A brief overview of the linked page leads me to the impression that there aren't even any vulnerabilities found or mentioned, it's just speculated that it's conceivable that implementations of a file format may be overflowable because the file format has, gasp, a length field. Lame. (no pun)
I am growing more and more towards the believe that Japanese culture is flat out superior to ours in terms of diversity and creativity. I recently watched in full two anime series: "The Law of Ueki" and "The Melancholy of Haruhi Suzumiya", both of which were excellent. Animation is a time-consuming and expensive process, yet the Japanese seem to produce the finest quality artwork in their sleep without even trying. Their stories tend to actually have consistency, meaning, and development, and they don't seem to like compromising the integrity of their work for marketing or network related reasons.
But I think it really dawned on me how far ahead of us they are when I was watching the Culture Day episode of Haruhi, when the main character sang a song at a talent competition. The song they used was composed specifically for the series so it must have just been an afterthought for the production crew, and yet it still blew most modern real American songs away. All the music I've heard in animes just combines an incredible number of influences from all over the world and perfects these elements into something that could never be generated by one of our artists. It makes me very jealous.
Just as I can't believe this article itself made it to the front page. Why the hell did someone think it was newsworthy to state that vulnerabilities are bad and flaws can be exploited? This just in: The NSA keeps secrets, Schneier fears the government, and bugs in hardware platforms can theoretically hurt their users.
I could very well use that for a poster slogan in my college's chapter of Students for Free Culture. I wonder if quotations like that are copyrightable when in distributed paper form...
Then there are parents who are terrified of all media besides books. A friend of mine claims to only be allowed to play video games of any kind at social LAN parties, and has similar troubles, though not as severe, with television. I do hope that no one in my generation will be so paranoid of the next technology as to govern its consumption in quotas rather reason.
This kind of article reinforces the management perspective mentioned in another recent article that firewalls == security. While most people would attack that equality by saying security is a process and thus much more than firewalls, I never really liked firewalls to begin with. All they are is a place to centralize some rules, which is great but it's not as if they give you any primitive or insight that you didn't already have before. Why block a port when the port isn't even open to begin with - besides centralization and efficiency, from a security perspective what does it offer you, especially when you trust the security policies of all the machines on your internal network?
That's a safety feature. Just as supergerms have self-destruct protein sequences, so they can't live outside a laboratory, so too are scientists protecting us from genetically engineered supermice.
I can't tell if you were being sarcastic, or if that was a call for gender-neutral dialogue, but just in case: Women already have "Motherboard", "Mothership", "Daughter Cells", and personifications of cruise vessels; let the men have this one.
I guess so. I normally don't hop on the Dvorak bashing, probably because I learned of his reputation relatively late, but that was a pretty pathetic piece of speculation. "The gPhone will suck because other phones suck." and "Why search the web when you can talk on the phone" are pretty sucky arguments.
That's the disclosure section of the patent. I was once told that if you can read a patent's claims and understand how the system works without even needing to read the disclosure, then it's a bad patent. In the case of Amazon's recent patent on search, I understood the system just by reading the title alone, indicating that it was an exceptionally poor patent. I was very curious to see what the disclosure would be for such an "invention", but sadly the complete patent is not available online, since it cites by reference application materials that seem to precede the practice of online publication.
> "I'm just saying that this is a poor representation of the case for banning underage smoking, and a poor argument for a change in policy."
Indeed. While I disagree with more than one step in the 6-point argument, it's the last one that I would say is most egregious:
> "Hence, if you believe that smoking should be banned for people under 18, then the same logic would lead to a ban on smoking for people over 18 as well."
because this assumes that we're using the same logic as the scarecrow.
> "Ok, so I messed up the numerical equivalent of milli and the case of the 'k', however as usual around here the point was completely missed in favour of pointing out minor errors."
You specifically quoted those examples and asserted that the "rest of the world", which "geeks" are apparently not part of, uses those terms and definitions. In this case, details are moderately relevant, as you blasted the parent over his deviation from perfect adherence to the standard while demonstrating variations of your own.
> "Have you ever been the only English speaker in a room full of, say, Cantonese? Really difficult to communicate, no? Try explaining your theory to a room full of NON geeks and see what their reactions are. Hint: Usually it's a combination of confusion, bewilderment and bemusement. "Sure, whatever, a small k means a thousand and like twenty eight or something. Whatever you say.""
I'm a bit bewildered by how much importance you attach to this matter. You imply that it is extremely difficult for an average person to comprehend even the *existence* of a technical fact or definition that does not exactly coincide with daily usage, and make it sound as though this would have disastrous consequences. I seriously doubt it would be that hard to explain to most people what a binary kilobyte is. If you wanted to try to be confusing to prove me wrong, you could start off yammering about representing numbers in different bases and quoting various powers of two, and in that case I can imagine their eyes glazing over. If on the other hand you wanted to tell someone that the prefix "kilo" doesn't necessarily mean "thousand" when people talk about disk storage, you could just *say* that and be done with it.
> "The point, and the underlying point to my posting, was that geeks like to re-invent things as convenient to them "
!! Of course, as does anyone with "technical" knowledge in any subject. I fail to see how this is the grievous sin you're making it out to be.
What, you didn't think all words in the English language were used in the same way by all professionals, did you? (By the way, did you know that engineers use the symbol "j" to represent the imaginary unit? Or that a "trusted" party in a security model is one that is capable of breaking your security policy?)
> "but that are completely outside the level of understanding of the rest of the civilized world."
Exaggerate much?
> "It's apparent that a contingent of geeks (and, apparently, lawyers) would like hard drive storage re-defined but refuse to accept any of the standard notations that have been proposed which would do exactly that."
Re-defined? How long have mega, giga, etc., been understood as binary prefixes when applied to hard drive storage? Was there ever a time when this was not the case? It's certainly not a recent development on the part of renegade geeks who aim to destroy society by attacking its language. Perhaps the proposed alternative notations have not been adopted by many because they *suck*. By your logic, we should all suddenly start using the prefixes "kibi", "mebi", and "gibi" in all conversations when we wish to indicate multiples of 1024. You can't possibly expect that to not raise some eyebrows among a crowd of non-geeks; they'll think you have a speech impediment.
> "Mega, kilo, giga, etc. have existed for quite some time both in and out of the computer industry as base-10 notations - even with the measure of network bandwidth, CPU Hz and other forms of measurement but when it comes to disk drive storage capacity suddenly the rules change and an exception is formed?!? Nonsense. Pure nonsense."
Prefixes, like all forms of vocabulary, mean what the speaker and listener agree upon. The English language has loads of exceptions that one could claim are pure nonsense under your logic, but that doesn't make it so. A significant portion of people use those words as binary prefixes, and while you can argue that it's suboptimal, the mere fact that it's a
Since it does not make an ounce of sense to talk about fractions of a bit unless we're describing averages, entropy, etc., I don't see why we can't reuse the lowercase half of the alphabet for the purpose the parent described. And yes, I believe I've seen it before.
> "{SIGH!} And people wonder why geeks are ostracized."
I'm currently reading "Copyrights and Copywrongs" by Siva Vaidhyanathan (which I'll recommend, although I've only read like, one chapter), and the first example he gives in the introduction is Warner Brothers threatening to sue Groucho Marx for using the film title "A Night in Casablanca".
From page 2: "Then Marx turned the issue of name ownership on the Warners. He conceded that they could claim control of "Warner," but certainly not "brothers." Marx claimed, "Professionally, we were brothers long before you were." Marc pointed out that even before the Marx Brothers, there were still the Smith Brothers, the Brothers Karamazov, Detroit Tigers outfielder Dan Brothers, and "Brother, Can You Spare a Dime?" which Marx asserted was originally plural, "but this was spreading a dime pretty thin, so they threw out one brother." Marx asked Jack Warner if he was the first "Jack," citing Jack the Ripper as a possible precursor. Marx told Harry Warner that he had known several Harrys in his life, so Harry Warner might have to relinquish his title as well. Marx concluded his letter with a call for solidarity among "brothers" in the face of attacks from ambitious young lawyers who might seek to cur their creative activities. "We are all brothers under the skin and we'll remain friends till the last reel of 'A Night in Casablanca' goes tumbling over the spool.""
Actually I'm on a 1.2 GHz K7 Sempron/Athlon-XP on a VIA board, with 512 MB RAM and I'm running KDE without any problems. It's not as super speedy as it could be, but it's still not noticeably degraded. Besides, the perks of using a non-MS desktop environment (such as being able to kill processes before the turn of the century) more than make up for the relatively lower computing power.
KDE does not need a gigabyte of memory. Or half a gig. Open up kinfocenter and click the memory tab for a graphical depiction. I never spill over into swap unless I'm doing something particularly intensive alongside my normal activities, like processing large 3d image data with ITK, or running firefox.
You jest, but it's a damn good book. I actively felt, as I was reading it for my Models of Computation class last semester, that it was the best textbook I've had up to that point. It's very precise.
Am I correct in thinking that the issue is that the given proof proves that the wrong system is universal? I read part of the proof but I'm unfamiliar with some of their terminology and models. From what I can tell, the question was whether a particular Turing machine was universal, and the answer was "yes, with some minor modifications", and the prize panel decided that this adaptation fit within the definition of the problem.
Now, my take on this whole thing was that the question is: whether there exists a way to encode* any Turing machine and string as input to this particular machine, such that it accepts if and only if the original machine does. And the answer seems to be, yes, if we can modify the machine's definition so that it's no longer a standard machine but rather one with an infinite repeating pattern on the tape. Well, that's fine, so long as we recognize that it's not the same as the original system. Otherwise, we'd be requiring the input encoding to contain as many bits of the pattern as necessary, and that required number is uncomputable.
* I'm not sure if I'm being too broad in that definition - how would I restrict this to exclude the possibility of the encoding process itself simulating the machine?
> "that for every UTM there exists another Turing Machine (possibly itself, but irrelevant) that can't be simulated on it?"
No, a universal Turing machine (or "The" Universal Turing Machine) can run all Turing machines including itself. Godel's theorem would have more to do with the fact that the UTM is not a decider, but the problem is still Turing-acceptable ("semi-decidable").
However, we can talk about higher-order Turing machines that make use of oracles to decide otherwise uncomputable problems, and then say that these are above the level of ordinary UTMs, but not above higher level UTMs... This is the basis for talking about Turing degrees - if we're given a means of solving one strange problem, what kinds of other things can we do. These other Turing machines cannot be constructed from the strict formal model, however, as the seven-tuple doesn't allow for the notion of an oracle.
Eh, my instinct is to twitch when we're so callously applying the addition operator to objects that are not natural numbers (or reals, etc). Maybe there's some definition or convention I'm unaware of, but I'd restate that as:
"If the set A union B contains an infinite number of elements, and B contains finitely many elements, then A's cardinality must be infinite."
But even that I'm not too sure about, as I know there's a lot more than meets the eye when it comes to set theory.
>> "All it takes is for the data published on the internet to be *structured*" > "All it takes is for the data published on the internet to be ** STOLEN **"
Careful, I heard somewhere that if you publish information to this thing called the web, other people can see it too!
Jeez, it's relatively straightforward to only make available information that you *want* the world to see. If you don't want your mother's maiden name to be public information, take it off your homepage/blog/profile. The only difference here is that information which was *already available* is now more readily accessible.
Slashdot Mirror
A brief overview of the linked page leads me to the impression that there aren't even any vulnerabilities found or mentioned, it's just speculated that it's conceivable that implementations of a file format may be overflowable because the file format has, gasp, a length field. Lame. (no pun)
I am growing more and more towards the believe that Japanese culture is flat out superior to ours in terms of diversity and creativity. I recently watched in full two anime series: "The Law of Ueki" and "The Melancholy of Haruhi Suzumiya", both of which were excellent. Animation is a time-consuming and expensive process, yet the Japanese seem to produce the finest quality artwork in their sleep without even trying. Their stories tend to actually have consistency, meaning, and development, and they don't seem to like compromising the integrity of their work for marketing or network related reasons.
But I think it really dawned on me how far ahead of us they are when I was watching the Culture Day episode of Haruhi, when the main character sang a song at a talent competition. The song they used was composed specifically for the series so it must have just been an afterthought for the production crew, and yet it still blew most modern real American songs away. All the music I've heard in animes just combines an incredible number of influences from all over the world and perfects these elements into something that could never be generated by one of our artists. It makes me very jealous.
Just as I can't believe this article itself made it to the front page. Why the hell did someone think it was newsworthy to state that vulnerabilities are bad and flaws can be exploited? This just in: The NSA keeps secrets, Schneier fears the government, and bugs in hardware platforms can theoretically hurt their users.
Tis ok, it was a good read anyway. Ah, the exploits of cartoon writers...
I miss Hanna and Barbera.
I could very well use that for a poster slogan in my college's chapter of Students for Free Culture. I wonder if quotations like that are copyrightable when in distributed paper form...
Then there are parents who are terrified of all media besides books. A friend of mine claims to only be allowed to play video games of any kind at social LAN parties, and has similar troubles, though not as severe, with television. I do hope that no one in my generation will be so paranoid of the next technology as to govern its consumption in quotas rather reason.
This kind of article reinforces the management perspective mentioned in another recent article that firewalls == security. While most people would attack that equality by saying security is a process and thus much more than firewalls, I never really liked firewalls to begin with. All they are is a place to centralize some rules, which is great but it's not as if they give you any primitive or insight that you didn't already have before. Why block a port when the port isn't even open to begin with - besides centralization and efficiency, from a security perspective what does it offer you, especially when you trust the security policies of all the machines on your internal network?
That's a safety feature. Just as supergerms have self-destruct protein sequences, so they can't live outside a laboratory, so too are scientists protecting us from genetically engineered supermice.
I can't tell if you were being sarcastic, or if that was a call for gender-neutral dialogue, but just in case: Women already have "Motherboard", "Mothership", "Daughter Cells", and personifications of cruise vessels; let the men have this one.
I guess so. I normally don't hop on the Dvorak bashing, probably because I learned of his reputation relatively late, but that was a pretty pathetic piece of speculation. "The gPhone will suck because other phones suck." and "Why search the web when you can talk on the phone" are pretty sucky arguments.
That's the disclosure section of the patent. I was once told that if you can read a patent's claims and understand how the system works without even needing to read the disclosure, then it's a bad patent. In the case of Amazon's recent patent on search, I understood the system just by reading the title alone, indicating that it was an exceptionally poor patent. I was very curious to see what the disclosure would be for such an "invention", but sadly the complete patent is not available online, since it cites by reference application materials that seem to precede the practice of online publication.
> "I'm just saying that this is a poor representation of the case for banning underage smoking, and a poor argument for a change in policy."
Indeed. While I disagree with more than one step in the 6-point argument, it's the last one that I would say is most egregious:
> "Hence, if you believe that smoking should be banned for people under 18, then the same logic would lead to a ban on smoking for people over 18 as well."
because this assumes that we're using the same logic as the scarecrow.
That is quite sinisterly subtle indeed. And that you of all users should be the one to clarify that... Quit messing with my head.
> "Ok, so I messed up the numerical equivalent of milli and the case of the 'k', however as usual around here the point was completely missed in favour of pointing out minor errors."
You specifically quoted those examples and asserted that the "rest of the world", which "geeks" are apparently not part of, uses those terms and definitions. In this case, details are moderately relevant, as you blasted the parent over his deviation from perfect adherence to the standard while demonstrating variations of your own.
> "Have you ever been the only English speaker in a room full of, say, Cantonese? Really difficult to communicate, no? Try explaining your theory to a room full of NON geeks and see what their reactions are. Hint: Usually it's a combination of confusion, bewilderment and bemusement. "Sure, whatever, a small k means a thousand and like twenty eight or something. Whatever you say.""
I'm a bit bewildered by how much importance you attach to this matter. You imply that it is extremely difficult for an average person to comprehend even the *existence* of a technical fact or definition that does not exactly coincide with daily usage, and make it sound as though this would have disastrous consequences. I seriously doubt it would be that hard to explain to most people what a binary kilobyte is. If you wanted to try to be confusing to prove me wrong, you could start off yammering about representing numbers in different bases and quoting various powers of two, and in that case I can imagine their eyes glazing over. If on the other hand you wanted to tell someone that the prefix "kilo" doesn't necessarily mean "thousand" when people talk about disk storage, you could just *say* that and be done with it.
> "The point, and the underlying point to my posting, was that geeks like to re-invent things as convenient to them "
!! Of course, as does anyone with "technical" knowledge in any subject. I fail to see how this is the grievous sin you're making it out to be.
What, you didn't think all words in the English language were used in the same way by all professionals, did you? (By the way, did you know that engineers use the symbol "j" to represent the imaginary unit? Or that a "trusted" party in a security model is one that is capable of breaking your security policy?)
> "but that are completely outside the level of understanding of the rest of the civilized world."
Exaggerate much?
> "It's apparent that a contingent of geeks (and, apparently, lawyers) would like hard drive storage re-defined but refuse to accept any of the standard notations that have been proposed which would do exactly that."
Re-defined? How long have mega, giga, etc., been understood as binary prefixes when applied to hard drive storage? Was there ever a time when this was not the case? It's certainly not a recent development on the part of renegade geeks who aim to destroy society by attacking its language. Perhaps the proposed alternative notations have not been adopted by many because they *suck*. By your logic, we should all suddenly start using the prefixes "kibi", "mebi", and "gibi" in all conversations when we wish to indicate multiples of 1024. You can't possibly expect that to not raise some eyebrows among a crowd of non-geeks; they'll think you have a speech impediment.
> "Mega, kilo, giga, etc. have existed for quite some time both in and out of the computer industry as base-10 notations - even with the measure of network bandwidth, CPU Hz and other forms of measurement but when it comes to disk drive storage capacity suddenly the rules change and an exception is formed?!? Nonsense. Pure nonsense."
Prefixes, like all forms of vocabulary, mean what the speaker and listener agree upon. The English language has loads of exceptions that one could claim are pure nonsense under your logic, but that doesn't make it so. A significant portion of people use those words as binary prefixes, and while you can argue that it's suboptimal, the mere fact that it's a
Where I come from, "k" is an acceptable abbreviation of "kilo", and wikipedia seems to agree: http://en.wikipedia.org/wiki/Si_prefix#List_of_SI_prefixes
Since it does not make an ounce of sense to talk about fractions of a bit unless we're describing averages, entropy, etc., I don't see why we can't reuse the lowercase half of the alphabet for the purpose the parent described. And yes, I believe I've seen it before.
> "{SIGH!} And people wonder why geeks are ostracized."
That statement is strange on several levels.
I'm currently reading "Copyrights and Copywrongs" by Siva Vaidhyanathan (which I'll recommend, although I've only read like, one chapter), and the first example he gives in the introduction is Warner Brothers threatening to sue Groucho Marx for using the film title "A Night in Casablanca".
From page 2:
"Then Marx turned the issue of name ownership on the Warners. He conceded that they could claim control of "Warner," but certainly not "brothers." Marx claimed, "Professionally, we were brothers long before you were." Marc pointed out that even before the Marx Brothers, there were still the Smith Brothers, the Brothers Karamazov, Detroit Tigers outfielder Dan Brothers, and "Brother, Can You Spare a Dime?" which Marx asserted was originally plural, "but this was spreading a dime pretty thin, so they threw out one brother." Marx asked Jack Warner if he was the first "Jack," citing Jack the Ripper as a possible precursor. Marx told Harry Warner that he had known several Harrys in his life, so Harry Warner might have to relinquish his title as well. Marx concluded his letter with a call for solidarity among "brothers" in the face of attacks from ambitious young lawyers who might seek to cur their creative activities. "We are all brothers under the skin and we'll remain friends till the last reel of 'A Night in Casablanca' goes tumbling over the spool.""
Actually I'm on a 1.2 GHz K7 Sempron/Athlon-XP on a VIA board, with 512 MB RAM and I'm running KDE without any problems. It's not as super speedy as it could be, but it's still not noticeably degraded. Besides, the perks of using a non-MS desktop environment (such as being able to kill processes before the turn of the century) more than make up for the relatively lower computing power.
KDE does not need a gigabyte of memory. Or half a gig. Open up kinfocenter and click the memory tab for a graphical depiction. I never spill over into swap unless I'm doing something particularly intensive alongside my normal activities, like processing large 3d image data with ITK, or running firefox.
> "Normally, this would simply mark it as unacceptably low-end for use with modern software."
Damnit, those are better than my specs*. Although I'm using KDE just fine.
* At least, if we accept the clockspeed war mentality. AMD K7 Sempron 1.2 GHz, via chipset.
I hope not. I (seriously) just finished memorizing 09f911029d74e35bd84156c5635688c0. And I've only had the T-shirt for a few months.
You jest, but it's a damn good book. I actively felt, as I was reading it for my Models of Computation class last semester, that it was the best textbook I've had up to that point. It's very precise.
Note that I'm not anonymous.
Am I correct in thinking that the issue is that the given proof proves that the wrong system is universal? I read part of the proof but I'm unfamiliar with some of their terminology and models. From what I can tell, the question was whether a particular Turing machine was universal, and the answer was "yes, with some minor modifications", and the prize panel decided that this adaptation fit within the definition of the problem.
That was a well put and concise explanation.
Now, my take on this whole thing was that the question is: whether there exists a way to encode* any Turing machine and string as input to this particular machine, such that it accepts if and only if the original machine does. And the answer seems to be, yes, if we can modify the machine's definition so that it's no longer a standard machine but rather one with an infinite repeating pattern on the tape. Well, that's fine, so long as we recognize that it's not the same as the original system. Otherwise, we'd be requiring the input encoding to contain as many bits of the pattern as necessary, and that required number is uncomputable.
* I'm not sure if I'm being too broad in that definition - how would I restrict this to exclude the possibility of the encoding process itself simulating the machine?
> "that for every UTM there exists another Turing Machine (possibly itself, but irrelevant) that can't be simulated on it?"
No, a universal Turing machine (or "The" Universal Turing Machine) can run all Turing machines including itself. Godel's theorem would have more to do with the fact that the UTM is not a decider, but the problem is still Turing-acceptable ("semi-decidable").
However, we can talk about higher-order Turing machines that make use of oracles to decide otherwise uncomputable problems, and then say that these are above the level of ordinary UTMs, but not above higher level UTMs... This is the basis for talking about Turing degrees - if we're given a means of solving one strange problem, what kinds of other things can we do. These other Turing machines cannot be constructed from the strict formal model, however, as the seven-tuple doesn't allow for the notion of an oracle.
Please correct me if I'm mistaken about this.
> "This is TRUE for numbers."
Eh, my instinct is to twitch when we're so callously applying the addition operator to objects that are not natural numbers (or reals, etc). Maybe there's some definition or convention I'm unaware of, but I'd restate that as:
"If the set A union B contains an infinite number of elements, and B contains finitely many elements, then A's cardinality must be infinite."
But even that I'm not too sure about, as I know there's a lot more than meets the eye when it comes to set theory.
>> "All it takes is for the data published on the internet to be *structured*"
> "All it takes is for the data published on the internet to be ** STOLEN **"
Careful, I heard somewhere that if you publish information to this thing called the web, other people can see it too!
Jeez, it's relatively straightforward to only make available information that you *want* the world to see. If you don't want your mother's maiden name to be public information, take it off your homepage/blog/profile. The only difference here is that information which was *already available* is now more readily accessible.