Back in the early 1990s the US government was trying to push a similar security-on-a-chip scheme called Clipper. The idea was that communication between two Clipper devices could only be understood by the two devices... or a Federal agent using the secret law-enforcement key. The technology and legitimate non-government security communities were all up in arms over this, but the law enforcement and spy agencies had more influence over Congress than we did.
Suddenly, the whole thing fell apart. First, (though officially neither-denied-nor-confirmed) the top-secret law enforcement key was rumored to have been one of the things sold to the Soviets by a highly placed mole. Second, a researcher at Bell Labs figured out how to "vandalize" the chips so that they still worked with each other but the law-enforcement key could no longer decrypt the conversation. Clipper vanished overnight.
This story underlines some key failings of Palladium: 1. The Feds won't allow it unless they get a back door. 2. Someone's going to steal the Feds' skeleton key. 3. Someone's going to break the chips at the hardware level.
I can't see this working. The black hats will have enormous incentive to break it if it does.
Unfortunately, our "leaders" in business and government are too easily coerced and/or purchased for us to trust them not to go along with it.
Match the medium to its use. If it's something that will be read and digested "offline", the standard dead-tree format will probably be best -- at least until PDAs become more ubiquitous. If the book is more likely to be used as a reference, ring or spiral binding would be best. (Ring bound books are especially handy if you need to temporarily remove pages from different parts of the book for simultaneous viewing). An online reference, be it on a CD or a Web site, should be heavily cross-referenced and indexed.
In short, material that will be accessed sequentially should be a permanent bound book or PDF; random-access information should be spiral or ring bound or HTML.
1. Consider the amount of data that needs to be collected and mined for each individual. Is all of this data going to be stored in one place and updated continuously, or gathered per individual on request? Since associations are going to be traced, they'll want to gather all of this information up front. This is going to require a hell of a lot of storage space and some ungodly bandwidth to maintain.
The level of detail they want to put into your dossier is considerably more than a private investigator could come up with, and PIs charge hundreds (maybe thousands) of dollars for such a report. These guys are going to keep the price down around $2 per ticket. Yeah. Sure.
2. A background check on one prospective passenger will be rather intensive. They're talking about using phone records here, which alone could bind the average person to several dozen other people. Let's call this number "a". Now, they're going to explore seven degrees of association. This means that 1+a**7 people need to be checked to vet one passenger. (Current population of Earth: about 6*10**9). How far in advance do I need to make my reservation?
3. Remember Kevin Bacon? I remember reading a couple of years ago that between any two people on Earth chosen at random there are on the average LESS THAN SIX degrees of separation. Yep, that applies to Ashcroft and Bin Laden as well.
4. Bad data is worse than no data, and it won't take much pollution to render the whole thing completely useless. The Feds will need to tamper with the data to allow their agents to work undercover and to operate the Witness Protection programs. This database will be an irresistable cracker target. And where would we get data on non-citizens?
Both major (and probably some of the minor) political parties will have their private cracks into the database and neither will hesitate very long to use those cracks to find or create dirt on their opponents and to try to clean their own candidates' records. It won't take long for them to dispel anyone's delusion that this thing is in any way accurate.
In short, it's just not going to work. I suspect someone's looking for free publicity or maybe some "venture capital".
It sounds like the judge considered the act of downloading to be "publishing". Of course this means that the plaintiff is the one who actually published the defamatory material in Australia.
Not that logic and law have anything to do with one another.
Carnivore could intercept a copyright-protected transmission without permission from the copyright holder. The fact that it is not intended to do so (at least according to its supporters) and is not supposed to be used for that is moot; if it can, its mere existence is illegal under the DMCA.
Of course, IANAL and with "justice" going to the highest bidder I'm not optimistic about this technicality being worth anything.
I haven't seen the text of the patent so this may not be applicable, but cartographic projection methods may well be another example of prior art.
They address the same issue: presenting a spherical surface on flat plane through controlled distortion. The inverse, photogrammetry, involves distorting flat tiles (aerial or satellite imagery) to model/recreate a 3D surface.
The math dates back to the 1860s or earlier. The U.S. government released FORTRAN source code at least a decade ago. Prior enough?
Any game or entertainment program qualifies as Art. If one argues that the code for such a program is not art, it would have to be because the code itself is not viewed but rather interpreted and expressed through a display device. According to that reasoning, however, neither a written play nor a recording of a performace could be considered art -- in both cases they are merely instructions to be interpreted and expressed by artists and/or display devices.
The source code submitted to the Obfuscated C and Perl contests bypasses even that shaky argument. That code is not primarily intended to accomplish anything practical. Its primary purpose is to express humor, and said humor is experienced by viewing the unperformed and uninterpreted code itself.
Slashdot Mirror
I'd include tweezers, needlenose pliers, and a handfull of jumpers.
How big is the toolbox you were planning on lugging all this stuff around in?
Back in the early 1990s the US government was trying to push a similar security-on-a-chip scheme called Clipper. The idea was that communication between two Clipper devices could only be understood by the two devices ... or a Federal agent using the secret law-enforcement key. The technology and legitimate non-government security communities were all up in arms over this, but the law enforcement and spy agencies had more influence over Congress than we did.
Suddenly, the whole thing fell apart. First, (though officially neither-denied-nor-confirmed) the top-secret law enforcement key was rumored to have been one of the things sold to the Soviets by a highly placed mole. Second, a researcher at Bell Labs figured out how to "vandalize" the chips so that they still worked with each other but the law-enforcement key could no longer decrypt the conversation. Clipper vanished overnight.
This story underlines some key failings of Palladium: 1. The Feds won't allow it unless they get a back door. 2. Someone's going to steal the Feds' skeleton key. 3. Someone's going to break the chips at the hardware level.
I can't see this working. The black hats will have enormous incentive to break it if it does.
Unfortunately, our "leaders" in business and government are too easily coerced and/or purchased for us to trust them not to go along with it.
Match the medium to its use. If it's something that will be read and digested "offline", the standard dead-tree format will probably be best -- at least until PDAs become more ubiquitous. If the book is more likely to be used as a reference, ring or spiral binding would be best. (Ring bound books are especially handy if you need to temporarily remove pages from different parts of the book for simultaneous viewing). An online reference, be it on a CD or a Web site, should be heavily cross-referenced and indexed.
In short, material that will be accessed sequentially should be a permanent bound book or PDF; random-access information should be spiral or ring bound or HTML.
1. Consider the amount of data that needs to be collected and mined for each individual. Is all of this data going to be stored in one place and updated continuously, or gathered per individual on request? Since associations are going to be traced, they'll want to gather all of this information up front. This is going to require a hell of a lot of storage space and some ungodly bandwidth to maintain.
The level of detail they want to put into your dossier is considerably more than a private investigator could come up with, and PIs charge hundreds (maybe thousands) of dollars for such a report. These guys are going to keep the price down around $2 per ticket. Yeah. Sure.
2. A background check on one prospective passenger will be rather intensive. They're talking about using phone records here, which alone could bind the average person to several dozen other people. Let's call this number "a". Now, they're going to explore seven degrees of association. This means that 1+a**7 people need to be checked to vet one passenger. (Current population of Earth: about 6*10**9). How far in advance do I need to make my reservation?
3. Remember Kevin Bacon? I remember reading a couple of years ago that between any two people on Earth chosen at random there are on the average LESS THAN SIX degrees of separation. Yep, that applies to Ashcroft and Bin Laden as well.
4. Bad data is worse than no data, and it won't take much pollution to render the whole thing completely useless. The Feds will need to tamper with the data to allow their agents to work undercover and to operate the Witness Protection programs. This database will be an irresistable cracker target. And where would we get data on non-citizens?
Both major (and probably some of the minor) political parties will have their private cracks into the database and neither will hesitate very long to use those cracks to find or create dirt on their opponents and to try to clean their own candidates' records. It won't take long for them to dispel anyone's delusion that this thing is in any way accurate.
In short, it's just not going to work. I suspect someone's looking for free publicity or maybe some "venture capital".
It sounds like the judge considered the act of downloading to be "publishing". Of course this means that the plaintiff is the one who actually published the defamatory material in Australia.
Not that logic and law have anything to do with one another.
Ask Dmitry Sklyarov.
Carnivore could intercept a copyright-protected transmission without permission from the copyright holder. The fact that it is not intended to do so (at least according to its supporters) and is not supposed to be used for that is moot; if it can, its mere existence is illegal under the DMCA.
Of course, IANAL and with "justice" going to the highest bidder I'm not optimistic about this technicality being worth anything.
I haven't seen the text of the patent so this may not be applicable, but cartographic projection methods may well be another example of prior art. They address the same issue: presenting a spherical surface on flat plane through controlled distortion. The inverse, photogrammetry, involves distorting flat tiles (aerial or satellite imagery) to model/recreate a 3D surface. The math dates back to the 1860s or earlier. The U.S. government released FORTRAN source code at least a decade ago. Prior enough?
The source code submitted to the Obfuscated C and Perl contests bypasses even that shaky argument. That code is not primarily intended to accomplish anything practical. Its primary purpose is to express humor, and said humor is experienced by viewing the unperformed and uninterpreted code itself.