going to worry about stuff that could kill you, worry about cancer and heart disease. Because it's about 80% likely that that is what will kill you. Terrorists well let's see they're like #2000 on the list of stuff that is likely to get you killed, if that. So, no, it is not we who are not understanding what's happening around us. It is you who needs a crash course in statistics. Badly.
---linuxrocks123
It's simply a wrong comment. The NPAPI version of Flash is _NOT_ unsupported. 11.2 is the last version that will be made available as an NPAPI Linux plugin, but Adobe plans to keep fixing security issues in the 11.2 version plugin indefinitely.
---linuxrocks123
> What online dating suffers from is 75% of the participants are dudes...
No true, except possibly on AdultFriendFinder.
http://www.nextadvisor.com/blo...
Question for anyone who knows: how the hell did KlearGear report a debt to a credit reporting agency in the first place? The credit reports are indexed by SSN, and they only have other identifiers like credit card numbers to go by if you don't have that. They paid by PayPal. Doesn't PayPal hide your credit card number from the merchant? With just a name, how did they report it? Does anyone know?
Sorry, the keys never leave the debug registers -- the keys are not computable from anything in RAM.
The security gain comes from the fact that it is feasible to perform a side-channel attack on RAM but infeasible to perform a side-channel attack on CPU registers. The data to recreate the keys is scrubbed from RAM; the keys never leave RAM. I have done work on a similar project to TRESOR, called Loop-Amnesia, which uses MSRs instead of the debug registers to perform the same task and does not require AES-NI support.
---linuxrocks123
Canadian courts do, eh?
Was this footage made public? If not, why not?
The report states that "the engine N1's were at about 55%". I'm not a pilot, but 55% doesn't sound like anywhere near full thrust to me.
I can't find the section you're referring to, and my research seems to indicate that laws passed by Congress supersede earlier conflicting international treaties. Perhaps you could indicate what you're referring to?
Parent is trolling. I took the bait so I'm giving up karma bonus.
What I'm saying is that Linux on minority architectures that doesn't get much love, such as SPARC, has QA problems, because, no, unlike the assertion I was replying to, "just recompiling" doesn't cut it. Anything written in a C derivative (other than C#, which is really a Java derivative) on Windows will suffer from the same QA concerns that hit Linux on SPARC, which is why some companies don't even port stuff to AMD64. In the proprietary world, you just don't get software on your favorite platform; in the open source world, you get it, but you might not work. I know which I think is better, especially since I'm using much of the same software I ran on Linux/SPARC on Solaris/SPARC now using OpenCSW, and it works better now since more people use Solaris and so it gets more bug reports and therefore more QA.
You want to make an argument that Windows developers -- who typically have no experience with cross-architecture development -- will be better at writing cross-architecture software than Linux/UNIX developers, who typically do, go ahead and make it. I'd love to hear it.
Oh, suuuuure. Just rebuild. It'll be fiiiiiiine.
Yeah, man, go see how that works for you. Take it from someone who spent 3 years running Linux on SPARC before giving up in frustration. Even though most things on Linux are MAINTAINED ACTIVELY for basically every architecture thanks to Debian, you'll run into weird little issues like Firefox crashing randomly, X crashing randomly, GhostScript for some reason crashing quite consistently unless you use a crazy old version, and tons of other little annoying problems that eventually force you to switch to Solaris. And this is with people maintaining the code, just not looking at it too much since few people run or report bugs on Linux/SPARC.
"It's not a big deal. You just rebuild your software." What a laugh. News flash: it's very easy to write totally nonportable C if you don't know what you're doing without even realizing it if you're not careful. News flash 2: most Linux/UNIX developers are careful, and they still mess up (per above). News flash 3: most crap Windows developers aren't careful at all.
Microsoft can't even get people onboard for AMD64 yet. And that's an architecture transition that's actually important.
It does seem unfair. For instance, even if the police "know" someone is a drug pusher (like he got convicted of it before), they still need probable cause to search him if he's walking around town minding his own business. The difference here is that there are NO LIMITS to reasonableness of searches conducted by Customs at the border (the real one, not the alarmist ACLU one). They could give every US citizen returning home a body-cavity search if they wanted to. It's up to the people to use the democratic process to decide how much we want to balance the risk of being personally humiliated at the border with the risk that we will tie Customs' hands and prevent them from doing their job. Now, my personal opinion is that transferring digital files across the border is not something Customs should be guarding against, since there's this little thing called the Internet that bad guys can use to transfer digital files across the border just as easily. My vote would go to a politician who supported reducing Customs' role here. But that's just me. I'm one voter out of many.
Huh. According to http://en.wikipedia.org/wiki/United_States_Border_Patrol_Interior_Checkpoints they shouldn't be able to do that unless they have probable cause. Now, looking through the windows probably does not qualify as "search" in that sense, but if they got in the car and opened the glove compartment, that's a search alright. Or, did you consent (which you shouldn't have if you did)?
I mean, it was an accident. 3 years in prison for a poor woman scavenging for metal doesn't seem too fair, at least assuming her goal wasn't to steal copper wire. Hopefully they won't charge her, or will give her a slap on the wrist.
---linuxrocks123
Actually, under the current system, only evidence A STATE ACTOR illegally obtained is inadmissible. Under the current system, the used condom would be admissible evidence. Your sister's rapist would most likely be convicted. You could theoretically be charged with trespassing, but I doubt there'd be a DA willing to take that case -- especially considering DAs are usually elected officials.
iPads sell well for the same reason jeans with holes in them do.
I don't believe M$ is making the phones, however, Microsoft can still be sued for making the software.
The last I heard, Motorola was still fighting them. Good on them, even though they make crap phones and DRMed the Droid 2 to hell.
I'm actually not too worried about the Android patent battles. Phone companies are big, have a lot of patents, and are used to suing the crap out of each other. That's probably why the whole industry was locked-down and stagnant until Google finally liberated it, but that's beside the point. The point is that phone manufacturers are used to patent wars and can afford to fight them. It's like how the Greeks never built an empire because they were too busy beating each other up but nevertheless managed to stop the Persians because their experiences beating each other up actually made them pretty good at fighting.
I doubt Microsoft will be able to stop Android with lawsuits. Microsoft will have to compete on the merits against Android, which has the functionality advantage, and Apple, which has the fashion advantage. It's in pretty serious trouble.