I expect that the RSA business will drop, perhaps totally. And some "free-market theorists" will see this as the free market at work. But I will note that this won't result in those who trusted them being made whole. And that there are lots of companies in the wings...one (or more) of them will probably get considerable undercover push towards dominance.
So while I expect that EMC will need to write-off RSA (possibly being recompensed with nice government contracts) I don't expect the customers to benefit. And *I* consider this a failure of the free market.
The interesting part is that they could then be put under a gag order and forbidden from letting anyone know of their relationship to the NSA.
So they could, at best, say "No comment"...unless that could also be forbidden.
Please note that all these restrictions SHOULD come under the heading of "prior restraint", and be ruled unconstitutional. But that will only happen when the government decides to allow it to happen. I don't think any honest foreign company will knowingly do business with any company subject to current laws. Which means that we only do business with those that are greedy liars...and willing to take significant risks under the prodding of greed. Fortunately then numbers of those is not few.
There wasn't any evidence. All that was known was the name. So there's really nothing to look at anew.
I always found their explanations dubious, but possible. I still do. Presumably someone somewhere actually knows, but either they can't prove it, or they don't choose to.
That one needs to stay in the category "suspicious, but undecided" until some real evidence turns up.
I really doubt those numbers. I suspect, e.g., that unemployed people are drastically undercounted. (This has been true in the past. I don't KNOW that it's true this time.) I'm certain that income from criminal activities is underreported. Etc. (There's also racial bias in the counts, and even some political bias...though that's usually accidental. You could say that about racial undercounting, also. Census takers are reluctant to interview truculent people in low-income areas, often due to fears of personal safety.)
IFAIK the census *tries* to be accurate. This doesn't mean that it *is* accurate, or that it doesn't have systematic biases.
He included China. He did, however, admit that he couldn't estimate the fraction above the limit. Still, if he's right, and over 1% of Chinese make over $40,000 (due to inequality in distribution), then many on Slashdot may NOT be a part of the global 1%. China is the dominating factor here, and it's an unknown.
Somehow I don't think you realize how difficult it is to define a good crypto. And how easy it is to end up with a bad one.
One time pads, OTOH, have a lot to reccommend them, if you're in a situation where you can use one. But it was public key that made the web business model possible.
Given enough computational resources, and the knowledge of the algorithm, and depending on the ration of pseudo-random bits to random jumps, this might well be crackable. It just wouldn't be easy. For many purposes it would be sufficient.
Focus a camera on a candle, and overdrive the amplification to where you're amplifying noise. (Does ovedriving the amplification still work? If not you need to pick a set of pixels that are on about half the time, and use them.) Knowing what you're doing, or the algorithm, doesn't help. But, as with any real random generator it requires a hardware assist. (In this case the camera and the candle.)
N.B.: This is just ONE approach. but it's one that lets you generate reasonably large quantities of actually random numbers with easily accessible equipment. If you live near a freeway, you could probably do something similar with a microphone, but as before you need to standardize the numbers against a background. Otherwise you get variation by time of day. Or buy a geiger counter, and have the mic listen to that. But that gives you a lower rate of accumulation.
Thermal noise from an overdriven amplifier is one decent source of noise. But it requires hardware support. So does any other source of true randomness (as opposed to psuedo-randomness),
One method that works in many applications is to use a low bit-rate source of true randomness (disk seek times, keyboard timing, etc. to amplify a pseudorandom source. I think that/dev/urandom sort of does this, but I had in mind repeatedly using/dev/random to initialize a pseudorandom generator for a short run before repeating the initialization. If you use multiple pseudorandom generators this should be nearly unbreakable...but it clearly isn't purely random. Perhaps this is what/dev/urandom does, but that's not what the things I've read imply.
I really doubt that they are complete. The internal electronics are probably broken, plastics deteriorate with age, etc. But the real loss is the skill sets needed to build it. There were a LOT of failures before we got a working version.
FWIW, I believe that NASA has officially said that they couldn't build another Saturn.
Sorry, but what's wrong with the aggregated information? If information is properly aggregated, you can't use it to identify individuals. Perhaps you meant a different term? Only, I can't guess what term you meant.
I doubt that it's a plot, largely because I tend not to believe in plots rather than because it's intrinsicly unlikely. However:
While increasing the income disparity has the long term effect of decreasing the wealth of those at the top, that's only in the long term. In the short term it's to their monetary benefit.
Also, an increased disparity in wealth translates into in increased disparity in social status, and that advantage doesn't evaporate in the long term. (Well, not until the revolution.)
Because of these two factors, it's to the advantage of every individual wealthy AND powerful person to increase the disparity. It's only a long term disadvantage, so it's to his immediate advantage, and even in the long term, it advantageous WRT status. (And because of this, you don't need a plot. It's the natural way for the system to evolve. Plots are only needed to drive it in an unnatural way, such as towards greater social equality. Even there they usually fail drastically, and sometimes bloodily.)
I can accept that you hope that, but I consider this foolish.
OTOH... My chief beef against GMO foods is the way the patent laws are interpreted and enforced. I do think that Monsanto should be put out of business by fair means or foul...and if I thought banning GMO plants outright would to that, I might be in favor of it for THAT reason. It wouldn't, however, so I'm more in favor of requiring extensive safety testing.... And of changing the laws so that the descendants of a plant belong to the farmer that grew them.
FWIW, cows prefer to eat foods that people can't digest. (Except for things like apples, and even those aren't particularly good for cows.) Cows do quite well on grass and alfalfa, and CAN even live for awhile on a mixture of partially sawdust. Clover tends to be too rich for them. (OTOH, they do like a bit of molasses added to their fodder. It helps keep them quiet while you are milking. But don't add too much, or you'll have a sick cow.)
Cows are not horses. Horses compete much more with humans for food. Horses prefer grain. Cows do better with grass and hay.
Right. And that means they can refuse to accept food that doesn't meet their standards. If they were desperate, they couldn't afford to be so picky. (That's oversimplifying, though. It wouldn't be the people in the government that were starving. They'd just be in danger of being overthrown.)
Not quite, but sort of. In many ways Mao Tse Tung was a "Chinese Emperor", even if not of the same dynasty. And the government now has a bit of the appearance of a Mandrinate. How true this is I'm not certain. Certainly changes are in process. Many of the changes have to do with speed of communication more than anything else, however. I wouldn't be surprised to see another Chinese Emperor emerge from the current Mandrinate, and he might be only a figure-head, as many Emperors have been before him.
There's ideology, and then there's what the government does and how it acts. These are rarely closely aligned.
The reason for singling out GMOs is that they can vary more wildly while being apparently the same species. I'm not saying it wouldn't be easy to breed a potatoe that was poisonous. (The wild variant of the potatoe was nearly poisonous, and even the modern potatoe has poisonous leaves. OTOH, the genetics of potatoes is quite complex, so that was a bad example. But it was the first that came to mind.)
No. The basic GUI design of KDE4 is inferior to the design of KDE3. I never had any hardware issues with it, but the design issues caused me to change to Gnome. And caused me to STAY with Gnome until the idiots at Gnome designed, and the distros supported, Gnome3.
At that point KDE4 became more usable than Gnome, but it's still not nearly as good as KDE3 was. It may be better than KDE2 was, I don't remember that clearly enough to be certain.
But it's design issues, not technical issues that are and have been the problems with the GUIs from day 1. (I rather liked Gnome1. I switched to KDE2 because the Gnome release at that time was unstable. I think they were switching to Gnome2, but memory's a bit vague on that point. I remember being impressed that in KDE I didn't need to drop into the command line to unpack a file...but that wasn't enough to decide me. I switched to KDE because the Gnome maintainers were pushing out code that wasn't ready. And removing features.)
It's definitely worth considering, and if you haven't read the original article, you should. I don't agree with ALL of his points. E.g., I'm not at all convinced that an untraceable currency is a bad idea. Other of his points, however, are plausible. And one reason I came here was to see if anyone had and counters.
E.g., an inherently deflationary currency (becoming arbitrarily expensive to generate new values as the easy ones are computed) does seem like a bad idea. What would be better would be a currency where each new value required a fixed amount of effort, and where it would slowly decay in value if it weren't being transferred. (A half-life of 20 years seems about right, if the original creation of the value isn't too difficult.) But actual radio-isotopes aren't desirable, because you don't want it to be dangerous. Also it would be nice if trading it increased it's value. N.B.: I don't say that such a currency is any ideal. Or has any plausibility. But I have my doubts about a currency with a fixed number of tokens of value.
It's also true that malware that mints bitcoins has already been discovered. So that worry about them isn't a hypothesis. One might wonder just how common such malware would become, but if they increase in value, one can expect that it will become more common. So that point of his is plausible.
I think you missed the part about the transition from pre-Singularity to post-Singularity involving a removal of 90% of the population, and THEN experiencing a large die-off.
Yes, what they ended up was was pretty good, though it certainly had it's problems (see "unscheduled criticality excursion", or Rachael's old job), but getting there was miserable. And that's the part we would get to try to live through.
Believe it if you want to. It's not impossible that it's true, but they also aren't noted for being honest with their customers.
I find it more likely that when there was a surge of bad PR, they changed their mind about what they were going to do, but I don't have any proof. And all they're offering as proof is their "honor and good name", of which they don't have much.
Be warned: They have designed their products to allow them to arbitrarily remove things that you have already purchased. Now ask yourself why.
But if they're doing reasonable de-duping, then only the first person to click on the image will register. Everyone else will hit the cache. To avoid this every email would need a separate link to the picture.
Slashdot Mirror
I expect that the RSA business will drop, perhaps totally. And some "free-market theorists" will see this as the free market at work. But I will note that this won't result in those who trusted them being made whole. And that there are lots of companies in the wings...one (or more) of them will probably get considerable undercover push towards dominance.
So while I expect that EMC will need to write-off RSA (possibly being recompensed with nice government contracts) I don't expect the customers to benefit. And *I* consider this a failure of the free market.
The interesting part is that they could then be put under a gag order and forbidden from letting anyone know of their relationship to the NSA.
So they could, at best, say "No comment"...unless that could also be forbidden.
Please note that all these restrictions SHOULD come under the heading of "prior restraint", and be ruled unconstitutional. But that will only happen when the government decides to allow it to happen. I don't think any honest foreign company will knowingly do business with any company subject to current laws. Which means that we only do business with those that are greedy liars...and willing to take significant risks under the prodding of greed. Fortunately then numbers of those is not few.
There wasn't any evidence. All that was known was the name. So there's really nothing to look at anew.
I always found their explanations dubious, but possible. I still do. Presumably someone somewhere actually knows, but either they can't prove it, or they don't choose to.
That one needs to stay in the category "suspicious, but undecided" until some real evidence turns up.
And while they turned off three of them, apparently they didn't turn them back on before leaving. So they weren't hiding that they had been there.
OTOH, they also didn't go out of their way to create a mess. So they weren't police.
But it doesn't stop them from keeping it on a windows machine that isn't properly maintained or isolated.
I really doubt those numbers. I suspect, e.g., that unemployed people are drastically undercounted. (This has been true in the past. I don't KNOW that it's true this time.) I'm certain that income from criminal activities is underreported. Etc. (There's also racial bias in the counts, and even some political bias...though that's usually accidental. You could say that about racial undercounting, also. Census takers are reluctant to interview truculent people in low-income areas, often due to fears of personal safety.)
IFAIK the census *tries* to be accurate. This doesn't mean that it *is* accurate, or that it doesn't have systematic biases.
He included China. He did, however, admit that he couldn't estimate the fraction above the limit. Still, if he's right, and over 1% of Chinese make over $40,000 (due to inequality in distribution), then many on Slashdot may NOT be a part of the global 1%. China is the dominating factor here, and it's an unknown.
Somehow I don't think you realize how difficult it is to define a good crypto. And how easy it is to end up with a bad one.
One time pads, OTOH, have a lot to reccommend them, if you're in a situation where you can use one. But it was public key that made the web business model possible.
It really depends on your application.
Given enough computational resources, and the knowledge of the algorithm, and depending on the ration of pseudo-random bits to random jumps, this might well be crackable. It just wouldn't be easy. For many purposes it would be sufficient.
Focus a camera on a candle, and overdrive the amplification to where you're amplifying noise. (Does ovedriving the amplification still work? If not you need to pick a set of pixels that are on about half the time, and use them.) Knowing what you're doing, or the algorithm, doesn't help. But, as with any real random generator it requires a hardware assist. (In this case the camera and the candle.)
N.B.: This is just ONE approach. but it's one that lets you generate reasonably large quantities of actually random numbers with easily accessible equipment. If you live near a freeway, you could probably do something similar with a microphone, but as before you need to standardize the numbers against a background. Otherwise you get variation by time of day. Or buy a geiger counter, and have the mic listen to that. But that gives you a lower rate of accumulation.
Thermal noise from an overdriven amplifier is one decent source of noise. But it requires hardware support. So does any other source of true randomness (as opposed to psuedo-randomness),
One method that works in many applications is to use a low bit-rate source of true randomness (disk seek times, keyboard timing, etc. to amplify a pseudorandom source. I think that /dev/urandom sort of does this, but I had in mind repeatedly using /dev/random to initialize a pseudorandom generator for a short run before repeating the initialization. If you use multiple pseudorandom generators this should be nearly unbreakable...but it clearly isn't purely random. Perhaps this is what /dev/urandom does, but that's not what the things I've read imply.
I really doubt that they are complete. The internal electronics are probably broken, plastics deteriorate with age, etc. But the real loss is the skill sets needed to build it. There were a LOT of failures before we got a working version.
FWIW, I believe that NASA has officially said that they couldn't build another Saturn.
Sorry, but what's wrong with the aggregated information? If information is properly aggregated, you can't use it to identify individuals. Perhaps you meant a different term? Only, I can't guess what term you meant.
I doubt that it's a plot, largely because I tend not to believe in plots rather than because it's intrinsicly unlikely. However:
While increasing the income disparity has the long term effect of decreasing the wealth of those at the top, that's only in the long term. In the short term it's to their monetary benefit.
Also, an increased disparity in wealth translates into in increased disparity in social status, and that advantage doesn't evaporate in the long term. (Well, not until the revolution.)
Because of these two factors, it's to the advantage of every individual wealthy AND powerful person to increase the disparity. It's only a long term disadvantage, so it's to his immediate advantage, and even in the long term, it advantageous WRT status. (And because of this, you don't need a plot. It's the natural way for the system to evolve. Plots are only needed to drive it in an unnatural way, such as towards greater social equality. Even there they usually fail drastically, and sometimes bloodily.)
I can accept that you hope that, but I consider this foolish.
OTOH... ... And of changing the laws so that the descendants of a plant belong to the farmer that grew them.
My chief beef against GMO foods is the way the patent laws are interpreted and enforced. I do think that Monsanto should be put out of business by fair means or foul...and if I thought banning GMO plants outright would to that, I might be in favor of it for THAT reason. It wouldn't, however, so I'm more in favor of requiring extensive safety testing.
FWIW, cows prefer to eat foods that people can't digest. (Except for things like apples, and even those aren't particularly good for cows.) Cows do quite well on grass and alfalfa, and CAN even live for awhile on a mixture of partially sawdust. Clover tends to be too rich for them. (OTOH, they do like a bit of molasses added to their fodder. It helps keep them quiet while you are milking. But don't add too much, or you'll have a sick cow.)
Cows are not horses. Horses compete much more with humans for food. Horses prefer grain. Cows do better with grass and hay.
Right. And that means they can refuse to accept food that doesn't meet their standards. If they were desperate, they couldn't afford to be so picky. (That's oversimplifying, though. It wouldn't be the people in the government that were starving. They'd just be in danger of being overthrown.)
Not quite, but sort of. In many ways Mao Tse Tung was a "Chinese Emperor", even if not of the same dynasty. And the government now has a bit of the appearance of a Mandrinate. How true this is I'm not certain. Certainly changes are in process. Many of the changes have to do with speed of communication more than anything else, however. I wouldn't be surprised to see another Chinese Emperor emerge from the current Mandrinate, and he might be only a figure-head, as many Emperors have been before him.
There's ideology, and then there's what the government does and how it acts. These are rarely closely aligned.
The reason for singling out GMOs is that they can vary more wildly while being apparently the same species. I'm not saying it wouldn't be easy to breed a potatoe that was poisonous. (The wild variant of the potatoe was nearly poisonous, and even the modern potatoe has poisonous leaves. OTOH, the genetics of potatoes is quite complex, so that was a bad example. But it was the first that came to mind.)
No. The basic GUI design of KDE4 is inferior to the design of KDE3. I never had any hardware issues with it, but the design issues caused me to change to Gnome. And caused me to STAY with Gnome until the idiots at Gnome designed, and the distros supported, Gnome3.
At that point KDE4 became more usable than Gnome, but it's still not nearly as good as KDE3 was. It may be better than KDE2 was, I don't remember that clearly enough to be certain.
But it's design issues, not technical issues that are and have been the problems with the GUIs from day 1. (I rather liked Gnome1. I switched to KDE2 because the Gnome release at that time was unstable. I think they were switching to Gnome2, but memory's a bit vague on that point. I remember being impressed that in KDE I didn't need to drop into the command line to unpack a file...but that wasn't enough to decide me. I switched to KDE because the Gnome maintainers were pushing out code that wasn't ready. And removing features.)
It's definitely worth considering, and if you haven't read the original article, you should. I don't agree with ALL of his points. E.g., I'm not at all convinced that an untraceable currency is a bad idea. Other of his points, however, are plausible. And one reason I came here was to see if anyone had and counters.
E.g., an inherently deflationary currency (becoming arbitrarily expensive to generate new values as the easy ones are computed) does seem like a bad idea. What would be better would be a currency where each new value required a fixed amount of effort, and where it would slowly decay in value if it weren't being transferred. (A half-life of 20 years seems about right, if the original creation of the value isn't too difficult.) But actual radio-isotopes aren't desirable, because you don't want it to be dangerous. Also it would be nice if trading it increased it's value.
N.B.: I don't say that such a currency is any ideal. Or has any plausibility. But I have my doubts about a currency with a fixed number of tokens of value.
It's also true that malware that mints bitcoins has already been discovered. So that worry about them isn't a hypothesis. One might wonder just how common such malware would become, but if they increase in value, one can expect that it will become more common. So that point of his is plausible.
Etc.
I think you missed the part about the transition from pre-Singularity to post-Singularity involving a removal of 90% of the population, and THEN experiencing a large die-off.
Yes, what they ended up was was pretty good, though it certainly had it's problems (see "unscheduled criticality excursion", or Rachael's old job), but getting there was miserable. And that's the part we would get to try to live through.
Yes. And I've stopped buying PC games. Funny thing about that.
Believe it if you want to. It's not impossible that it's true, but they also aren't noted for being honest with their customers.
I find it more likely that when there was a surge of bad PR, they changed their mind about what they were going to do, but I don't have any proof. And all they're offering as proof is their "honor and good name", of which they don't have much.
Be warned: They have designed their products to allow them to arbitrarily remove things that you have already purchased. Now ask yourself why.
But if they're doing reasonable de-duping, then only the first person to click on the image will register. Everyone else will hit the cache. To avoid this every email would need a separate link to the picture.