In the last 200 years, sea level has been about 30cm higher than today. About 1970, something remarkable happened; sea level fell down to its present position.
There was not a global 30cm drop in seal level around 1970, which would be quite noticeable.
Local mean sea level varies quite a bit due to geological factors and local weather effects such as atmospheric pressure. This produces statistical variation on various scales; individual locations might well see contrary trends; even aggregate trends smoothed over three year weighted averages tend to have considerable noise.
In any case, an island like the one in the video is a poor choice as a benchmark because mean sea level varies across the Pacific by as much as 60cm at any given time due to atmospheric effects. The El Nino/Southern Oscillation could well produce dramatic shifts in local MSL on an island like this. Pressure driven changes in MSL in this region can reach 30cm or in rare cases even more.
Well, I agree with your analysis of the economic purpose of copyright. However, in the US, at least, the power to grant copyright is not limited to the intent for which the Constitution grants that power to Congress. Were that so, then copyright extension would be unconstitutional. Since there is no way to send information back in time, there is no way to incent authors of pre-existing works by copyright extension.
The state of the law, as I understand it, is that the bit about "To promote the progress of science and useful arts" is considered "dicta", in other words an extraneous expression of opinion that is not binding. Once the power is granted, then Congress can use it for any purposes, even to retard the advancement of science. Furthermore, this means that the term of "limited times" mentioned in the next clause is unlimited, at least by considerations of economics. I do not think the state of the law is clear on this; the Constitution clearly states there must be limits but doesn't set a limit on those limits. Possibly a bald faced attempt to establish de facto perpetual copyright, say by specifying terms of millions of years, would fail Constitutional scrutiny, but extensions with the pretense of justification, say bringing terms into agreement with international agreements might.
Well... you are supposed to make a good faith effort to describe prior art in your patent application.
This makes the existence of such a "business method" questionable. An infringement suit under a patent obtained in the manner described amounts to using the court for purposes of defrauding the defendant. The only way to argue such a suit is to deliberately lie to the court in your arguments. If this business method patent were granted, I don't see how you could claim infringement damages under it, because you can't make money legally from patent fraud. Furthermore, you couldn't legally license this "invention", because you'd be contributing to fraud.
So, I don't see any value to a patent like this, unless the patent office and the courts want to advertise that they are simply instruments of fraud. I think this patent reflects the "when in doubt it's safer to patent anything that pops into your head" school of thought.
This is basic Operating Systems 101 stuff: throughput is not the same as responsiveness. Nor does average performance reflect how often you find yourself with not enough performance.
Once you have enough performance, say faster UI response than you can perceive as one example, getting more performance helps your average but is not that much of a practical benefit. So given two systems with the same average performance, if that average is adequate, then the system with less variance is preferable.
I'd say that Vista's average responsiveness is fine. It's throughput is fine. It's just perceptibly variable in performance on a lot of hardware. On hardware where there are occasional performance "glitches", a slightly less responsive operating system that was absolutely consistent would be perceived as better by users.
in all fairness McCain had lots of positive messages but they were flatly refused to be reported.
Actually, I think McCain was at fault here. I have nothing against negative advertising per se; it's perfectly fair, provided the attacks are accurate. If attacks are not accurate, then they'd better work.
McCain's attacks were too poorly focused. He should have focused mainly on Obama's relative inexperience, and built on that. Without a point of focus none of the negative impressions McCain wanted to create had a chance to gel. Their factual weakness made them treacherous to McCain. In 2000, many of Bush's negative ads were factually weak, but it didn't matter because they worked.
Note, however, that more favorable coverage is not the same thing as bias. For example, in covering the debate between advocates of intelligent design and advocates of evolution, "balanced" coverage -- coverage that is not favorable to one side or the other -- is biased, because intelligent design is not science. This is not to deny that bias might have existed in election coverage, but that isn't the only source of disparity between the treatments of the candidates.
One important source in this case was the quality of the campaigns. Obama ran a superb campaign. It was organized, disciplined and consistently on-message. McCain's campaign was none of these things. They kept searching for a new message, then circling back to ones that weren't working, like the Ayers issue. They could have raised Ayers again if momentum was swinging their way, but it wasn't; it was just an issue that hadn't stuck that they they were stuck on because they didn't know how to swing the election back their way. This lack of focus created a vacuum into which negative coverage expanded.
McCain himself couldn't stick to the script, and had to cut off press access, which is bad for a candidate who based his career on accessibility. Palin's lack of polish really undermined McCain's strongest issue in this election: even Obama supporters have to admit it would be better if he had a full term in the Senate under his belt.
This was a Democratic year; to overcome that, McCain's campaign needed to put together a message that resonated, and slowly dig itself out of the hole over the course of weeks. Obama showed how to do this. He started in a hole against Hilary Clinton, and his campaign demonstrated the staying power to wear down her lead over months and months.
McCain isn't like that; he's mercurial, given to dramatic gestures and sudden improvisations. That might work in an even year, but not this year. None of the big things he did that were supposed to sway the election, such as selecting Palin or "suspending" his campaign, had staying power to carry him through to election day.
This election was most emphatically not McCain's to lose. It was Obama's, and the McCain campaign simply failed to seize the initiative. Obama was vulnerable, but McCain's campaign was simply not able to put those vulnerabilities into focus. The press did not snatch victory from McCain's grasp; he just never put himself in a position to grasp victory. His poor press coverage simply reflected this. A well run campaign, say Bush's 2000 campaign, determines what the press is covering and how it is covering it.
The law of the sea supposedly governs this kind of thing. Determining the extent of the continental shelf can extend the exclusive national right for minerals up to 150 nm past the EEZ, so in theory the documentation of the shelf should be a benign action. But ultimately international law is enforced by warfare on various scales of intensity, starting at diplomatic sanctions, through economic sanctions, and all the way up as high as warfare can go.
International law is only what you can force a country to accept as international law. We know Russia wants to claim these resources, and gainsaying them can lead to armed conflict.
Back in 1991, I remarked that the course of the twenty first century would be determined by the integration of the former Soviet states into the world political, security and economic systems. The opportunity to do this, if it ever existed, was bungled by the first Bush administration, and now we are dealing with a militarily powerful, mineral rich nation with a paranoid persecution complex and authoritarian instincts. Do they have more invested in stability than they can get out of grabbing territory?
Well, the fact that they don't want to build a fiber network doesn't mean they aren't selling network services in the city. They might be selling DSL or cable Internet, and the city fiber infrastructure would certainly put that business out of service.
It gets down to philosophical differences about economic value and the role of government. According to one point of view, the government should above all do no harm to any business. If a business wants to sell government weather data, the government ought to make it hard for individuals to get the data directly. By the same token, if a private company wants to provide network services in an area, the government has no business providing better or cheaper services.
The deleterious effect of static stretching on muscle power has been known for years.
It's not a matter of static stretching being "bad for you", what's "bad for you" depends on context. Static stretching is a developmental exercise. You wouldn't go to the weight room for serious strength training before a competition, and the same applies to static stretching.
Well coached athletes have been doing the kind of warm-up exercises described in the article for years, it's just that the word hasn't trickled down.
Well, you have to make a distinction between what an individual "needs" to survive and what a species "needs" to support certain evolutionary trends.
There is seems to be considerable scientific opinion to the effect that meat eating played an important role in making the large brained, linguistically gifted and tool making species H. sapiens possible. However, this doesn't mean that individual humans or pre-humans have ever "needed" meat to survive, or that eating meat to any degree was more "healthy" as we'd define it.
Humans in primitive conditions no doubt ate opportunistically. If you're a primitive person out hunting, you aren't going to ignore a bush full of berries because today's hunting day. Edible herbs probably went straight into the mouth without a second thought. So I expect people were constantly grazing on plants. On the other hand, they'd also gorge themselves on meat when given a chance -- check out the huge prehistoric shell middens they left behind. And of course, when you kill a mammoth, you'd better have a refrigerator for all that meat.
And, of course, humans didn't have refrigerators; they just became extremely adept at turning those meat calories into fat. They had to store calories, to support their very thermally expensive brains.
A modern person, of course, can graze on herbs and roots all the time. It's probably healthier for him too, although regular, small inclusions of meat in his diet do no harm and simplify getting the amino acids he needs.
Well, I don't doubt there was an element of political cowardice, but that's not the whole story. People were anxious to do something, and it's not as if every idea in the Patriot Act is bad. There's a lot of stuff that could be useful in there.
The Act is a typical case of a product rushed out the door without a sufficient requirements analysis.
For example, it has muddled priorities. A lot of the more objectionable features are built around the assumption that responding to a ticking time bomb is the highest priority. If, indeed, it comes to that, it certainly will be the highest priority, but the highest priority before it comes to that is keeping it from coming to that.
Even so, stipulating what can be done in such scenarios is probably a good idea, but those stipulations should be carefully bounded. That's the kind of talk that would go over like a lead balloon back in 2002, but it's true.
This leads us to another sign of a rush job: inattention to unintended consequences.
Probably every new investigatory power in the act would be justifiable in some highly specific circumstances. Just like bashing in somebody's front door is justifiable in certain high specific circumstances. But that shouldn't amount to carte blanche for door bashing. The powers granted should come with accountability for their use, and this is where the Patriot Act fails.
A serious minded civil libertarian could go over the Act and find better ways to accomplish everything the Act is supposed to do. The real problem was that the administration wanted -- had always wanted really -- a completely free hand unfettered by any kind of oversight.
Yes, we're on the same page when it comes to crypto digest functions in this application.
However the definitional issue is more than pedantry. Any kind of practical large scale document indexing system requires the calculation of hash functions, and in general functions useful for cryptography might not be useful for indexing and vice versa.
A non-injective hash function is not "bad"; it's just not useful in certain circumstances. In fact, so far as I know cryptographic hash functions are not injective. It is enough that collisions cannot be constructed in a way for an arbitrary payload to be mixed in with some carefully cooked up binary gobbledygook producing two messages that hash the same way. On the other hand such a function, if it is cheap to compute, is quite useful when used for data retrieval.
In any case, this is an area of current research for text documents: attempting to discover derivative and co-derivative documents (e.g. as in the case of plagiarism). It's quite an interesting topic. Naturally, you are looking for an approximate needle in a vast and fuzzy haystack when it comes to the Internet.
Of course there is such thing as what I call a fuzzy hash.
Any function which maps an array of bits into a number is a hash. It's just that very few of such functions are useful for as a cryptographic digest function. They still serve other purposes, such as indexing data structures. It's not hard to imagine an algorithm generating small array of numbers, such that the Cartesian distance between two such arrays is meaningful. Broadly speaking, such an array would be a hash.
The rhetoric that's been spewing out of some people's mouths when they can't talk for more than a minute about his actual stated policies is really starting to get nauseating.
Actually, geneticists have shown there is African ancestry (recent as in the last couple of thousands of years) in many European populations. I wouldn't be surprised if much of it spread through Muslim Spain and the Ottoman empire, but we also know that Mediterranean trade missions went as far afield as Britain during the Bronze Age, and where people swap goods, they also have a tendency of leaving genes behind as well. Think about it: Nubia to Egypt; Egypt to Phoenicia; Phoenicia to Celtic lands and so on.
Genes exist to be spread. There are certain Y chromosome markers that are only found in two places: an Indian tribe in the US Midwest, and in Finland. Nobody knows how this happened... well we know how it happened obviously, but the precise history of how this particular Y chromosome spread is unknown.
Likewise, take some historical figure like Alexander the Great. The statistical operation of gene transfer means that after well over a thousand years, either (a) he has no descendants or (b) practically everybody in the world is his descendant.
One of the ways propaganda works is by erasing fine, but important distinctions in the minds of the public. One of the most important distinctions is between bias (which everybody has) and propaganda.
I don't believe Obama will fix anything "magically", because I don't believe in magic. However, there is reason for everybody here to hope, even Republicans.
Democracy does not consistently select good leaders. No system can do that, which is why democracy is important: democracy works by making throwing out bad leaders easier. You don't have to take to the streets with Molotov cocktails or worry about your relatives disappearing into some secret police dungeon to get regime change.
Competition improves the breed, and so a spot of extinction pressure is a hopeful thing. It means the system is still working on some level.
It's just like the free market. The market doesn't mean you are consistently satisfied with your purchases. No economic system could do that. It means that it is easy to switch vendors, which keeps the vendors working hard to satisfy you and redress your grievances.
Our system is designed to prevent overnight changes. 2008 is the continuation of a process that started in 2006, of throwing out the Republican party. Republicans should be glad in the long term of this, just as Democrats should be glad that they can lose most of their gains in the next election. Just as competition improves the breed, complacency ruins it. The Democrats, for now, are aware the country has just thrown out the old regime, and that come January there will be no more excuses. As long as they are mindful of this, they will at least try to do better.
Republicans will of course be looking at the mistakes that the Democrats will make. If they're smart, they'll look at the mistakes they themselves made. There are those in the party who say the answer is to become more like the party that got thrown out in 2006. I think the Republicans have a good shot of convincing the American people the Democrats have overreached if they deemphasize the ultra socially conservative southern wing and turn to the fiscally conservative, socially moderate and libertarian leaning Republicans of the West and Northeast.
I hope they do, I hope they put together and honest, credible attack on the Democrats by 2012. And I'm a Democrat. No party that thinks itself unassailable, as the Republicans did in 2005, can be trusted.
There's an even simpler objection to this scheme. The most common change (other than scaling) that somebody is going to do to a picture is going to be cropping, which of course totally throws the idea hashing thumbnails out the window.
Still, the fingerprinting idea is still a good one. It's how search engines decide whether they've indexed a document already: they fingerprint. The case of two exactly identical files is an important special case.
Naturally, a cryptographic fingerprint algorithm is going to be a poor choice if you take the possibility of editing into account, since it is supposed to produce uncorrelatable results for minor differences. You'd want something that could efficiently calculate a set of fuzzy hashes for various croppings and rotations.
Well, I think I didn't quite express myself clearly. It isn't just a matter of not trusting the systems to operate in accordance with the rules; it's a matter of rules not performing in accordance with expectations.
This gets right to your point about which documents are classified.
What I'm saying is you can't count on the classification process to do the right thing, and even when it does the right thing you can't count on that to be enough. Either way, sensistive intelligence can be leaked that is not necessarily something that would be flagged as a "dataspill".
Of course organizations have to rely on their rules, but one of the goals for defense systems, even non-classified ones, should be that information should only be released through officially sanctioned channels. If an FOIA request comes through, then information gets released; that's fine. If documents are released through public affairs, they are purged of any hidden editorial history. If computers are retired, then the disks are destroyed or at least reliably erased before they become surplus.
So, I think, adding a bit of paranoia around non-classified defense networks is good.
This probably is a good goal for any government system: the system should be as transparent as possible, but not transparent through any accidental channels. I'm a big believer in open government, but government agencies also handle a great deal of sensitive personal information. I don't want some IRS contractor to leave a laptop with my tax return data in a coffee shop somewhere, but I do want everyone to know how tax enforcement policy is set.
I'm not necessarily disagreeing with what you are saying.
However, the power of networked computing is that with enough data and the right tools for sifting through it, you can make quite useful inferences. Verification of any data really amounts to checking it against different sources, after all. For example, if you know the Eurofighter can climb at a certain rate, and you know that in an exercise an F35 was unable to outclimb it, you can infer that the F35 cannot climb as quickly. Or vice versa.
This really comes back to "loose lips sink ships". Idle talk isn't very useful, but put it in the right context and it becomes intelligence.
I think you are right, in that removing the source of anger is important. That, after all, is the point of anger, isn't it? Except that in so many situations you can't remove the source of anger by being angry. That's civilization for you: you can't haul of and smack somebody you disagree with into submission.
I agree that if continually provoked to anger (or tempted to do something you are addicted to), you'll end up giving in. But getting angry (unless you can remove the source of anger that way) is no more useful than getting drunk to relieve your alcoholism.
Correction: any computer which is supposed to be allowed to access Secret information is not allowed to be hooked up to the Internet. I suspect there is no way to enforce the rule as you state it without possibly divulging what is secret and what is not. For example if I'm monitoring a computer and find that a bunch of files have been deleted, I might look at one of the files I downloaded that was purged, and say, "hey, this memo implies the F35 can climb at over 330 meters/second."
What I'm saying is that it's best not to trust in systems to operate according to the rules.
Asynchronous anger is something different altogether, although I think you have to be married to experience.
Your spouse is always angry about something you did in the past or are about to do in the future. Theoretically they can be mad at you for what you are doing right now, but I think in practice there isn't enough room for that, so it gets shoved into some kind of complex priority queue.
Slashdot Mirror
Sorry, that video is scientific garbage.
There was not a global 30cm drop in seal level around 1970, which would be quite noticeable.
Local mean sea level varies quite a bit due to geological factors and local weather effects such as atmospheric pressure. This produces statistical variation on various scales; individual locations might well see contrary trends; even aggregate trends smoothed over three year weighted averages tend to have considerable noise.
In any case, an island like the one in the video is a poor choice as a benchmark because mean sea level varies across the Pacific by as much as 60cm at any given time due to atmospheric effects. The El Nino/Southern Oscillation could well produce dramatic shifts in local MSL on an island like this. Pressure driven changes in MSL in this region can reach 30cm or in rare cases even more.
Well, I agree with your analysis of the economic purpose of copyright. However, in the US, at least, the power to grant copyright is not limited to the intent for which the Constitution grants that power to Congress. Were that so, then copyright extension would be unconstitutional. Since there is no way to send information back in time, there is no way to incent authors of pre-existing works by copyright extension.
The state of the law, as I understand it, is that the bit about "To promote the progress of science and useful arts" is considered "dicta", in other words an extraneous expression of opinion that is not binding. Once the power is granted, then Congress can use it for any purposes, even to retard the advancement of science. Furthermore, this means that the term of "limited times" mentioned in the next clause is unlimited, at least by considerations of economics. I do not think the state of the law is clear on this; the Constitution clearly states there must be limits but doesn't set a limit on those limits. Possibly a bald faced attempt to establish de facto perpetual copyright, say by specifying terms of millions of years, would fail Constitutional scrutiny, but extensions with the pretense of justification, say bringing terms into agreement with international agreements might.
Well... you are supposed to make a good faith effort to describe prior art in your patent application.
This makes the existence of such a "business method" questionable. An infringement suit under a patent obtained in the manner described amounts to using the court for purposes of defrauding the defendant. The only way to argue such a suit is to deliberately lie to the court in your arguments. If this business method patent were granted, I don't see how you could claim infringement damages under it, because you can't make money legally from patent fraud. Furthermore, you couldn't legally license this "invention", because you'd be contributing to fraud.
So, I don't see any value to a patent like this, unless the patent office and the courts want to advertise that they are simply instruments of fraud. I think this patent reflects the "when in doubt it's safer to patent anything that pops into your head" school of thought.
This is basic Operating Systems 101 stuff: throughput is not the same as responsiveness. Nor does average performance reflect how often you find yourself with not enough performance.
Once you have enough performance, say faster UI response than you can perceive as one example, getting more performance helps your average but is not that much of a practical benefit. So given two systems with the same average performance, if that average is adequate, then the system with less variance is preferable.
I'd say that Vista's average responsiveness is fine. It's throughput is fine. It's just perceptibly variable in performance on a lot of hardware. On hardware where there are occasional performance "glitches", a slightly less responsive operating system that was absolutely consistent would be perceived as better by users.
in all fairness McCain had lots of positive messages but they were flatly refused to be reported.
Actually, I think McCain was at fault here. I have nothing against negative advertising per se; it's perfectly fair, provided the attacks are accurate. If attacks are not accurate, then they'd better work.
McCain's attacks were too poorly focused. He should have focused mainly on Obama's relative inexperience, and built on that. Without a point of focus none of the negative impressions McCain wanted to create had a chance to gel. Their factual weakness made them treacherous to McCain. In 2000, many of Bush's negative ads were factually weak, but it didn't matter because they worked.
Note, however, that more favorable coverage is not the same thing as bias. For example, in covering the debate between advocates of intelligent design and advocates of evolution, "balanced" coverage -- coverage that is not favorable to one side or the other -- is biased, because intelligent design is not science. This is not to deny that bias might have existed in election coverage, but that isn't the only source of disparity between the treatments of the candidates.
One important source in this case was the quality of the campaigns. Obama ran a superb campaign. It was organized, disciplined and consistently on-message. McCain's campaign was none of these things. They kept searching for a new message, then circling back to ones that weren't working, like the Ayers issue. They could have raised Ayers again if momentum was swinging their way, but it wasn't; it was just an issue that hadn't stuck that they they were stuck on because they didn't know how to swing the election back their way. This lack of focus created a vacuum into which negative coverage expanded.
McCain himself couldn't stick to the script, and had to cut off press access, which is bad for a candidate who based his career on accessibility. Palin's lack of polish really undermined McCain's strongest issue in this election: even Obama supporters have to admit it would be better if he had a full term in the Senate under his belt.
This was a Democratic year; to overcome that, McCain's campaign needed to put together a message that resonated, and slowly dig itself out of the hole over the course of weeks. Obama showed how to do this. He started in a hole against Hilary Clinton, and his campaign demonstrated the staying power to wear down her lead over months and months.
McCain isn't like that; he's mercurial, given to dramatic gestures and sudden improvisations. That might work in an even year, but not this year. None of the big things he did that were supposed to sway the election, such as selecting Palin or "suspending" his campaign, had staying power to carry him through to election day.
This election was most emphatically not McCain's to lose. It was Obama's, and the McCain campaign simply failed to seize the initiative. Obama was vulnerable, but McCain's campaign was simply not able to put those vulnerabilities into focus. The press did not snatch victory from McCain's grasp; he just never put himself in a position to grasp victory. His poor press coverage simply reflected this. A well run campaign, say Bush's 2000 campaign, determines what the press is covering and how it is covering it.
This could get very ugly.
The law of the sea supposedly governs this kind of thing. Determining the extent of the continental shelf can extend the exclusive national right for minerals up to 150 nm past the EEZ, so in theory the documentation of the shelf should be a benign action. But ultimately international law is enforced by warfare on various scales of intensity, starting at diplomatic sanctions, through economic sanctions, and all the way up as high as warfare can go.
International law is only what you can force a country to accept as international law. We know Russia wants to claim these resources, and gainsaying them can lead to armed conflict.
Back in 1991, I remarked that the course of the twenty first century would be determined by the integration of the former Soviet states into the world political, security and economic systems. The opportunity to do this, if it ever existed, was bungled by the first Bush administration, and now we are dealing with a militarily powerful, mineral rich nation with a paranoid persecution complex and authoritarian instincts. Do they have more invested in stability than they can get out of grabbing territory?
Well, the fact that they don't want to build a fiber network doesn't mean they aren't selling network services in the city. They might be selling DSL or cable Internet, and the city fiber infrastructure would certainly put that business out of service.
It gets down to philosophical differences about economic value and the role of government. According to one point of view, the government should above all do no harm to any business. If a business wants to sell government weather data, the government ought to make it hard for individuals to get the data directly. By the same token, if a private company wants to provide network services in an area, the government has no business providing better or cheaper services.
The deleterious effect of static stretching on muscle power has been known for years.
It's not a matter of static stretching being "bad for you", what's "bad for you" depends on context. Static stretching is a developmental exercise. You wouldn't go to the weight room for serious strength training before a competition, and the same applies to static stretching.
Well coached athletes have been doing the kind of warm-up exercises described in the article for years, it's just that the word hasn't trickled down.
Well, you have to make a distinction between what an individual "needs" to survive and what a species "needs" to support certain evolutionary trends.
There is seems to be considerable scientific opinion to the effect that meat eating played an important role in making the large brained, linguistically gifted and tool making species H. sapiens possible. However, this doesn't mean that individual humans or pre-humans have ever "needed" meat to survive, or that eating meat to any degree was more "healthy" as we'd define it.
Humans in primitive conditions no doubt ate opportunistically. If you're a primitive person out hunting, you aren't going to ignore a bush full of berries because today's hunting day. Edible herbs probably went straight into the mouth without a second thought. So I expect people were constantly grazing on plants. On the other hand, they'd also gorge themselves on meat when given a chance -- check out the huge prehistoric shell middens they left behind. And of course, when you kill a mammoth, you'd better have a refrigerator for all that meat.
And, of course, humans didn't have refrigerators; they just became extremely adept at turning those meat calories into fat. They had to store calories, to support their very thermally expensive brains.
A modern person, of course, can graze on herbs and roots all the time. It's probably healthier for him too, although regular, small inclusions of meat in his diet do no harm and simplify getting the amino acids he needs.
Well, I don't doubt there was an element of political cowardice, but that's not the whole story. People were anxious to do something, and it's not as if every idea in the Patriot Act is bad. There's a lot of stuff that could be useful in there.
The Act is a typical case of a product rushed out the door without a sufficient requirements analysis.
For example, it has muddled priorities. A lot of the more objectionable features are built around the assumption that responding to a ticking time bomb is the highest priority. If, indeed, it comes to that, it certainly will be the highest priority, but the highest priority before it comes to that is keeping it from coming to that.
Even so, stipulating what can be done in such scenarios is probably a good idea, but those stipulations should be carefully bounded. That's the kind of talk that would go over like a lead balloon back in 2002, but it's true.
This leads us to another sign of a rush job: inattention to unintended consequences.
Probably every new investigatory power in the act would be justifiable in some highly specific circumstances. Just like bashing in somebody's front door is justifiable in certain high specific circumstances. But that shouldn't amount to carte blanche for door bashing. The powers granted should come with accountability for their use, and this is where the Patriot Act fails.
A serious minded civil libertarian could go over the Act and find better ways to accomplish everything the Act is supposed to do. The real problem was that the administration wanted -- had always wanted really -- a completely free hand unfettered by any kind of oversight.
Yes, we're on the same page when it comes to crypto digest functions in this application.
However the definitional issue is more than pedantry. Any kind of practical large scale document indexing system requires the calculation of hash functions, and in general functions useful for cryptography might not be useful for indexing and vice versa.
A non-injective hash function is not "bad"; it's just not useful in certain circumstances. In fact, so far as I know cryptographic hash functions are not injective. It is enough that collisions cannot be constructed in a way for an arbitrary payload to be mixed in with some carefully cooked up binary gobbledygook producing two messages that hash the same way. On the other hand such a function, if it is cheap to compute, is quite useful when used for data retrieval.
In any case, this is an area of current research for text documents: attempting to discover derivative and co-derivative documents (e.g. as in the case of plagiarism). It's quite an interesting topic. Naturally, you are looking for an approximate needle in a vast and fuzzy haystack when it comes to the Internet.
Of course there is such thing as what I call a fuzzy hash.
Any function which maps an array of bits into a number is a hash. It's just that very few of such functions are useful for as a cryptographic digest function. They still serve other purposes, such as indexing data structures. It's not hard to imagine an algorithm generating small array of numbers, such that the Cartesian distance between two such arrays is meaningful. Broadly speaking, such an array would be a hash.
The rhetoric that's been spewing out of some people's mouths when they can't talk for more than a minute about his actual stated policies is really starting to get nauseating.
<irony>I wonder how that happens.</irony>
Actually, geneticists have shown there is African ancestry (recent as in the last couple of thousands of years) in many European populations. I wouldn't be surprised if much of it spread through Muslim Spain and the Ottoman empire, but we also know that Mediterranean trade missions went as far afield as Britain during the Bronze Age, and where people swap goods, they also have a tendency of leaving genes behind as well. Think about it: Nubia to Egypt; Egypt to Phoenicia; Phoenicia to Celtic lands and so on.
Genes exist to be spread. There are certain Y chromosome markers that are only found in two places: an Indian tribe in the US Midwest, and in Finland. Nobody knows how this happened... well we know how it happened obviously, but the precise history of how this particular Y chromosome spread is unknown.
Likewise, take some historical figure like Alexander the Great. The statistical operation of gene transfer means that after well over a thousand years, either (a) he has no descendants or (b) practically everybody in the world is his descendant.
One of the ways propaganda works is by erasing fine, but important distinctions in the minds of the public. One of the most important distinctions is between bias (which everybody has) and propaganda.
I don't believe Obama will fix anything "magically", because I don't believe in magic. However, there is reason for everybody here to hope, even Republicans.
Democracy does not consistently select good leaders. No system can do that, which is why democracy is important: democracy works by making throwing out bad leaders easier. You don't have to take to the streets with Molotov cocktails or worry about your relatives disappearing into some secret police dungeon to get regime change.
Competition improves the breed, and so a spot of extinction pressure is a hopeful thing. It means the system is still working on some level.
It's just like the free market. The market doesn't mean you are consistently satisfied with your purchases. No economic system could do that. It means that it is easy to switch vendors, which keeps the vendors working hard to satisfy you and redress your grievances.
Our system is designed to prevent overnight changes. 2008 is the continuation of a process that started in 2006, of throwing out the Republican party. Republicans should be glad in the long term of this, just as Democrats should be glad that they can lose most of their gains in the next election. Just as competition improves the breed, complacency ruins it. The Democrats, for now, are aware the country has just thrown out the old regime, and that come January there will be no more excuses. As long as they are mindful of this, they will at least try to do better.
Republicans will of course be looking at the mistakes that the Democrats will make. If they're smart, they'll look at the mistakes they themselves made. There are those in the party who say the answer is to become more like the party that got thrown out in 2006. I think the Republicans have a good shot of convincing the American people the Democrats have overreached if they deemphasize the ultra socially conservative southern wing and turn to the fiscally conservative, socially moderate and libertarian leaning Republicans of the West and Northeast.
I hope they do, I hope they put together and honest, credible attack on the Democrats by 2012. And I'm a Democrat. No party that thinks itself unassailable, as the Republicans did in 2005, can be trusted.
There's an even simpler objection to this scheme. The most common change (other than scaling) that somebody is going to do to a picture is going to be cropping, which of course totally throws the idea hashing thumbnails out the window.
Still, the fingerprinting idea is still a good one. It's how search engines decide whether they've indexed a document already: they fingerprint. The case of two exactly identical files is an important special case.
Naturally, a cryptographic fingerprint algorithm is going to be a poor choice if you take the possibility of editing into account, since it is supposed to produce uncorrelatable results for minor differences. You'd want something that could efficiently calculate a set of fuzzy hashes for various croppings and rotations.
Well, I think I didn't quite express myself clearly. It isn't just a matter of not trusting the systems to operate in accordance with the rules; it's a matter of rules not performing in accordance with expectations.
This gets right to your point about which documents are classified.
What I'm saying is you can't count on the classification process to do the right thing, and even when it does the right thing you can't count on that to be enough. Either way, sensistive intelligence can be leaked that is not necessarily something that would be flagged as a "dataspill".
Of course organizations have to rely on their rules, but one of the goals for defense systems, even non-classified ones, should be that information should only be released through officially sanctioned channels. If an FOIA request comes through, then information gets released; that's fine. If documents are released through public affairs, they are purged of any hidden editorial history. If computers are retired, then the disks are destroyed or at least reliably erased before they become surplus.
So, I think, adding a bit of paranoia around non-classified defense networks is good.
This probably is a good goal for any government system: the system should be as transparent as possible, but not transparent through any accidental channels. I'm a big believer in open government, but government agencies also handle a great deal of sensitive personal information. I don't want some IRS contractor to leave a laptop with my tax return data in a coffee shop somewhere, but I do want everyone to know how tax enforcement policy is set.
seriously wtf is with that! girls suck!
Ah. An optimist.
I'm not necessarily disagreeing with what you are saying.
However, the power of networked computing is that with enough data and the right tools for sifting through it, you can make quite useful inferences. Verification of any data really amounts to checking it against different sources, after all. For example, if you know the Eurofighter can climb at a certain rate, and you know that in an exercise an F35 was unable to outclimb it, you can infer that the F35 cannot climb as quickly. Or vice versa.
This really comes back to "loose lips sink ships". Idle talk isn't very useful, but put it in the right context and it becomes intelligence.
I think you are right, in that removing the source of anger is important. That, after all, is the point of anger, isn't it? Except that in so many situations you can't remove the source of anger by being angry. That's civilization for you: you can't haul of and smack somebody you disagree with into submission.
I agree that if continually provoked to anger (or tempted to do something you are addicted to), you'll end up giving in. But getting angry (unless you can remove the source of anger that way) is no more useful than getting drunk to relieve your alcoholism.
Correction: any computer which is supposed to be allowed to access Secret information is not allowed to be hooked up to the Internet. I suspect there is no way to enforce the rule as you state it without possibly divulging what is secret and what is not. For example if I'm monitoring a computer and find that a bunch of files have been deleted, I might look at one of the files I downloaded that was purged, and say, "hey, this memo implies the F35 can climb at over 330 meters/second."
What I'm saying is that it's best not to trust in systems to operate according to the rules.
Asynchronous anger is something different altogether, although I think you have to be married to experience.
Your spouse is always angry about something you did in the past or are about to do in the future. Theoretically they can be mad at you for what you are doing right now, but I think in practice there isn't enough room for that, so it gets shoved into some kind of complex priority queue.
No, people don't need to vent. They need to feel like they are control of things that are important to them. Venting is a poor substitute.
I agree, it is nice to have people agree with you. That doesn't mean you stop being mad. In fact, I suspect it trains you to be mad.