I would not want to trade e.g. US cancer survival rates for the European cancer survival rates
Again, citation needed. And with all of them, no neocon sites or insurance company sites.
Are you retarded? What part of "Lancet Oncology Journal" do you not understand (I gave that example because it is the most recent)? The only way you could not find a myriad of supporting citations in reputable medical journals is if you refused to look -- this is a well-known and undisputed fact in medical academia.
I actually heard about the Lancet Oncology data in the British press, mostly because they wanted to know why the UK NHS has such abysmal cancer survival rates compared to the major economies of Europe. (What the British press did not say is that the same study shows the US out-performing every industrialized country in the world for which we have data in this regard.)
Like I said, your attitude is an exemplar of why nothing good will come from the politics of healthcare.
Our lifespan is shorter, our infant mortality is higher, and we pay more per capita for our health care.
This is the kind of superficial reasoning that will prevent any good from coming from these discussions. For example:
- American life spans are shorter due to higher incidence of accidents and homicide. If you remove these from the statistics, Americans are among the longest lived people in the industrialized world. Healthcare has nothing to do with it. The only person that lives longer than a Japanese person in Japan is a Japanese person in the US.
- Infant mortality in the industrialized world is a function of genetics, with some ethnic groups having integer factor higher rates. If you control for genetic/ethnic demographics, all industrialized countries have roughly the same infant mortality rate. Infant mortality in the non-industrialized world is a function of infectious disease -- those people might actually benefit from more healthcare.
- Americans pay more for their healthcare, but by every direct measure of medical outcomes they also receive superior results, particularly in terms of diagnostic accuracy and disease treatment. For example, in the recent Lancet Oncology study, the average US survival rate across all cancers exceeds that of every other country in the world by a large margin. Relative to western Europe, your odds of surviving cancer is 20-40% higher in the US, which is no small margin. Since these are the results I am actually paying for with healthcare, it seems that Americans get more when they pay more.
All of which ignores the fact that every state has socialized healthcare services for the poor -- it is not a Federal function under the US constitution. There is government healthcare, just not at the Federal level. I am okay with the idea of universal healthcare, but let us be accurate when characterizing the current situation. For starters, I would not want to trade e.g. US cancer survival rates for the European cancer survival rates considering that cancer is one of the leading killers in the industrialized world. If the European model is so great, how come their disease survival rates are so poor relative to the US?
They have first rate low tech preventative and pre/post natal health care. Which gives them a lower infant mortality rate than the US and a life expectancy just a bit lower that the US.
In the industrialized world infant mortality rates are primarily a function of genetics, not healthcare. Some ethnic groups have integer factor higher rates of miscarriage even after controlling for the environmental and medical factors. If you control for ethnic and racial demographics, infant mortality rates are more or less the same in the industrialized world. In the non-industrialized world, infant mortality rates are primarily a function of infectious disease (a pretty low bar to cross before genetics take over).
A similar story exists with life expectancy, which largely measures environmental factors (e.g. car accidents and homicide in the case of the US). The statistics you are using do not measure what you think they are measuring. For example, if you look at survival rates for cancer -- a leading cause of death in the industrialized world -- the US leads the world by a significant margin (e.g. per the recent Lancet Oncology study), yet this is not obvious in the aggregate life expectancy statistics.
The problem with software patents is that anything that's really novel, like your impossible compression algorithm is basically a mathematical algorithm. Since you can't patent mathematical algorithms, there shouldn't be any need for software patents. I'm not sure if I've ever seen anything really inventive in software that wasn't a mathematical algorithm.
The problem is that all patents are mathematical algorithms, and as long as we continue to pretend this is not the case we will have people peddling non-solutions to the real problems. People who do not recognize this are not credible participants in the discussion of whether or not patents (of any type) should be abolished.
More generally though, algorithm patents are no more "patenting mathematics" than engine patents are "patenting thermodynamics". Your assertion is silly. The idea of "sorting" in computer science cannot be patented, but an implementation of the idea e.g. "bubblesort" can be, and the specific reduction to practice (e.g. your code) is protected by copyright. Same as with hardware patents or chemical process patents. No one is preventing you from inventing another algorithm to do "sorting" any more than a chemical process patent prevents you from inventing another synthesis path to some molecular end product. Saying that algorithm patents are "patenting mathematics" and other patents are not is a fallacy of modifying definitions in the middle of an argument to generate a particular outcome that comports with your desires.
I'm not precisely sure; that is what lawyers are for, and I got this from lawyers that do a lot of international tech patents.:-)
The gist of it is that if you invent an algorithm and implement it in source code, the compiler target effectively determines whether the resulting product can be patent protected in Europe as a practical matter. Both a CPU and FPGA are general purpose computing silicon (albeit optimized for different tasks), but selecting one or the other as the compilation target for the source code can determine patent protectibility; an FPGA is not "fixed silicon", but for patent purposes it is often treated as such and so you start there and work your way back further into more hardware-like substrates. The problem, obviously, is that in this day there is a very smooth continuum between silicon and software in practice such that the pretense of a difference does not pass casual inspection, never mind that in theory there has never been a difference. Consequently, there is a gray area where it is not obvious whether the subject matter is patentable in Europe, and that fact is exploited quite liberally by patent lawyers since "software" or "hardware" is purely a function of framing. For true algorithm patents (not crap like "business process" patents), this is apparently a very functional loophole that allows de facto algorithm patents.
Because of the essential similarity between CPUs and many other types of silicon, once a patent has been granted for the "hardware" machinery it is relatively easy to swim upstream to make a kind of inverted argument that CPU implementations are infringing on the patent. Again, the only way to really fix this is to address the inconsistency of pretending bits and molecules have a meaningful difference from the standpoint of process invention. As we continue to become better at manipulating molecular structures at a very fine level, this distinction will only become blurrier.
Why is software so special that it's the only thing that I know of covered by both copyright and extensive patents?
This is a trivially falsifiable assertion, and it does not do anyone's credibility any good that it is repeated so often.
It is the general case for patents, e.g. chemical process patents, that the new abstract process/algorithm (the part valuable to a third party) is protected by patent and specific reduction to practice is protected by copyright (which may or may not have any practical value). In some other venerable patent areas, these are largely independent works, though in some cases the reduction to practice may be licensed as well. One argument that can be made for algorithm patents is that they are structurally indistinguishable in both theory and practice from an chemical process patents -- swap "bits" for "molecules". Software receives no special protection like you assert above, the algorithm/process and implementation protected by patent and copyright respectively, much like it is for everything else.
Of course, the elephant in the room that everything is an algorithm and that there is no mathematical difference between bits, molecules, hardware, software, or data. It is a distinction with no theoretical difference and people keep trying to patch up the law so that we can pretend a distinction exists in the face of clear evidence to the contrary; copyright has a similar issue, by the way. It has led to absurdities such as an algorithm on Intel x86 not being patentable in Europe but the same algorithm on a Xilinx Vertex-5 is -- the distinction between the two is arbitrary and capricious. The problem is not algorithm patents (and much of what we are talking about here is business process patents, not algorithm patents per se) but that so many patents are frivolous, but that is not a problem unique to any particular field of patentability. The two biggest problems are really frivolous patents being rubberstamped, and theoretically inconsistent treatment in a few narrow areas that are then bleeding over into other areas as the inconsistency becomes obvious in real cases. The only way to actually fix these inconsistencies is by adopting an all or nothing policy; I have no particular opinion on which way that goes.
Undoubtedly, the most important "non-medical" factor is the fact that close to 1/4 of the population doesn't have proper access to the medical system in the first place, thereby exacerbating any medical problems that they have until it's too late. You can't just sweep that under the rug.
If you eliminate accidents and homicide from the statistics, Americans live longer. Period. Do you realize how radically better US healthcare would have to be for your assertion to hold up in the statistics? Direct healthcare outcomes for the average American, rich or poor, are better than the rest of the industrialized world, but not that much better. You cannot juggle the numbers to make that fact go away, and its reality is well documented in the medical journals (e.g. the recent Lancet Oncology study). The problem with Americans is demographic, genetic, environmental, and behavioral.
You can try and dodge the elephant, but it is quite large. A poor person with cancer in the US has better survival odds than a poor person in the UK's NHS. Every state in the US has a public healthcare system for the poor, incidentally, since the US Federal government has no jurisdiction in these matters. The idea that there are people without access to medical care in the US is false; most of the argument is over whether there is a more efficient way. As someone raised on free public healthcare in the US in multiple states, it never ceases to amaze me that it supposedly does not exist. This is mostly just ignorance from foreign perspectives that fail to realize that individual US States are essentially the equivalent of countries in the EU and that public healthcare is dealt with at that level of government.
Funny how you think that you "know" that, given that we're essentially the only developed country that doesn't provide some form of national health care, we pay almost twice as much for healthcare as the next most expensive country, and even with all that money we're spending, we're nowhere near the top of the list of healthiest or longest living populations.
You are conflating demographic and environmental factors with healthcare outcomes.
If you, for example, remove non-medical causes of premature death (car accidents, homicide, etc) Americans outlive other industrialized countries. Healthcare is only a small factor in life expectancy, and average healthiness is almost completely unrelated to healthcare in the industrialized world. The environmental and demographic factors are atypically poor in the US relative to the industrialized world.
If you look at direct measures of healthcare outcomes, such as diagnostic accuracy and disease survival rates, the US leads the industrialized world by a large margin. The elephant in the room in the recent Lancet Oncology study, for example, was that cancer survival rates in the US are much higher than in any other industrialized country in the world -- about 20-40% on average depending on the country and the cancer. So in this sense, Americans are paying more but they are also receiving much more.
The real situation is that the US has terrible non-medical factors that drag down its statistics but compensates with the best average medical outcomes by a huge margin. In most of the rest of the industrialized world, you have middling to good non-medical factors and middling to poor medical outcomes. In other words, the aggregate statistics are not measuring the same thing. Since we pay the medical establishment to produce positive medical outcomes, it would seem prudent to evaluate their efficacy based on those results and not on the number of automobile accidents people are involved in.
At a minimum, it would be foolish to trash a medical system that produces results such as cancer survival rates that no other system is currently coming close to. The US system may be byzantine and inefficient, but it also outperforms the rest of the world in the key metric of medical results. Let's not throw out the baby with the bath water, at least not until a national healthcare system exists with equivalent medical outcomes.
Re:Too many 'this stuff sucks' moments
on
The Future of XML
·
· Score: 2, Interesting
Finally, my application had poor performance because XML is so slow and bloated to read in as a wire protocol.
When was the last time you tried it, 1995? Nowadays, compression algorithms require so little processing power that XML adds only a minimal amount of overhead when transfered over the wire.
Actually, you are demonstrating some cluelessness here. Size bloat is only a small part of why XML massively sucks as a wire protocol compared to functionally equivalent universal representations such as ASN.1 which were designed to be extremely fast wire protocol encodings; compression does not address much of the performance suckiness. The reality is that good wire protocol representations are about an order of magnitude faster than equivalent XML formats, which matters a lot of you are moving any kind of traffic. Since ASN.1 is an ITU standard, it is usually used for networking protocols (as opposed to application protocols). There is even an ASN.1 encoding standard called XER that losslessly encodes XML as ASN.1 and back so as to improve its characteristics as a wire protocol. There are other wire protocol encodings out there, though ASN.1 is probably the most prevalent.
If you think compressing XML makes it a good wire protocol, you are simply too ignorant or too inexperienced to be designing or implementing wire protocols. Study the design of some of the binary wire encodings and maybe you will figure out the subtleties that make them so much faster. There is a hell of a lot more to wire protocol scaling and performance than the number of bits going over the wire. Methinks you stepped a wee bit outside your area of expertise.
"I just think you've never been outside USA (and that's a mild way of putting this)."
Former European citizenship aside, I think I spent more of my life in Asia than Europe (I've never really done the arithmetic). Is that what you are referring to? Do not confuse me with one of those people that has never been anywhere. On the other hand, one might suspect that of you...
As for health care outcomes, that is standard fare for health care economics and the most scathing indictments of the European health care system in terms of health care outcomes actually come from Europe. Go ahead and Google on cancer survival rates between Europe and the US. The statistical difference is pretty stark. Europeans pay less, but if you have anything serious you do not want to be anywhere but the US. Remember, those survival rates are *average* and outcomes track in income in Europe the same way they do in the US (another dirty secret well known to health care economists).
I love the subject of health care. It brings out the finest kinds of ignorance. Some random facts to consider (Google for the references):
- As a financial instrument, insurance exists to distribute risk, not cost. Anybody who does not understand what the distinction is please vacate the discussion. Technically speaking, insurance is how one distributes risk and some approximation of communist government (in a literal rather than pejorative sense) is how one distributes cost. Trying to use the former to approximate the latter is inefficient and raises the costs for everyone.
- For all the inefficiency and expense of what passes for health care in the US, the US also has the best health care outcomes in the industrialized world and generally by a wide margin. If you have cancer, your survival rates in the US are much better than Europe on average and the best in the world in absolute terms. This is true by a number of other direct metrics of health outcomes and holds across the population even if you are an average person and not a wealthy person. Americans are paying more but they are getting more, and survival rates for a pretty broad swath of nasty things is 20-40% better, not buried in the noise floor. This is the good part of the US health care system that no socialized system has ever emulated. Americans compensate for being unhealthy (and car accidents, etc) in mortality rates with really good medical outcomes. If you normalize for genetics and environment, Americans live longer than anyone else. Of course, many Americans have crap genetics and have a crap diet as far as longevity is concerned.
- Before the Americans get too smug, the American health care "system" (there is no system, it is a market) is byzantine and inefficient. It should cost nowhere near what it does even for what Americans get.
- All Americans have health care, even those that cannot afford it, and the idea that there are people without access to health care is a myth that inflames the clueless and serves the purposes of political propaganda. The quality is mediocre, but what do you expect with socialized medicine. It is not hypothetical, I was one of those invisible souls raised on government health care for the destitute.
What we really have is a number of facts. The European system produces mediocre results in terms of actual health care outcomes (what we are nominally paying for), but it is relatively inexpensive. Americans pay a lot but have the best health care outcomes in the world. Americans pay far more than is strictly necessary by any reasonable metric, but I guess they can afford it. Americans are also bearing the cost of most medical technology innovation, amortized in the American medical market; when is the rest of the industrialized world going to carry their fair share of that burden?
In short, all the systems suck. That said, I would be reluctant to give up the superior health outcomes (what I pay doctors for) and medical innovation of the American medical environment. On the other hand, I wish they were more efficient at what they do. Clearly there has to be a better way, but by every metric that matters to someone getting health care, replicating the European system is not it.
"*cough*JDAM*cough*Tomahawk*cough*F-16*cough* Sorry got something stupid stuck in my throat. Anyways as I was saying the US has plenty of weapons sytems that use GPS systems and would either become completely ineffective or seriously crippled without it."
As was pointed out elsewhere, neither JDAMs nor Tomahawks (nor F16s for that matter) use GPS guidance -- only technically ignorant tools claim that any US weapon systems use GPS guidance. Even rudimentary research shows that systems like JDAMs and Tomahawk get their navigation data from ultra-precise laser ring interferometers (an extremely precise solid-state optical accelerometer with a purpose similar to mechanical gyroscopes), with the ability to optionally accept GPS fine-tuning corrections. The precision of the inertial systems is classified, but it is generally known that it is apparently not much worse than the GPS corrected version for weapon targeting purposes (they may have already converged for all we know -- INS has been continuously improved, and it wasn't bad to start with). Note that this is also why jamming or toying with the GPS signal does not send bombs and such flying way off course; the inertial system only accepts corrections within the computed error bars of its own positioning data. If the GPS signal is outside those bars, the navigation system assumes the GPS has been compromised.
I just gotta love the armchair weapon designers who think that of the thousands of bright engineers that built all these weapons, it never occurred to anyone that someone might jam or disable the weak RF signal that is GPS. GPS is a convenience, not a necessity, and it would only have small impact on the efficacy of American weaponry -- by design. The reason the US military embeds inertial navigation in everything is precisely because it is essentially impervious to countermeasures short of altering the physics of the universe. For all we know, nominal GPS weapon targeting is largely for show and indirection, since it costs next to nothing to strap a GPS receiver on an inertial navigation system (and it would seem to have worked in that case, considering how much breathless idiocy is fixated on GPS weapon targeting).
As a point of trivia, the laser inertial navigation systems used by the US military were invented in the 1960s for some (failed) ballistic missile interceptor research. Even though the ABM research was a dead end, the laser ring gyro technology developed for the project paid dividends for the US military way out of proportion to the money expended on the ABM research.
"The US military is completely dependant on their technology and the rest of the world knows it. Do their cruise missiles even work without GPS?"
The US has no weapon systems that are GPS guided and never has, precisely because it is vulnerable. The Chinese may have just now gotten around to developing anti-satellite technology, but the Soviet Union had it ages ago.
The core guidance package of US weapon systems is extremely high precision inertial navigation (all systems described as "GPS-guided" are actually inertial -- the media is a bit stupid about these things, as GPS is an optional untrusted overlay on inertial navigation systems). Some intelligent terrain following weapons also use optical geo-referencing. As a matter of policy going back to the Soviet Union days, the US military machine views satellite systems as "nice to have" but its infrastructure is pervasively designed to operate under the presumption that there are no satellites in orbit. The vulnerability of the US military to massive system outages is greatly overstated; the Soviet Union was a much bigger threat on this scale than the Chinese are, and the US military has always been pretty religious about designing systems whose functionality was robust and in the face of rapidly degrading military infrastructure and relatively decentralized. It is easy to forget it, but the Chinese have nothing on the old Soviet Union in terms of technology and force numbers, and that was the doctrinal enemy of much of the modern US military.
"They need to add a GOOGLE function to allow queries to be searched nicer.
SELECT * FROM articles WHERE GOOGLE('boobies');"
I don't know about MySQL, but you can pretty trivially write extensions to PostgreSQL that do exactly this kind of thing. In fact, I've written a number of such extensions to PostgreSQL that make Google resources seem like local resources in a PostgreSQL database. (This kind of deep and easy customizability is one of the things I find to be a killer feature of PostgreSQL relative to many/most other databases.)
The "pay raise" is a bribe by the board for key executives to not jump ship. It is a common pattern when companies are in trouble to put golden handcuffs on key people to keep them around to see the trouble through to the end rather than leaving for better climes with fewer risks. If you think about it, it makes a certain amount of economic sense.
...smart terrorists only use peroxide-based explosives (like the London subway bombing et al), oxidized halide based explosives (e.g. chlorate), and various other dirt cheap and ubiquitous explosives. While many of the most famous explosive chemistries might be subject to nitrate tests, the range of explosive chemistries that have been used at various times is far more diverse than nitrates. First World War mortar explosives are as dangerous today as they were back then, even if some of them do not contain nitrates.
The fixation on the detection of nitrate and related chemistry is a bit of a blind spot in explosive detection technology.
"Anyone want to guess how long it'll be before Victor finds himself out of a job?... Unfortunately..."
Before you don the tinfoil, a basic education on the US system of government might be in order. The Federal judiciary are appointed for life and only substantial evidence of gross misconduct puts them at risk of removal by a super-majority of Congress.
"No, copyrights are for written works or other forms of expression, not implementations of arbitrary ideas. The "implementation that produces" the chemical is also not copyrightable. A document which describes the implementation is copyrightable, the process for making the chemical is patentable at best."
Definitely incorrect. There have been numerous copyright infringement suits, rather large ones, for cloning hardware implementations. Or at least, this has been true in the US for a very long time. There are many types of technical hardware that you can buy overseas that is not sold in the US because it was copied from US-based designs. In many such cases, it is not a matter of patent protection or even trade dress but e.g. the circuit board layout being copied.
Copyright covers a hell of a lot more than "written works". Forms of expression included, broadly, implementation.
Chemical process patents are basically algorithms, but done with atoms and molecules instead of bits. The implementation of the abstract process is subject to its own IP protection independent of the patented process. In theory, a more efficient way to move atoms around to achieve a particular result should have no more and no less protection than a more efficient way to move bits around to achieve a particular result.
Software patents are a bit of a misnomer, further polluted by frivolous patents (which all fields suffer from). A software patent, ideally, is a new method/algorithm for doing something in the bit domain e.g. a more efficient method for doing something we can already do. Good software patents generally do not restrict the abstract implementation of a result, just a particular method of obtaining a result that may be more efficient than previous methods in some fashion. I will readily acknowledge that some so-called "software patents" are idiotic, but that is independent of whether or not the idea is valid in the abstract -- no accounting for USPTO incompetence.
Frivolous patents are hardly unique to software, and are one of the main problems with the patent system.
All implementations of patents are copyrighted. It is the nature of the thing, so let's stop pretending like this is unique to software algorithms. Chemical process patents are fundamentally indistinguishable from software patents in all respects -- including copyright -- yet we ignore them and their long history. I get the impression that posters on slashdot are so clueless about other fields that they think software is special in this regard.
This issue will not be solved until people accept the mathematical truism that hardware patents and process patents are indistinguishable from software patents; one of the accomplishments of 20th century mathematics was proving that fact. Either we accept that algorithms can be patented or not, recognizing that it is *all* algorithms, or we fumble around with broken law that refuses to acknowledge reality (which would not be a first for government).
"An attack with those babies will not be stopped by the current generation of missile defense systems. It is _not_ a completely ballistic rocket. In other words predicting, calculating it's trajectory and using all the billions of dollars of infrastructure designed for ballistic missiles is not as useful anymore."
You severely underestimate and misunderstand the operation of modern ABM systems. Current US ABM missile technology uses broad-spectrum imaging based terminal guidance, exactly like air-to-air missiles and similar (in fact the terminal homing software is largely identical). It has the same ability to follow a non-ballistic trajectory that all other missiles with this type of guidance do, and these targets are far less agile than, say, a fighter aircraft. The "new technology" in state-of-the-art ABM was the ability to marry this kind of terminal guidance package to a rocket that operates in the ABM envelope (though most of that "technology" is little more than extremely advanced materials science).
The Russians are talking nonsense and preying on the ignorance of most people regarding how modern US ABM systems actually function. The design of the US ABM system was to defeat all targets, agile or otherwise, but extending the envelope of target capability well into the ABM range rather than just local air defense.
"The transportation and residential sectors combined make up more than half of our greenhouse gas emissions. I'm not talking about the switch from a Tahoe to a Prius, but from a Tahoe to something like a Chevy Volt (sized for real-world use, of course). Given most people's driving patterns, that could cut our CO2 emissions from transportation by more than half."
You are playing games with statistics. Coal and natural gas alone account for more than half of all CO2 emissions in the US. How many people are using coal and natural gas in their vehicles? Of the less-than-half that is left, less than half of *that* is due to motor vehicle emissions. And most of those motor vehicles are not SUVs.
In short, you buy something like a 5% net reduction in CO2 output for some very heavy-handed and moderately costly motor vehicle regulation. Pissing in the ocean. Simply moving away from fossil-fuel based power production to nuclear and geothermal (and others when it makes sense) would slash CO2 output in approximately half for a moderate one-time cost that would be no more than the cost of making everyone drive tiny cars. It would also be a largely transparent change that would not significantly impact quality of life for most of the country.
A much, much simpler way to reduce fuel consumption in vehicles is to stop taxing diesel much more heavily than gasoline. Diesel is a less expensive fuel *and* it gets substantially better mileage (moreso than gasoline hybrids in most comparable cases).
"If political will existed we could have vehicles that do everything SUVs do now, and houses just as big and comfy as the ones we have now, that had only a small fraction of the environmental impact. The technology is there; it's just a question of making it economically feasible."
Considering that the environmental impact of SUVs versus, say, a Prius is negligible in the big environmental calculus, I would say that it is an entirely unimportant point. On the list of things that will have the most impact on the environment that is way, way down the list in both terms of impact and return on investment as far as the environment goes.
People routinely conflate oil trade issues with environmental issues -- they are not the same. Eliminating all SUVs tomorrow will have a negligible impact on CO2 production or most other environmental issues people seem to care about. Eliminating SUVs tomorrow would impact the geopolitics of the oil trade, but that has very little to do with environmental impact. A lot of people get hot and bothered about SUVs, but they are pretty much inconsequential in terms of the environment if you look at actual relative impact numbers. If environmental results are all we are looking for, spending energy on SUVs offers an incredibly low return on investment.
Slashdot Mirror
I actually heard about the Lancet Oncology data in the British press, mostly because they wanted to know why the UK NHS has such abysmal cancer survival rates compared to the major economies of Europe. (What the British press did not say is that the same study shows the US out-performing every industrialized country in the world for which we have data in this regard.)
Like I said, your attitude is an exemplar of why nothing good will come from the politics of healthcare.
This is the kind of superficial reasoning that will prevent any good from coming from these discussions. For example:
- American life spans are shorter due to higher incidence of accidents and homicide. If you remove these from the statistics, Americans are among the longest lived people in the industrialized world. Healthcare has nothing to do with it. The only person that lives longer than a Japanese person in Japan is a Japanese person in the US.
- Infant mortality in the industrialized world is a function of genetics, with some ethnic groups having integer factor higher rates. If you control for genetic/ethnic demographics, all industrialized countries have roughly the same infant mortality rate. Infant mortality in the non-industrialized world is a function of infectious disease -- those people might actually benefit from more healthcare.
- Americans pay more for their healthcare, but by every direct measure of medical outcomes they also receive superior results, particularly in terms of diagnostic accuracy and disease treatment. For example, in the recent Lancet Oncology study, the average US survival rate across all cancers exceeds that of every other country in the world by a large margin. Relative to western Europe, your odds of surviving cancer is 20-40% higher in the US, which is no small margin. Since these are the results I am actually paying for with healthcare, it seems that Americans get more when they pay more.
All of which ignores the fact that every state has socialized healthcare services for the poor -- it is not a Federal function under the US constitution. There is government healthcare, just not at the Federal level. I am okay with the idea of universal healthcare, but let us be accurate when characterizing the current situation. For starters, I would not want to trade e.g. US cancer survival rates for the European cancer survival rates considering that cancer is one of the leading killers in the industrialized world. If the European model is so great, how come their disease survival rates are so poor relative to the US?
In the industrialized world infant mortality rates are primarily a function of genetics, not healthcare. Some ethnic groups have integer factor higher rates of miscarriage even after controlling for the environmental and medical factors. If you control for ethnic and racial demographics, infant mortality rates are more or less the same in the industrialized world. In the non-industrialized world, infant mortality rates are primarily a function of infectious disease (a pretty low bar to cross before genetics take over).
A similar story exists with life expectancy, which largely measures environmental factors (e.g. car accidents and homicide in the case of the US). The statistics you are using do not measure what you think they are measuring. For example, if you look at survival rates for cancer -- a leading cause of death in the industrialized world -- the US leads the world by a significant margin (e.g. per the recent Lancet Oncology study), yet this is not obvious in the aggregate life expectancy statistics.
The problem is that all patents are mathematical algorithms, and as long as we continue to pretend this is not the case we will have people peddling non-solutions to the real problems. People who do not recognize this are not credible participants in the discussion of whether or not patents (of any type) should be abolished.
More generally though, algorithm patents are no more "patenting mathematics" than engine patents are "patenting thermodynamics". Your assertion is silly. The idea of "sorting" in computer science cannot be patented, but an implementation of the idea e.g. "bubblesort" can be, and the specific reduction to practice (e.g. your code) is protected by copyright. Same as with hardware patents or chemical process patents. No one is preventing you from inventing another algorithm to do "sorting" any more than a chemical process patent prevents you from inventing another synthesis path to some molecular end product. Saying that algorithm patents are "patenting mathematics" and other patents are not is a fallacy of modifying definitions in the middle of an argument to generate a particular outcome that comports with your desires.
I'm not precisely sure; that is what lawyers are for, and I got this from lawyers that do a lot of international tech patents. :-)
The gist of it is that if you invent an algorithm and implement it in source code, the compiler target effectively determines whether the resulting product can be patent protected in Europe as a practical matter. Both a CPU and FPGA are general purpose computing silicon (albeit optimized for different tasks), but selecting one or the other as the compilation target for the source code can determine patent protectibility; an FPGA is not "fixed silicon", but for patent purposes it is often treated as such and so you start there and work your way back further into more hardware-like substrates. The problem, obviously, is that in this day there is a very smooth continuum between silicon and software in practice such that the pretense of a difference does not pass casual inspection, never mind that in theory there has never been a difference. Consequently, there is a gray area where it is not obvious whether the subject matter is patentable in Europe, and that fact is exploited quite liberally by patent lawyers since "software" or "hardware" is purely a function of framing. For true algorithm patents (not crap like "business process" patents), this is apparently a very functional loophole that allows de facto algorithm patents.
Because of the essential similarity between CPUs and many other types of silicon, once a patent has been granted for the "hardware" machinery it is relatively easy to swim upstream to make a kind of inverted argument that CPU implementations are infringing on the patent. Again, the only way to really fix this is to address the inconsistency of pretending bits and molecules have a meaningful difference from the standpoint of process invention. As we continue to become better at manipulating molecular structures at a very fine level, this distinction will only become blurrier.
This is a trivially falsifiable assertion, and it does not do anyone's credibility any good that it is repeated so often.
It is the general case for patents, e.g. chemical process patents, that the new abstract process/algorithm (the part valuable to a third party) is protected by patent and specific reduction to practice is protected by copyright (which may or may not have any practical value). In some other venerable patent areas, these are largely independent works, though in some cases the reduction to practice may be licensed as well. One argument that can be made for algorithm patents is that they are structurally indistinguishable in both theory and practice from an chemical process patents -- swap "bits" for "molecules". Software receives no special protection like you assert above, the algorithm/process and implementation protected by patent and copyright respectively, much like it is for everything else.
Of course, the elephant in the room that everything is an algorithm and that there is no mathematical difference between bits, molecules, hardware, software, or data. It is a distinction with no theoretical difference and people keep trying to patch up the law so that we can pretend a distinction exists in the face of clear evidence to the contrary; copyright has a similar issue, by the way. It has led to absurdities such as an algorithm on Intel x86 not being patentable in Europe but the same algorithm on a Xilinx Vertex-5 is -- the distinction between the two is arbitrary and capricious. The problem is not algorithm patents (and much of what we are talking about here is business process patents, not algorithm patents per se) but that so many patents are frivolous, but that is not a problem unique to any particular field of patentability. The two biggest problems are really frivolous patents being rubberstamped, and theoretically inconsistent treatment in a few narrow areas that are then bleeding over into other areas as the inconsistency becomes obvious in real cases. The only way to actually fix these inconsistencies is by adopting an all or nothing policy; I have no particular opinion on which way that goes.
If you eliminate accidents and homicide from the statistics, Americans live longer. Period. Do you realize how radically better US healthcare would have to be for your assertion to hold up in the statistics? Direct healthcare outcomes for the average American, rich or poor, are better than the rest of the industrialized world, but not that much better. You cannot juggle the numbers to make that fact go away, and its reality is well documented in the medical journals (e.g. the recent Lancet Oncology study). The problem with Americans is demographic, genetic, environmental, and behavioral.
You can try and dodge the elephant, but it is quite large. A poor person with cancer in the US has better survival odds than a poor person in the UK's NHS. Every state in the US has a public healthcare system for the poor, incidentally, since the US Federal government has no jurisdiction in these matters. The idea that there are people without access to medical care in the US is false; most of the argument is over whether there is a more efficient way. As someone raised on free public healthcare in the US in multiple states, it never ceases to amaze me that it supposedly does not exist. This is mostly just ignorance from foreign perspectives that fail to realize that individual US States are essentially the equivalent of countries in the EU and that public healthcare is dealt with at that level of government.
You are conflating demographic and environmental factors with healthcare outcomes.
If you, for example, remove non-medical causes of premature death (car accidents, homicide, etc) Americans outlive other industrialized countries. Healthcare is only a small factor in life expectancy, and average healthiness is almost completely unrelated to healthcare in the industrialized world. The environmental and demographic factors are atypically poor in the US relative to the industrialized world.
If you look at direct measures of healthcare outcomes, such as diagnostic accuracy and disease survival rates, the US leads the industrialized world by a large margin. The elephant in the room in the recent Lancet Oncology study, for example, was that cancer survival rates in the US are much higher than in any other industrialized country in the world -- about 20-40% on average depending on the country and the cancer. So in this sense, Americans are paying more but they are also receiving much more.
The real situation is that the US has terrible non-medical factors that drag down its statistics but compensates with the best average medical outcomes by a huge margin. In most of the rest of the industrialized world, you have middling to good non-medical factors and middling to poor medical outcomes. In other words, the aggregate statistics are not measuring the same thing. Since we pay the medical establishment to produce positive medical outcomes, it would seem prudent to evaluate their efficacy based on those results and not on the number of automobile accidents people are involved in.
At a minimum, it would be foolish to trash a medical system that produces results such as cancer survival rates that no other system is currently coming close to. The US system may be byzantine and inefficient, but it also outperforms the rest of the world in the key metric of medical results. Let's not throw out the baby with the bath water, at least not until a national healthcare system exists with equivalent medical outcomes.
Actually, you are demonstrating some cluelessness here. Size bloat is only a small part of why XML massively sucks as a wire protocol compared to functionally equivalent universal representations such as ASN.1 which were designed to be extremely fast wire protocol encodings; compression does not address much of the performance suckiness. The reality is that good wire protocol representations are about an order of magnitude faster than equivalent XML formats, which matters a lot of you are moving any kind of traffic. Since ASN.1 is an ITU standard, it is usually used for networking protocols (as opposed to application protocols). There is even an ASN.1 encoding standard called XER that losslessly encodes XML as ASN.1 and back so as to improve its characteristics as a wire protocol. There are other wire protocol encodings out there, though ASN.1 is probably the most prevalent.
If you think compressing XML makes it a good wire protocol, you are simply too ignorant or too inexperienced to be designing or implementing wire protocols. Study the design of some of the binary wire encodings and maybe you will figure out the subtleties that make them so much faster. There is a hell of a lot more to wire protocol scaling and performance than the number of bits going over the wire. Methinks you stepped a wee bit outside your area of expertise.
"I just think you've never been outside USA (and that's a mild way of putting this)."
Former European citizenship aside, I think I spent more of my life in Asia than Europe (I've never really done the arithmetic). Is that what you are referring to? Do not confuse me with one of those people that has never been anywhere. On the other hand, one might suspect that of you...
As for health care outcomes, that is standard fare for health care economics and the most scathing indictments of the European health care system in terms of health care outcomes actually come from Europe. Go ahead and Google on cancer survival rates between Europe and the US. The statistical difference is pretty stark. Europeans pay less, but if you have anything serious you do not want to be anywhere but the US. Remember, those survival rates are *average* and outcomes track in income in Europe the same way they do in the US (another dirty secret well known to health care economists).
As I said, ignorance knows no bounds.
- As a financial instrument, insurance exists to distribute risk, not cost. Anybody who does not understand what the distinction is please vacate the discussion. Technically speaking, insurance is how one distributes risk and some approximation of communist government (in a literal rather than pejorative sense) is how one distributes cost. Trying to use the former to approximate the latter is inefficient and raises the costs for everyone.
- For all the inefficiency and expense of what passes for health care in the US, the US also has the best health care outcomes in the industrialized world and generally by a wide margin. If you have cancer, your survival rates in the US are much better than Europe on average and the best in the world in absolute terms. This is true by a number of other direct metrics of health outcomes and holds across the population even if you are an average person and not a wealthy person. Americans are paying more but they are getting more, and survival rates for a pretty broad swath of nasty things is 20-40% better, not buried in the noise floor. This is the good part of the US health care system that no socialized system has ever emulated. Americans compensate for being unhealthy (and car accidents, etc) in mortality rates with really good medical outcomes. If you normalize for genetics and environment, Americans live longer than anyone else. Of course, many Americans have crap genetics and have a crap diet as far as longevity is concerned.
- Before the Americans get too smug, the American health care "system" (there is no system, it is a market) is byzantine and inefficient. It should cost nowhere near what it does even for what Americans get.
- All Americans have health care, even those that cannot afford it, and the idea that there are people without access to health care is a myth that inflames the clueless and serves the purposes of political propaganda. The quality is mediocre, but what do you expect with socialized medicine. It is not hypothetical, I was one of those invisible souls raised on government health care for the destitute.
What we really have is a number of facts. The European system produces mediocre results in terms of actual health care outcomes (what we are nominally paying for), but it is relatively inexpensive. Americans pay a lot but have the best health care outcomes in the world. Americans pay far more than is strictly necessary by any reasonable metric, but I guess they can afford it. Americans are also bearing the cost of most medical technology innovation, amortized in the American medical market; when is the rest of the industrialized world going to carry their fair share of that burden?
In short, all the systems suck. That said, I would be reluctant to give up the superior health outcomes (what I pay doctors for) and medical innovation of the American medical environment. On the other hand, I wish they were more efficient at what they do. Clearly there has to be a better way, but by every metric that matters to someone getting health care, replicating the European system is not it.
"*cough*JDAM*cough*Tomahawk*cough*F-16*cough*
Sorry got something stupid stuck in my throat. Anyways as I was saying the US has plenty of weapons sytems that use GPS systems and would either become completely ineffective or seriously crippled without it."
As was pointed out elsewhere, neither JDAMs nor Tomahawks (nor F16s for that matter) use GPS guidance -- only technically ignorant tools claim that any US weapon systems use GPS guidance. Even rudimentary research shows that systems like JDAMs and Tomahawk get their navigation data from ultra-precise laser ring interferometers (an extremely precise solid-state optical accelerometer with a purpose similar to mechanical gyroscopes), with the ability to optionally accept GPS fine-tuning corrections. The precision of the inertial systems is classified, but it is generally known that it is apparently not much worse than the GPS corrected version for weapon targeting purposes (they may have already converged for all we know -- INS has been continuously improved, and it wasn't bad to start with). Note that this is also why jamming or toying with the GPS signal does not send bombs and such flying way off course; the inertial system only accepts corrections within the computed error bars of its own positioning data. If the GPS signal is outside those bars, the navigation system assumes the GPS has been compromised.
I just gotta love the armchair weapon designers who think that of the thousands of bright engineers that built all these weapons, it never occurred to anyone that someone might jam or disable the weak RF signal that is GPS. GPS is a convenience, not a necessity, and it would only have small impact on the efficacy of American weaponry -- by design. The reason the US military embeds inertial navigation in everything is precisely because it is essentially impervious to countermeasures short of altering the physics of the universe. For all we know, nominal GPS weapon targeting is largely for show and indirection, since it costs next to nothing to strap a GPS receiver on an inertial navigation system (and it would seem to have worked in that case, considering how much breathless idiocy is fixated on GPS weapon targeting).
As a point of trivia, the laser inertial navigation systems used by the US military were invented in the 1960s for some (failed) ballistic missile interceptor research. Even though the ABM research was a dead end, the laser ring gyro technology developed for the project paid dividends for the US military way out of proportion to the money expended on the ABM research.
"The US military is completely dependant on their technology and the rest of the world knows it. Do their cruise missiles even work without GPS?"
The US has no weapon systems that are GPS guided and never has, precisely because it is vulnerable. The Chinese may have just now gotten around to developing anti-satellite technology, but the Soviet Union had it ages ago.
The core guidance package of US weapon systems is extremely high precision inertial navigation (all systems described as "GPS-guided" are actually inertial -- the media is a bit stupid about these things, as GPS is an optional untrusted overlay on inertial navigation systems). Some intelligent terrain following weapons also use optical geo-referencing. As a matter of policy going back to the Soviet Union days, the US military machine views satellite systems as "nice to have" but its infrastructure is pervasively designed to operate under the presumption that there are no satellites in orbit. The vulnerability of the US military to massive system outages is greatly overstated; the Soviet Union was a much bigger threat on this scale than the Chinese are, and the US military has always been pretty religious about designing systems whose functionality was robust and in the face of rapidly degrading military infrastructure and relatively decentralized. It is easy to forget it, but the Chinese have nothing on the old Soviet Union in terms of technology and force numbers, and that was the doctrinal enemy of much of the modern US military.
"They need to add a GOOGLE function to allow queries to be searched nicer.
SELECT * FROM articles WHERE GOOGLE('boobies');"
I don't know about MySQL, but you can pretty trivially write extensions to PostgreSQL that do exactly this kind of thing. In fact, I've written a number of such extensions to PostgreSQL that make Google resources seem like local resources in a PostgreSQL database. (This kind of deep and easy customizability is one of the things I find to be a killer feature of PostgreSQL relative to many/most other databases.)
...will be a machine, which may or may not be controlled by a techie in an air-conditioned office.
The "pay raise" is a bribe by the board for key executives to not jump ship. It is a common pattern when companies are in trouble to put golden handcuffs on key people to keep them around to see the trouble through to the end rather than leaving for better climes with fewer risks. If you think about it, it makes a certain amount of economic sense.
...smart terrorists only use peroxide-based explosives (like the London subway bombing et al), oxidized halide based explosives (e.g. chlorate), and various other dirt cheap and ubiquitous explosives. While many of the most famous explosive chemistries might be subject to nitrate tests, the range of explosive chemistries that have been used at various times is far more diverse than nitrates. First World War mortar explosives are as dangerous today as they were back then, even if some of them do not contain nitrates.
The fixation on the detection of nitrate and related chemistry is a bit of a blind spot in explosive detection technology.
"Anyone want to guess how long it'll be before Victor finds himself out of a job?... Unfortunately..."
Before you don the tinfoil, a basic education on the US system of government might be in order. The Federal judiciary are appointed for life and only substantial evidence of gross misconduct puts them at risk of removal by a super-majority of Congress.
"No, copyrights are for written works or other forms of expression, not implementations of arbitrary ideas. The "implementation that produces" the chemical is also not copyrightable. A document which describes the implementation is copyrightable, the process for making the chemical is patentable at best."
Definitely incorrect. There have been numerous copyright infringement suits, rather large ones, for cloning hardware implementations. Or at least, this has been true in the US for a very long time. There are many types of technical hardware that you can buy overseas that is not sold in the US because it was copied from US-based designs. In many such cases, it is not a matter of patent protection or even trade dress but e.g. the circuit board layout being copied.
Copyright covers a hell of a lot more than "written works". Forms of expression included, broadly, implementation.
The same argument could be applied to every other industry, for better or worse. Which is my only point.
A patent system that is not consistently applied is worse than useless.
Chemical process patents are basically algorithms, but done with atoms and molecules instead of bits. The implementation of the abstract process is subject to its own IP protection independent of the patented process. In theory, a more efficient way to move atoms around to achieve a particular result should have no more and no less protection than a more efficient way to move bits around to achieve a particular result.
Software patents are a bit of a misnomer, further polluted by frivolous patents (which all fields suffer from). A software patent, ideally, is a new method/algorithm for doing something in the bit domain e.g. a more efficient method for doing something we can already do. Good software patents generally do not restrict the abstract implementation of a result, just a particular method of obtaining a result that may be more efficient than previous methods in some fashion. I will readily acknowledge that some so-called "software patents" are idiotic, but that is independent of whether or not the idea is valid in the abstract -- no accounting for USPTO incompetence.
Frivolous patents are hardly unique to software, and are one of the main problems with the patent system.
All implementations of patents are copyrighted. It is the nature of the thing, so let's stop pretending like this is unique to software algorithms. Chemical process patents are fundamentally indistinguishable from software patents in all respects -- including copyright -- yet we ignore them and their long history. I get the impression that posters on slashdot are so clueless about other fields that they think software is special in this regard.
This issue will not be solved until people accept the mathematical truism that hardware patents and process patents are indistinguishable from software patents; one of the accomplishments of 20th century mathematics was proving that fact. Either we accept that algorithms can be patented or not, recognizing that it is *all* algorithms, or we fumble around with broken law that refuses to acknowledge reality (which would not be a first for government).
"An attack with those babies will not be stopped by the current generation of missile defense systems. It is _not_ a completely ballistic rocket. In other words predicting, calculating it's trajectory and using all the billions of dollars of infrastructure designed for ballistic missiles is not as useful anymore."
You severely underestimate and misunderstand the operation of modern ABM systems. Current US ABM missile technology uses broad-spectrum imaging based terminal guidance, exactly like air-to-air missiles and similar (in fact the terminal homing software is largely identical). It has the same ability to follow a non-ballistic trajectory that all other missiles with this type of guidance do, and these targets are far less agile than, say, a fighter aircraft. The "new technology" in state-of-the-art ABM was the ability to marry this kind of terminal guidance package to a rocket that operates in the ABM envelope (though most of that "technology" is little more than extremely advanced materials science).
The Russians are talking nonsense and preying on the ignorance of most people regarding how modern US ABM systems actually function. The design of the US ABM system was to defeat all targets, agile or otherwise, but extending the envelope of target capability well into the ABM range rather than just local air defense.
"The transportation and residential sectors combined make up more than half of our greenhouse gas emissions. I'm not talking about the switch from a Tahoe to a Prius, but from a Tahoe to something like a Chevy Volt (sized for real-world use, of course). Given most people's driving patterns, that could cut our CO2 emissions from transportation by more than half."
You are playing games with statistics. Coal and natural gas alone account for more than half of all CO2 emissions in the US. How many people are using coal and natural gas in their vehicles? Of the less-than-half that is left, less than half of *that* is due to motor vehicle emissions. And most of those motor vehicles are not SUVs.
In short, you buy something like a 5% net reduction in CO2 output for some very heavy-handed and moderately costly motor vehicle regulation. Pissing in the ocean. Simply moving away from fossil-fuel based power production to nuclear and geothermal (and others when it makes sense) would slash CO2 output in approximately half for a moderate one-time cost that would be no more than the cost of making everyone drive tiny cars. It would also be a largely transparent change that would not significantly impact quality of life for most of the country.
A much, much simpler way to reduce fuel consumption in vehicles is to stop taxing diesel much more heavily than gasoline. Diesel is a less expensive fuel *and* it gets substantially better mileage (moreso than gasoline hybrids in most comparable cases).
"If political will existed we could have vehicles that do everything SUVs do now, and houses just as big and comfy as the ones we have now, that had only a small fraction of the environmental impact. The technology is there; it's just a question of making it economically feasible."
Considering that the environmental impact of SUVs versus, say, a Prius is negligible in the big environmental calculus, I would say that it is an entirely unimportant point. On the list of things that will have the most impact on the environment that is way, way down the list in both terms of impact and return on investment as far as the environment goes.
People routinely conflate oil trade issues with environmental issues -- they are not the same. Eliminating all SUVs tomorrow will have a negligible impact on CO2 production or most other environmental issues people seem to care about. Eliminating SUVs tomorrow would impact the geopolitics of the oil trade, but that has very little to do with environmental impact. A lot of people get hot and bothered about SUVs, but they are pretty much inconsequential in terms of the environment if you look at actual relative impact numbers. If environmental results are all we are looking for, spending energy on SUVs offers an incredibly low return on investment.