512 bits made from 2 hashes, one weak and one strong will be weaker than a single 512 bit hash from the stronger algorithm.
True. However, using 2 different algorithms that are not known to be weak is probably stronger than using a single algorithm that is not known to be weak but produces twice as many bits.
This follows from the fact that similar methods will be used to generate all of the bits in the latter case, therefore if there is some systematic flaw it is reasonably likely to apply to all of the bits. Whereas in the former case, you'd have to find 2 systematic flaws to get you as far (assuming that the algorithms used to generate them were dissimilar, and therefore unlikely to both contain the same flaw).
Well, I'm not up on the latest virus developments, but AFAIK most viruses exploit known vulnerabilities rather than trying to crack algorithmically hard problems. Maybe there's some subtle way a virus could use a powerful P-time factorizer, but I think the author really just stated "hackers and viruses" because it sounded nice.
Neither am I, but if you asked me to hazard a guess, I would probably say that investigating the code of an operating system in order to find security flaws with it is an NP problem. Being able to do this automatically in a reasonable time frame would assist virus writers. It would also, of course, assist operating system designers.
How is it any easier to produce a file that will cause a collision in MD5 when I hash it in MD5 after an arbitrary (and unknown to the attacker) string has been hashed?
and has seriously considered reducing the burden of proof for serious offences to "balance of probabilities".
You know, I'd forgotten about this. You're talking about the idea that was suggested for last year's Sexual Offences Act that in cases of alleged rape, it should be up to the alleged offender to show that consent was given (a task that is nearly impossible in most cases), aren't you? I take it that didn't make it to the final version of the act...?
CCTV is also frequently of such exceptionally poor quality that facial recognition is all but impossible. Typically we're shown a grainy black and white and asked if we know anyone who was wearing a dark top with a white stripe across it.
Do bare in mind that we're only asked to identify the people that the police can't recognise themselves. Given that the local police generally know 90% or more of the criminals operating in their area, this means that you only see the 10% of unknown criminals mixed with only the instances from those 90% where the CCTV footage is too poor for the police to look at it and immediately say, "Isn't that Joe Foster from the old tower block on Lebanon Close?"
Gun bans are not a major obstacle in the way of a revolutionary. When you commit revolution, you will always break laws. Breaking the gun ban is minor compared to the ban on killing government officials.
That argument doesn't work. Sure, a revolutionary is going to break laws, but that isn't the only thing preventing a revolution. Arranging them is difficult. Having to acquire your weapons from a source outside of your own country and then import them makes them harder, and increases the chance that you will be noticed before you manage to stage your coup.
Absolutley nobody carried (legitimately) a handgun as a crime deterrent and anybody waving a legally-held handgun at a mugger would find themselves locked up pretty quickly.
Are you sure? I know somebody who did exactly that, and got away with it. What do you think's going to happen? Mugger walks into police station and says, "I was mugging this guy and he pulled a gun on me!"?
What kind of equipment is needed to achieve the necessary Disk I/O to match the network throughput?
I suspect that a 16 disk RAID 0 array of high speed disks could keep up with this. However, I think you'd need a pretty specialised computer system to keep up with such an array. I'm not sure what kind of architecture you'd need, but I am convinced it would involve multiple I/O buses and some kind of crossbar arrangement for shared access between DMA controllers. Memory would have to be interleaved carefully also.
Chances are they were transferring localhost:/dev/zero to remote:/dev/null.
Would a Next Cube be considered a PC? I would, because it's present day brother would be my PowerMac G5.
Just because you're using an operating system that is a close relative in terms of technology used doesn't make the computers the same category.
UNIX was an operating system originally developed for minicomputers. This doesn't mean that my IA32 *BSD machine is a minicomputer -- it _is_ a PC.
Although, that said, I've never used a Next Cube, but I understand they were marketed for single user desktop use, which in my book means they are PCs (or at least, personal computers).
Huh? Who says you have to fling them out in an even distribution? Why not send those 100 rocks out towards the 100 nearest stars?
Well, why not use a focused EM signal pointed toward the 100 nearest stars? Exactly the same argument applies. If you focus it tightly enough you will lose very little power.
Your rocks, by the way, will either make it or not make it (perhaps because they crash into something else orbiting the destination star before they're noticed). Admittedly, the star's gravity means you have a large target to aim for.
It is in fact highly unlikely that a signal we received at this stage from outside of our star system was broadcast omnidirectionally. To broadcast the feintest signal we can detect at that kind of range omnidirectionally would take a _huge_ amount of power, probably a significant fraction of the output of a small star.
Whereas with a tightly focussed signal, the amount of power required is much lower.
I look at a squid or octopus, and I question the claim that evolution in water necessarily favors "the mouth at one end and the anus at the other."
They do fill an unusual niche. In order to do so, they require a lot of extraordinary capabilities -- they are quite intelligent, very flexible and many species have remarkable camouflage abilities. I would hazard a guess that at least some of these are necessary preconditions to their body shape being viable. It is remarkably good for a certain type of 'hide & grab' hunting. It isn't very good at escaping potential predators. They have to hide, instead, using a combination of the above abilities.
I'd be interested to see a history of the evolution of these creatures, to see how close I am to reality. I don't know how much of this is known?
Also the HF would quickly dissolve various minerals etc. and form insoluble salts thereby sequestering all the F out of the atmosphere and into the lithosphere in solid bonded form in a short (geological) time period.
You could say the same thing about O2. There would have to be some kind of ongoing reaction to produce it, as there is on Earth to produce oxygen. We call it "life".
(Seriously: simple life lives off available heat and eventually evolves to use whatever energy store it can find. Sooner or later, it would start releasing that stored fluorine into the atmosphere.)
Of course CGTAGTAGTAAT.... isn't going to be too useful to them, unless we include the molecular structure of the bases
I think they would also need to know the structure of the key enzymes that are involved in the process of actually using the stuff. I don't think DNA/RNA by itself is adequate to form life -- you need most of the rest of the components of a cell nucleus to actually make it useful.
IIRC the general conclusion was that it was an urban myth. Certainly, nobody could track down the researchers quoted in the article, and there were plenty of counterexamples that followed the rules as specified but were still hard to read.
If you drew the letters PLEORBM in a Scrabble game, it might take a while to see the word staring at you.
I dunno, I got that pretty quickly. Short anagrams are easy. Scrabble's hard because you have to use a tile from the board too, which means you're not looking only in one place.
I dug through their CSS and this is an example of how the font is set - font:70% Tahoma, Helvetica. This is ok CSS but it is a matter of what the browser is considering the parent size. Obviously Firefox and IE don't agree. I don't have any other browser besides those two on this machine so I can't common on how other browser handle the page.
I think (by default) they both use the same base size. The big problem is, I think, that IE interprets the percentage as being a percentage of the enclosing or tag's size, whereas Gecko interprets it as being a percentage of the enclosing tag, whatever that happens to be. So with IE the text remains the same size, but with Gecko browsers the text gets progressively smaller as it is nested more and more deeply in structural markup.
I believe this to be an IE bug, but am not certain.
The relationship is probably a lot more complicated than "thought comes before language". I suspect they are both highly dependent on each other.
For instance, it is clear that many non-verbal animals are able to think, in at least some limited fashion. Larger rodents, for instance, are able to build models of their world and solve simple problems (not limited to learning by trial and error). It is exactly this kind of modelling that concepts like one object being inside another stem from -- spatial reasoning is almost certainly the most deeply embedded and instinctive part of thought, and therefore the least likely to depend on language.
However, the ability to form complex theories and plans may or may not be entirely dependent on our ability to express them. Could primitive man, for instance, have looked at the weather and decided whether it would be best to go hunting today or finish building that shelter first, if he didn't have words for 'rain', 'shelter', and 'later'? The question might be too complex to approach without some kind of symbolism that can be internalised. Or it might not. Its very hard to tell.
Is right to left, or left to right the best way to go.
I remember reading about an interesting study into this. Apparently, there are a small number of people who have a particular form of brain damage which effectively reverses their perception. These people, if they were originally educated to read/write left to right, would afterwards naturally read/write right to left, or vice versa.
Apparently, once they get used to using their right hand with a style similar to that a left-hander would use (or vice-versa) they can read & write in the opposite direction at roughly the same rate a normal person can in the usual direction. The conclusion: the difference is not noticeable; neither left to right nor right to left is substantially more efficient (or any difference is also negated by the brain damage these people have suffered).
No, I can't cite references. I just came across it about 10 years ago, I don't even remember what I was studying at the time.
1. I want my old input method back. Many years ago, I learnt to use a chord based system where 5 buttons could be used to enter all letters and common punctuation, and a shift mode could be used for digits and other less common items. Failing that, an input mechanism that has been specifically designed for use on a small area so that I'm not constantly repositioning the pen. Unistrokes is supposedly quite good at this, but I've not found a useful implementation so far. Graffiti isn't good enough -- there isn't enough opportunity to use right-to-left strokes.
2. Programability. Developing software for it should be _easy_. Palms are currently better for this than WinCE based systems, because they have a much simpler UI (which is all you need, really), but the motorola emulation and stupid handling of global data[1] are annoying backward-compatibility features that I would like to see dropped.
3. Expandability. There should be a variety of reasonably priced expansion modules that can be connected easily to the device. They should probably hook onto it, making it slightly thicker.
-- Note 1: you can only access your apps global data in certain circumstances, which means that writing an object-oriented framework for certain kinds of task is impossible.
I might actually take my laptop to Rittenhouse square and chill out staring at the chicks and playing doom 3 (hey i need all the light i can get to see in that game).
As good as 3D graphics hardware is, it has not (yet) reached the point where light falling on the screen is reflected by the objects being rendered. Sorry.
Slashdot Mirror
512 bits made from 2 hashes, one weak and one strong will be weaker than a single 512 bit hash from the stronger algorithm.
True. However, using 2 different algorithms that are not known to be weak is probably stronger than using a single algorithm that is not known to be weak but produces twice as many bits.
This follows from the fact that similar methods will be used to generate all of the bits in the latter case, therefore if there is some systematic flaw it is reasonably likely to apply to all of the bits. Whereas in the former case, you'd have to find 2 systematic flaws to get you as far (assuming that the algorithms used to generate them were dissimilar, and therefore unlikely to both contain the same flaw).
Well, I'm not up on the latest virus developments, but AFAIK most viruses exploit known vulnerabilities rather than trying to crack algorithmically hard problems. Maybe there's some subtle way a virus could use a powerful P-time factorizer, but I think the author really just stated "hackers and viruses" because it sounded nice.
Neither am I, but if you asked me to hazard a guess, I would probably say that investigating the code of an operating system in order to find security flaws with it is an NP problem. Being able to do this automatically in a reasonable time frame would assist virus writers. It would also, of course, assist operating system designers.
Your explanation makes no sense to me.
How is it any easier to produce a file that will cause a collision in MD5 when I hash it in MD5 after an arbitrary (and unknown to the attacker) string has been hashed?
and has seriously considered reducing the burden of proof for serious offences to "balance of probabilities".
You know, I'd forgotten about this. You're talking about the idea that was suggested for last year's Sexual Offences Act that in cases of alleged rape, it should be up to the alleged offender to show that consent was given (a task that is nearly impossible in most cases), aren't you? I take it that didn't make it to the final version of the act...?
CCTV is also frequently of such exceptionally poor quality that facial recognition is all but impossible. Typically we're shown a grainy black and white and asked if we know anyone who was wearing a dark top with a white stripe across it.
Do bare in mind that we're only asked to identify the people that the police can't recognise themselves. Given that the local police generally know 90% or more of the criminals operating in their area, this means that you only see the 10% of unknown criminals mixed with only the instances from those 90% where the CCTV footage is too poor for the police to look at it and immediately say, "Isn't that Joe Foster from the old tower block on Lebanon Close?"
Gun bans are not a major obstacle in the way of a revolutionary. When you commit revolution, you will always break laws. Breaking the gun ban is minor compared to the ban on killing government officials.
That argument doesn't work. Sure, a revolutionary is going to break laws, but that isn't the only thing preventing a revolution. Arranging them is difficult. Having to acquire your weapons from a source outside of your own country and then import them makes them harder, and increases the chance that you will be noticed before you manage to stage your coup.
Absolutley nobody carried (legitimately) a handgun as a crime deterrent and anybody waving a legally-held handgun at a mugger would find themselves locked up pretty quickly.
Are you sure? I know somebody who did exactly that, and got away with it. What do you think's going to happen? Mugger walks into police station and says, "I was mugging this guy and he pulled a gun on me!"?
What kind of equipment is needed to achieve the necessary Disk I/O to match the network throughput?
I suspect that a 16 disk RAID 0 array of high speed disks could keep up with this. However, I think you'd need a pretty specialised computer system to keep up with such an array. I'm not sure what kind of architecture you'd need, but I am convinced it would involve multiple I/O buses and some kind of crossbar arrangement for shared access between DMA controllers. Memory would have to be interleaved carefully also.
Chances are they were transferring localhost:/dev/zero to remote:/dev/null.
Would a Next Cube be considered a PC? I would, because it's present day brother would be my PowerMac G5.
Just because you're using an operating system that is a close relative in terms of technology used doesn't make the computers the same category.
UNIX was an operating system originally developed for minicomputers. This doesn't mean that my IA32 *BSD machine is a minicomputer -- it _is_ a PC.
Although, that said, I've never used a Next Cube, but I understand they were marketed for single user desktop use, which in my book means they are PCs (or at least, personal computers).
Huh? Who says you have to fling them out in an even distribution? Why not send those 100 rocks out towards the 100 nearest stars?
Well, why not use a focused EM signal pointed toward the 100 nearest stars? Exactly the same argument applies. If you focus it tightly enough you will lose very little power.
Your rocks, by the way, will either make it or not make it (perhaps because they crash into something else orbiting the destination star before they're noticed). Admittedly, the star's gravity means you have a large target to aim for.
It is in fact highly unlikely that a signal we received at this stage from outside of our star system was broadcast omnidirectionally. To broadcast the feintest signal we can detect at that kind of range omnidirectionally would take a _huge_ amount of power, probably a significant fraction of the output of a small star.
Whereas with a tightly focussed signal, the amount of power required is much lower.
I look at a squid or octopus, and I question the claim that evolution in water necessarily favors "the mouth at one end and the anus at the other."
They do fill an unusual niche. In order to do so, they require a lot of extraordinary capabilities -- they are quite intelligent, very flexible and many species have remarkable camouflage abilities. I would hazard a guess that at least some of these are necessary preconditions to their body shape being viable. It is remarkably good for a certain type of 'hide & grab' hunting. It isn't very good at escaping potential predators. They have to hide, instead, using a combination of the above abilities.
I'd be interested to see a history of the evolution of these creatures, to see how close I am to reality. I don't know how much of this is known?
Also the HF would quickly dissolve various minerals etc. and form insoluble salts thereby sequestering all the F out of the atmosphere and into the lithosphere in solid bonded form in a short (geological) time period.
You could say the same thing about O2. There would have to be some kind of ongoing reaction to produce it, as there is on Earth to produce oxygen. We call it "life".
(Seriously: simple life lives off available heat and eventually evolves to use whatever energy store it can find. Sooner or later, it would start releasing that stored fluorine into the atmosphere.)
Of course CGTAGTAGTAAT.... isn't going to be too useful to them, unless we include the molecular structure of the bases
I think they would also need to know the structure of the key enzymes that are involved in the process of actually using the stuff. I don't think DNA/RNA by itself is adequate to form life -- you need most of the rest of the components of a cell nucleus to actually make it useful.
IIRC the general conclusion was that it was an urban myth. Certainly, nobody could track down the researchers quoted in the article, and there were plenty of counterexamples that followed the rules as specified but were still hard to read.
Actually, MS's typography unit are pretty good.
It's a shame the rest of MS doesn't actually listen to what they say, but...
If you drew the letters PLEORBM in a Scrabble game, it might take a while to see the word staring at you.
I dunno, I got that pretty quickly. Short anagrams are easy. Scrabble's hard because you have to use a tile from the board too, which means you're not looking only in one place.
Gah. Switch to extrans. Use Preview. Etc.
That should have been "enclosing <body> or <table> tag's size".
I dug through their CSS and this is an example of how the font is set - font:70% Tahoma, Helvetica. This is ok CSS but it is a matter of what the browser is considering the parent size. Obviously Firefox and IE don't agree. I don't have any other browser besides those two on this machine so I can't common on how other browser handle the page.
I think (by default) they both use the same base size. The big problem is, I think, that IE interprets the percentage as being a percentage of the enclosing or tag's size, whereas Gecko interprets it as being a percentage of the enclosing tag, whatever that happens to be. So with IE the text remains the same size, but with Gecko browsers the text gets progressively smaller as it is nested more and more deeply in structural markup.
I believe this to be an IE bug, but am not certain.
The relationship is probably a lot more complicated than "thought comes before language". I suspect they are both highly dependent on each other.
For instance, it is clear that many non-verbal animals are able to think, in at least some limited fashion. Larger rodents, for instance, are able to build models of their world and solve simple problems (not limited to learning by trial and error). It is exactly this kind of modelling that concepts like one object being inside another stem from -- spatial reasoning is almost certainly the most deeply embedded and instinctive part of thought, and therefore the least likely to depend on language.
However, the ability to form complex theories and plans may or may not be entirely dependent on our ability to express them. Could primitive man, for instance, have looked at the weather and decided whether it would be best to go hunting today or finish building that shelter first, if he didn't have words for 'rain', 'shelter', and 'later'? The question might be too complex to approach without some kind of symbolism that can be internalised. Or it might not. Its very hard to tell.
Is right to left, or left to right the best way to go.
I remember reading about an interesting study into this. Apparently, there are a small number of people who have a particular form of brain damage which effectively reverses their perception. These people, if they were originally educated to read/write left to right, would afterwards naturally read/write right to left, or vice versa.
Apparently, once they get used to using their right hand with a style similar to that a left-hander would use (or vice-versa) they can read & write in the opposite direction at roughly the same rate a normal person can in the usual direction. The conclusion: the difference is not noticeable; neither left to right nor right to left is substantially more efficient (or any difference is also negated by the brain damage these people have suffered).
No, I can't cite references. I just came across it about 10 years ago, I don't even remember what I was studying at the time.
a lot of them seem to spell things like "omg, wtf j00 l00s3r?! pwn3d!!!!11!
____________________
Watch Shitty Kung Fu Movie Clips
I fail to see the relevance of kung fu movie clips. Why is that there?
I'm sorry, they've got patents on launching stuff out of a tube by pumping compressed air into it?
Prior art?
1. I want my old input method back. Many years ago, I learnt to use a chord based system where 5 buttons could be used to enter all letters and common punctuation, and a shift mode could be used for digits and other less common items. Failing that, an input mechanism that has been specifically designed for use on a small area so that I'm not constantly repositioning the pen. Unistrokes is supposedly quite good at this, but I've not found a useful implementation so far. Graffiti isn't good enough -- there isn't enough opportunity to use right-to-left strokes.
2. Programability. Developing software for it should be _easy_. Palms are currently better for this than WinCE based systems, because they have a much simpler UI (which is all you need, really), but the motorola emulation and stupid handling of global data[1] are annoying backward-compatibility features that I would like to see dropped.
3. Expandability. There should be a variety of reasonably priced expansion modules that can be connected easily to the device. They should probably hook onto it, making it slightly thicker.
--
Note 1: you can only access your apps global data in certain circumstances, which means that writing an object-oriented framework for certain kinds of task is impossible.
I might actually take my laptop to Rittenhouse square and chill out staring at the chicks and playing doom 3 (hey i need all the light i can get to see in that game).
As good as 3D graphics hardware is, it has not (yet) reached the point where light falling on the screen is reflected by the objects being rendered. Sorry.