At the distances involved, optical measurements will have much more resolution. Radar loses out on the resolution to optical before you even get to the moon's orbit. More data points are great, but we need higher resolution data than radar can provide at that range.
Or at least that's what I've been reading from the astronomy folks.
The real advantage of AMD's 64-bit system architecture has less to do with its number crunching performance and more to do with the scalability and speed of its I/O and memory architecture e.g. HyperTransport. There is not much to differentiate Intel and AMD in the absolute number crunching arena, but if you start looking at scaling memory, I/O, and the number of cores that can work together, AMD64 leaps ahead.
If the big advantage of these new 64-bit processors is nominally found in servers, then AMD will clean house because their systems scale and perform VERY well in the server role compared to Intel. Sure, you may not be able to tell the difference between AMD and Intel on the desktop, but for most types of server loads, there is no contest. The Opterons are very, very good server systems, and for many types of loads e.g. database servers, they run rings around Xeon processors for a very low cost.
Unless Intel matches a very competent ccNUMA and I/O fabric to their EMT64 cores, they will not be competitive where it matters.
Probably server overload then. They have multiple servers, and I imagine the load is not distributed evenly depending on the kind of customer you are. My original point, was more that there was no way they were network limited. Disk I/O might be another matter altogether depending on what systems you are connected to.
If the most you've ever gotten is 36k, the reason is that your network connection is poor.
Their networks are fiber connected at the core on a high-performance ring from Neopolitan Networks, so there is no shortage of capacity on their end. That network can handle an order of magnitude more bandwidth than is used to drive Keyhole without even breaking a sweat. I'm also on Neopolitan's network, and Keyhole performance is great. I suspect the bottleneck is closer to your end.
Except that there are as many Democrats defecting to Republicans as there are Republicans defecting to Democrats, at least in my experience. The Democrats are not particularly motivated, and a great many I know think Kerry is a pompous asshat, such that they really don't care who wins even though they do not like Bush. They despise Bush, but they don't like Kerry either even though they'll vote for him.
And in fact, that is why the Democrats will lose the election. Out of all the people they could have selected, they select a flagrantly elitist blowhard with no definable position and an obvious lack of charisma. Ugh. There really is nothing to get excited about there, and it is apparent that a lot of Democrats don't really believe in Kerry. Other than the libertarian wing of the Republican party (which is, sadly, fringe), the Republicans genuinely seem to like Bush, for better or worse. I've definitely noticed an erosion of support among the old school blue collar life-long Democrats, many who feel that Kerry is completely out of touch with their reality.
The Democrats had a real shot, right up until the point they selected Kerry. Mind you, I don't think it was obvious just how lousy of a candidate he was going to be before they selected him. Howard Dean would at least have been interesting, and even someone like Gephardt would have done better shoring up the base. Right now, they are chasing down votes they should have already owned.
Which kind of begs the question as to how we ended up with a couple of clowns to choose from in the first place. What happened to really great candidates that you could feel good about voting for?
IMHO, some of what you're really looking at is population density. The rights of my fist end just before they touch your nose. The further away you are, the more rights my fist have - and yours, for that matter. Pack us in tighter, and there just isn't that much space to swing my arms.
This is a popular myth, a misconception based on dividing the population over the entire State or region, the vast majority of which is completely unpopulated due to Federal encroachment.
It surprises most people to find out that the mountain West is one of the most urbanized regions of the United States, ~80% urban (also the youngest population). Because the Federal government continues to acquire all the land in those States, the amount of land people can actually live on is quite small, and so everyone lives in a couple urban areas in those States. If you look at a list of US cities ordered by population, you'll find that many of the larger cities are actually in the mountain West. Basically, you have States the size of countries like Germany, while only being allowed to use an area more the size of New Jersey or New England states.
In fact, as State populations go, the mountain West States are right around the median for the country, and several very large metros (read: well over 1M people) are located there, e.g. Phoenix or Las Vegas.
The difference doesn't have to do with the amount of people in the space -- Westerners are actually packed into cities more than most people east of the Rockies -- but a fundamental difference in attitude. As I mentioned in the original post, the Federal government has had a long history of antagonizing and exploiting the Western States, which has made those States culturally and institutionally anti-government. Combined with certain historic economic realities of the mountain West, a very libertarian outlook evolved and became a part of the culture. The government used to get away with it because at one time those States really did have tiny populations. Now, those States have millions of people in them (though this fact hasn't caught up with the politicians back east), and the culture has a long memory.
It has long been a truism that there is a geographical philosophical split between those areas east of the Rockies exclusive and that west of the Rockies inclusive. Right now, this is mostly evident within the Republican party.
West of the Rockies, and in the mountain West in particular, the core political ideology of the region tends to revolve around a small-government, non-interference, live-and-let-live perspective -- real believers in rugged individualism. There are many historical reasons as to why this is that go back a century or two. While the people that live there are often conservative as individuals, they generally are not socially conservative in that they try and legislate the behaviors of society. A built-in distrust of government is stronger than their desire to control what other people do. East of the Rockies, big government social conservatism is deeply embedded in the culture.
Libertarians and similar have long held relatively strong positions in the mountain West due to the fact that Eastern conservatives often control conservative politics, primarily because of population differences. People like Bush reflect only the conservative issues that are unique to Eastern conservatives while not reflecting the issues shared by Eastern and Western conservatives. When more extreme examples of this come down the road in the Republican party, it tends to lead to defections to the Libertarian party out West. It is an old political and ideological tug-of-war.
In fact, if you look at the core philosophical components of Western conservatism, it is essentially libertarian. Which is why there are far fewer restrictions on what you can do and what you can own in the "conservative" mountain West than in "liberal" states, ironically. Nevada makes California look like a socially conservative police state by comparison if you actually compare laws, and they are next door.
From the standpoint of the publishers (and my girlfriend is a national sales manager for one of the very large publishers), this idea is incredibly bad and they have been extremely resistant to it. Amazon has been trying to push this for years without success, and it seems that now Google is getting in on the game. Or maybe Amazon is trying to use Google as additional weight to try and break the stubbornness of the publishers.
The problem, ultimately, is that showing the page you are looking for, plus or minus two pages, is often all the pages you need to see for a great many bookes e.g. books that are randomly accessed in a reference fashion. As an example of this, my girlfriend routinely searches cookbooks online using this very feature. It shows her the recipe she was looking for from an expensive cookbook, and plus or minus a couple pages, which means she gets the entire recipes -- the primary benefit of the book -- online for free. And she uses this as an example of why her publishing houses won't participate.
For STM publishers and similar, 90% of their product line could be used this way. Letting Amazon (or Google) give away book content in a searchable format five pages at a time would dramatically eat into their sales without generating any revenue. Most of the books you do see in this system are either 1) books from minor publishers too stupid to have thought this through, or 2) a very short list of throwaway books from major publishers to prove to Amazon and themselves that it actually eats sales rather than driving them -- the consensus of the publishing industry. It would have died a long time ago except that it is the pet project of someone high up in Amazon.
The research that went into many software algorithms was non-trivial. I work with (a still unpublished in any form) family of algorithms that actually required an 8-year R&D effort to develop, and with high commercial value. Why should this receive less (hypothetical) protection than some worthless mechanical trinket someone threw together in their garage?
It seems like you are merely arguing against frivolous patents, a completely separate issue. Most mechanical patents are crap, just like most software patents. But because this is slashdot, we focus on the software ones. I have seen no real valid argument why high-investment algorithm R&D should not be patent protected while a widget with similar investment in time and money should be. We already know there is no difference between "machine" and "software" theoretically, so that can't be the issue.
If algorithmic information theory has taught us anything, it is that everything is fundamentally a mathematical process. In this fashion, software is no different than anything else.
Again, we must either decide on a universal and consistent patent system, or none at all. Arbitrary and capricious decisions as to what can and can't be patented are certainly not the way to go. Personally, I'd be happy if we just cracked down on frivolous patents, a problem that plagues patents entirely unrelated to software as well.
What do you mean, "what data"? Try the Vostok ice core data for starters. People (on both sides) love to quote what they've heard about the data that seems to support their position but it seems that hardly anyone has actually studied it themselves.
There is a lot more ambiguity in the data we have than is commonly presented -- that's why both sides can use it. You soundbited your way to an argument without understanding the context. I'm not disagreeing with global warming, I'm disagreeing with the shoddy construction and specious reasoning of your position. All heat, no light. If you end up at a good result from poor premises, it is only because you were lucky.
Unfortunately, many of the most vocal people on both sides have about as deep an understanding of the issue as you do. I object to letting a mob of fools be heavy-handed with people using a specious argument. It is no different than fundamentalists forcing everyone to learn Christian Creationism in schools. If people don't want to be bothered to understand something from the first principles up, I don't want them making policy. The ability to assert an axiom does not make an axiom valid or relevant to the issue at hand.
Since the industrial age has begun, the amount of atmospheric carbon dioxide has increased from around 280ppmv to 380ppmv.
I see that you are trying to make a correlation here, so answer me this: How do you explain CO2 concentrations that rapidly increase with a similar magnitude, pre-industrial revolution, pre-history, and even pre-homo-sapien? And why is CO2 concentration a lagging indicator of warming in previous cycles?
The climate is changing, but people trying to propose solutions based on cherry picking and pulling data out of context to suit some particular ideological viewpoint is a lot of foolishness that will create more problems than it will solve. It would make things seem so easy and ideologically pure if anthropogenic CO2 was the fulcrum of climate change, but it isn't that way in fact. There are a hell of a lot of other processes that are major contributing factors that will have to be acknowledged if a real solution is to be found, and a great many of those have nothing to do with CO2. It could very well be the case that all manmade CO2 sources could be eliminated tomorrow and it still wouldn't make a substantial difference in the climate trajectory.
The climate isn't a sitcom. You can't get a neat and tidy solution to anything in a half hour, and as is true of many types of systems, we may in fact be impotent when it comes to directing outcomes in a meaningful sense. Changing our CO2 emissions is far from free, and it would be prudent to study the systems more thoroughly before putting ourselves on a path that could find us in an even worse position than if we did nothing at all. Jumping on a ridiculously simplistic bandwagon as a solution in what is largely an absence of good knowledge is a sure way to generate a hell of a lot of heat with precious little light.
A lot of people are eager to jump on a bandwagon completely oblivious to the long-term consequences and unintended consequences that might occur -- we could end up screwing ourselves far worse than climate change. Ecosystems, fundamental economics, and a great many other things are profoundly effected by CO2 levels. You can't change global temperature in isolation by playing with CO2 levels, and the result of doing so naively may end up causing more harm than it prevented. And because of the changes made and their consequences, we may find ourselves in a situation where we are far less capable of dealing with the new mess we made.
China seems to have achieved the social stability and unity of purpose normally associated with totalitarianism, without sacrificing the rising standards of living afforded by capitalism.
Ugh. Have you actually been to China? It is still clearly a "third world" type country, and will be for the foreseeable future. What makes you think this model scales?
If the standards of living are rising, it is only because they were so poor to begin with. Efficiency isn't such an important factor because you can buy big improvements for little money. But if you try to scale this to the standard of living of, say, Europe, you end up with a non-competitive economy because standards of living are pretty much defined by relative efficiency.
Centralized political economies do not scale well, and are not competitive at the high-end.
I can't wait to see this die, and I wish sunsetting would be used for all our laws.
The so-called "Assault Weapon Ban" was nothing more than a feel-good measure that had nothing to do with crime or safety. All it did manage to do was annoy and/or piss-off people who buy or own guns. No appreciable benefit to any constituency, and a big downside to a rather sizeable constituency. It is no wonder that most politicians don't want to touch this issue, and Bush knew full well that it would never end up on his desk. If you ignore the Democrats in "safe States" like California, who can soapbox on this issue all day without consequences, it is a "third-rail" issue everywhere else whether you are a Democrat or a Republican.
The 1994 Congressional blood bath pretty much assures that gun control won't be touched again for a long time.
Okay, a lot of people here (not just the poster I'm responding to) need to get themselves a serious education on the US form of government.
The President is NOT the representative of The People, the President is the Executive of the States -- idiots who vote in the Presidential elections who don't understand this shouldn't be voting. Up until the early 20th century, people didn't even vote for the President, the President was selected by the State governments, and was accomplished via the Electoral College. Short of discarding the Constitution and starting over, this will remain the case, as discarding this would break some of the basic axioms of the US Constitution. The President represents the States, not the People.
Hence why they are called "States" and not "Provinces". A "state" in the US is very much akin to a "country" in the EU, and is a sovereign entity. This is why laws concerning a great many things vary so widely across states. Hell, that's why even the system of Law varies across the States. Beside the default English Common, you officially have Spanish Common, and Code Napoleon (and probably others) which are recognized in some States.
The 25% increase in problem complexity creating a 100% increase in solution complexity is probably reflective of what is effectively a combinatorial explosion in implementation required to make all the various parts of the model play nicely together under all circumstances.
In fact, I've generally held that the complexity of implementation is generally an exponential function of the general complexity of the problem. Allowing additional degrees of freedom in a design is typically very expensive. You aren't just adding that additional degree of freedom to the design, you have to make the rest of the design aware of that new dimension and put it into implementation.
Liberals come from cities, Conservatives come from rural areas.
Not really true. Population density of a state has little to do with how urbanized a state is. A western "county" is as large as an east coast "state", and unlike the east coast, many western counties are largely owned by the federal government (read: cannot be populated). So the county will look sparse in terms of persons/square mile, but in reality everyone lives in one or two cities and towns -- totally urban. For example, the mountain West is one of the most urban (>80% in many of these states) and demographically youngest regions of the country and also quite conservative. The states only have average populations, but most of the people live in a few large cities (in no small part because the Federal government continues to take all the land in those States).
So I don't think it is a rural versus urban issue. States like Arizona and Nevada are very young and urban, moreso than most of the traditional "liberal" States, but are also notoriously conservative in many ways, though in the West it is a very libertarian kind of conservatism.
Seriously. I see all these reasonably well-paid geeks that were making decent money for years who are in the gutter in less than 12 months after becoming unemployed. There really is no excuse for not having a decent amount of money saved up after all this time.
I have only limited sympathy for people who don't save any money and the expect everyone to break out the violins when it bites them in the ass. It isn't the the fault of the government, Bush, or anyone else that you are fiscally irresponsible. I'm not exactly the definition of fiscally responsible, but I managed to scrape by for two years while unemployed and another year and a half working at a startup where I make so little I might as well be unemployed.
So quit yer whinin'. I don't want to hear it. You pissed all your money away on frivolous shit for years on end (and yes, without knowing anything about you I'll bet you spent at least half your money on frivolous shit -- I used to be the same way) and now its time to pay the piper. What a pathetic lot.
I don't know. We have been running Opterons here for some time and they rock. Some of the best performing iron you can buy for any amount of money.
I don't know what issues you are having, but the Opterons are friggin' golden; we run them on the Arima/Rioworks MBs and have had zero problems. A 1.6GHz part runs just over $200 and schools just about everything else for heavy duty work. It feels like high priced iron but comes at a fraction of the price. Everything mission critical runs on these puppies.
You may doubt AMD, but I don't and we use them for mission critical roles. The Opterons are ridiculously good for the price, a real professional product.
Perhaps Im wrong and this cariculum will teach excelent data structure usage, and algorithim analysis and AI and compiler design and low level architecture. But at this point i kind of doubt it.
Does it really matter? Most people with a 4-year CS degree have at best mediocre knowledge and abilities in these things as well out here in the real world. Most 4-year CS folks may have had to take a bunch of classes on more esoteric aspects of CS, but most don't leave school with useful knowledge beyond that of any other competent code monkey -- they largely forget all the esoteric stuff after 6 months in the real world.
The problem with CS is that you don't actually use 90% of the esoteric stuff ever once you leave school and get a real job, making them indistinguishable from the 2-year folks (or 0-year folks) in relatively short order. Above a certain baseline skill level, motivation and natural ability are pretty much everything. I work with CS PhDs who actually do real work, and even these folks are pretty clueless about theoretical computer science. They may have dabbled in it at one time, but working in the real world rotted their brains.
You've got to be a kidding. Setting up a website is not "bringing it to market". Only someone stuck in some kind of engineer or geek-centric perspective would think that it is. Bringing a software (or any) product to market is seriously capital intensive, particularly if you are competing against ripped off clones of your own product. Copyright doesn't protect you against incumbent billion dollar marketing machines.
So you spend 3-5 years on the ultimate piece of software with a whole collection of spectacular algorithms never seen before in computer science, and put it on the web at your little website that no one has heard of. Among the trickle of people that buy your product is BigCorp, who make their own implementation of your secret algorithm sauce and sell it through their billion dollar marketing machine to every PHB on the planet. You barely cover the cost of your website, and BigCorp makes hundreds of millions on your research and development effort. You disappear into obscurity, not making a dime.
In short, without a patent the algorithm inventor would be at a HUGE disadvantage in the market in virtually every single aspect of doing business. Copyright doesn't help with this, only a fool thinks "bringing something to market" is practical in the complete absence of design protection and very limited capital, particularly if the design is genuinely worth something in the open market.
So its either trade secrets or patents. Both have their upside and downside. The "copyright is protection enough" argument is a sham. Its virtually no protection at all.
People need to either accept the idea of all patents or no patents. You can't legitimately have it both ways and be consistent.
1.) The argument that software patents are bad because most software patents are frivolous is a strawman. Most non-software patents are also frivolous. This is not unique to software, people here only notice software patents because they work in software. If you don't like the frivolousness of patents granted then deal with the general frivolousness of the patent system, because "software" patents and "frivolous" patents only intersect and neither contains the other in its entirety.
2.) Software is hardware is software. If you can't patent software, then you shouldn't be able patent hardware either. This was settled the better part of a century ago, in case anyone was paying attention. Pretending that software is a special case that is different from hardware creates a distinction where none exists.
3.) A minority of software patents, just like a minority of normal hardware patents, cover inventions that took substantial research and development effort that no one could reasonable claim to be "obvious". This would seem to be precisely what is supposed to be protected by patents under ideal circumstances and I can't see a reasonable argument that says these inventions should be treated differently than all other patents. Otherwise we would be in the position of allowing frivolous non-software patents and disallowing heavy-duty substantive "software" patents.
That said, it seems to me that the biggest reason a lot of geeks don't like software patents is that it is inconvenient for them. Every rationalization that disallows "software" patents but allows other types of patents has been uniformly weak and inconsistent. If you think patents are bad, then ALL patents are bad, not just the ones you wish you didn't have to abide by.
Shared memory can easily be simulated on a gigabit network
Not just no, but hell no. One of THE major points of having shared memory is that such systems have an order of magnitude (or two) lower memory latency. Seeing as how memory latency defines the performance bound for most apps that just happen to run well on shared memory systems, there is no way you can reasonably use gigabit network fabrics as a substitute. Unless you can afford greatly reduced performance to save a few bucks.
When gigabit networks have typical latencies like that of a ccNUMA system, this will be worth considering. But we are still pretty far from that today, and it isn't a trivial engineering problem to get there.
There is already a thriving market for this
on
Ethernet at 10 Gbps
·
· Score: 1
Requests for multi-gigabit circuits across Ethernet fabrics are already moderately common, particularly as more longhaul and backbone infrastructure moves to pure Ethernet. You can do this today by channel-bonding lambdas on a fiber with Gigabit ethernet (the silicon will support it), but 10G would make it that much easier.
The most common uses for this: long-haul video transfer, massive network file systems, and extremely high-volume internet servers. Massive distributed private file systems and applications are hot, and will be a major driver of high-performance fiber ethernet fabrics.
For most purposes, a properly architected network will add almost no real latency beyond what is required due to physical limits i.e. the speed of light. Regional ethernet fabrics should be easily sub-millisecond between any two points in the region, which is quite adequate for most applications that might be amenable to being networked to begin with. Most people still think of networks in terms of routing bottlenecks and other topology biasing issues like that, but large-scale ethernet fabrics, an increasingly popular choice for network providers big and small, make most of these issues go away for a lot of purposes and mitigate the performance hit for most others. 10G will just make ubiquitous ethernet fabrics that much more scalable.
Slashdot Mirror
Or at least that's what I've been reading from the astronomy folks.
If the big advantage of these new 64-bit processors is nominally found in servers, then AMD will clean house because their systems scale and perform VERY well in the server role compared to Intel. Sure, you may not be able to tell the difference between AMD and Intel on the desktop, but for most types of server loads, there is no contest. The Opterons are very, very good server systems, and for many types of loads e.g. database servers, they run rings around Xeon processors for a very low cost.
Unless Intel matches a very competent ccNUMA and I/O fabric to their EMT64 cores, they will not be competitive where it matters.
Probably server overload then. They have multiple servers, and I imagine the load is not distributed evenly depending on the kind of customer you are. My original point, was more that there was no way they were network limited. Disk I/O might be another matter altogether depending on what systems you are connected to.
Their networks are fiber connected at the core on a high-performance ring from Neopolitan Networks, so there is no shortage of capacity on their end. That network can handle an order of magnitude more bandwidth than is used to drive Keyhole without even breaking a sweat. I'm also on Neopolitan's network, and Keyhole performance is great. I suspect the bottleneck is closer to your end.
And in fact, that is why the Democrats will lose the election. Out of all the people they could have selected, they select a flagrantly elitist blowhard with no definable position and an obvious lack of charisma. Ugh. There really is nothing to get excited about there, and it is apparent that a lot of Democrats don't really believe in Kerry. Other than the libertarian wing of the Republican party (which is, sadly, fringe), the Republicans genuinely seem to like Bush, for better or worse. I've definitely noticed an erosion of support among the old school blue collar life-long Democrats, many who feel that Kerry is completely out of touch with their reality.
The Democrats had a real shot, right up until the point they selected Kerry. Mind you, I don't think it was obvious just how lousy of a candidate he was going to be before they selected him. Howard Dean would at least have been interesting, and even someone like Gephardt would have done better shoring up the base. Right now, they are chasing down votes they should have already owned.
Which kind of begs the question as to how we ended up with a couple of clowns to choose from in the first place. What happened to really great candidates that you could feel good about voting for?
This is a popular myth, a misconception based on dividing the population over the entire State or region, the vast majority of which is completely unpopulated due to Federal encroachment.
It surprises most people to find out that the mountain West is one of the most urbanized regions of the United States, ~80% urban (also the youngest population). Because the Federal government continues to acquire all the land in those States, the amount of land people can actually live on is quite small, and so everyone lives in a couple urban areas in those States. If you look at a list of US cities ordered by population, you'll find that many of the larger cities are actually in the mountain West. Basically, you have States the size of countries like Germany, while only being allowed to use an area more the size of New Jersey or New England states.
In fact, as State populations go, the mountain West States are right around the median for the country, and several very large metros (read: well over 1M people) are located there, e.g. Phoenix or Las Vegas.
The difference doesn't have to do with the amount of people in the space -- Westerners are actually packed into cities more than most people east of the Rockies -- but a fundamental difference in attitude. As I mentioned in the original post, the Federal government has had a long history of antagonizing and exploiting the Western States, which has made those States culturally and institutionally anti-government. Combined with certain historic economic realities of the mountain West, a very libertarian outlook evolved and became a part of the culture. The government used to get away with it because at one time those States really did have tiny populations. Now, those States have millions of people in them (though this fact hasn't caught up with the politicians back east), and the culture has a long memory.
West of the Rockies, and in the mountain West in particular, the core political ideology of the region tends to revolve around a small-government, non-interference, live-and-let-live perspective -- real believers in rugged individualism. There are many historical reasons as to why this is that go back a century or two. While the people that live there are often conservative as individuals, they generally are not socially conservative in that they try and legislate the behaviors of society. A built-in distrust of government is stronger than their desire to control what other people do. East of the Rockies, big government social conservatism is deeply embedded in the culture.
Libertarians and similar have long held relatively strong positions in the mountain West due to the fact that Eastern conservatives often control conservative politics, primarily because of population differences. People like Bush reflect only the conservative issues that are unique to Eastern conservatives while not reflecting the issues shared by Eastern and Western conservatives. When more extreme examples of this come down the road in the Republican party, it tends to lead to defections to the Libertarian party out West. It is an old political and ideological tug-of-war.
In fact, if you look at the core philosophical components of Western conservatism, it is essentially libertarian. Which is why there are far fewer restrictions on what you can do and what you can own in the "conservative" mountain West than in "liberal" states, ironically. Nevada makes California look like a socially conservative police state by comparison if you actually compare laws, and they are next door.
The problem, ultimately, is that showing the page you are looking for, plus or minus two pages, is often all the pages you need to see for a great many bookes e.g. books that are randomly accessed in a reference fashion. As an example of this, my girlfriend routinely searches cookbooks online using this very feature. It shows her the recipe she was looking for from an expensive cookbook, and plus or minus a couple pages, which means she gets the entire recipes -- the primary benefit of the book -- online for free. And she uses this as an example of why her publishing houses won't participate.
For STM publishers and similar, 90% of their product line could be used this way. Letting Amazon (or Google) give away book content in a searchable format five pages at a time would dramatically eat into their sales without generating any revenue. Most of the books you do see in this system are either 1) books from minor publishers too stupid to have thought this through, or 2) a very short list of throwaway books from major publishers to prove to Amazon and themselves that it actually eats sales rather than driving them -- the consensus of the publishing industry. It would have died a long time ago except that it is the pet project of someone high up in Amazon.
It seems like you are merely arguing against frivolous patents, a completely separate issue. Most mechanical patents are crap, just like most software patents. But because this is slashdot, we focus on the software ones. I have seen no real valid argument why high-investment algorithm R&D should not be patent protected while a widget with similar investment in time and money should be. We already know there is no difference between "machine" and "software" theoretically, so that can't be the issue.
If algorithmic information theory has taught us anything, it is that everything is fundamentally a mathematical process. In this fashion, software is no different than anything else.
Again, we must either decide on a universal and consistent patent system, or none at all. Arbitrary and capricious decisions as to what can and can't be patented are certainly not the way to go. Personally, I'd be happy if we just cracked down on frivolous patents, a problem that plagues patents entirely unrelated to software as well.
There is a lot more ambiguity in the data we have than is commonly presented -- that's why both sides can use it. You soundbited your way to an argument without understanding the context. I'm not disagreeing with global warming, I'm disagreeing with the shoddy construction and specious reasoning of your position. All heat, no light. If you end up at a good result from poor premises, it is only because you were lucky.
Unfortunately, many of the most vocal people on both sides have about as deep an understanding of the issue as you do. I object to letting a mob of fools be heavy-handed with people using a specious argument. It is no different than fundamentalists forcing everyone to learn Christian Creationism in schools. If people don't want to be bothered to understand something from the first principles up, I don't want them making policy. The ability to assert an axiom does not make an axiom valid or relevant to the issue at hand.
I see that you are trying to make a correlation here, so answer me this: How do you explain CO2 concentrations that rapidly increase with a similar magnitude, pre-industrial revolution, pre-history, and even pre-homo-sapien? And why is CO2 concentration a lagging indicator of warming in previous cycles?
The climate is changing, but people trying to propose solutions based on cherry picking and pulling data out of context to suit some particular ideological viewpoint is a lot of foolishness that will create more problems than it will solve. It would make things seem so easy and ideologically pure if anthropogenic CO2 was the fulcrum of climate change, but it isn't that way in fact. There are a hell of a lot of other processes that are major contributing factors that will have to be acknowledged if a real solution is to be found, and a great many of those have nothing to do with CO2. It could very well be the case that all manmade CO2 sources could be eliminated tomorrow and it still wouldn't make a substantial difference in the climate trajectory.
The climate isn't a sitcom. You can't get a neat and tidy solution to anything in a half hour, and as is true of many types of systems, we may in fact be impotent when it comes to directing outcomes in a meaningful sense. Changing our CO2 emissions is far from free, and it would be prudent to study the systems more thoroughly before putting ourselves on a path that could find us in an even worse position than if we did nothing at all. Jumping on a ridiculously simplistic bandwagon as a solution in what is largely an absence of good knowledge is a sure way to generate a hell of a lot of heat with precious little light.
A lot of people are eager to jump on a bandwagon completely oblivious to the long-term consequences and unintended consequences that might occur -- we could end up screwing ourselves far worse than climate change. Ecosystems, fundamental economics, and a great many other things are profoundly effected by CO2 levels. You can't change global temperature in isolation by playing with CO2 levels, and the result of doing so naively may end up causing more harm than it prevented. And because of the changes made and their consequences, we may find ourselves in a situation where we are far less capable of dealing with the new mess we made.
More study, less slogan.
Several parts, but have spent most of my time in the regions around Beijing.
Ugh. Have you actually been to China? It is still clearly a "third world" type country, and will be for the foreseeable future. What makes you think this model scales?
If the standards of living are rising, it is only because they were so poor to begin with. Efficiency isn't such an important factor because you can buy big improvements for little money. But if you try to scale this to the standard of living of, say, Europe, you end up with a non-competitive economy because standards of living are pretty much defined by relative efficiency.
Centralized political economies do not scale well, and are not competitive at the high-end.
The so-called "Assault Weapon Ban" was nothing more than a feel-good measure that had nothing to do with crime or safety. All it did manage to do was annoy and/or piss-off people who buy or own guns. No appreciable benefit to any constituency, and a big downside to a rather sizeable constituency. It is no wonder that most politicians don't want to touch this issue, and Bush knew full well that it would never end up on his desk. If you ignore the Democrats in "safe States" like California, who can soapbox on this issue all day without consequences, it is a "third-rail" issue everywhere else whether you are a Democrat or a Republican.
The 1994 Congressional blood bath pretty much assures that gun control won't be touched again for a long time.
The President is NOT the representative of The People, the President is the Executive of the States -- idiots who vote in the Presidential elections who don't understand this shouldn't be voting. Up until the early 20th century, people didn't even vote for the President, the President was selected by the State governments, and was accomplished via the Electoral College. Short of discarding the Constitution and starting over, this will remain the case, as discarding this would break some of the basic axioms of the US Constitution. The President represents the States, not the People.
Hence why they are called "States" and not "Provinces". A "state" in the US is very much akin to a "country" in the EU, and is a sovereign entity. This is why laws concerning a great many things vary so widely across states. Hell, that's why even the system of Law varies across the States. Beside the default English Common, you officially have Spanish Common, and Code Napoleon (and probably others) which are recognized in some States.
In fact, I've generally held that the complexity of implementation is generally an exponential function of the general complexity of the problem. Allowing additional degrees of freedom in a design is typically very expensive. You aren't just adding that additional degree of freedom to the design, you have to make the rest of the design aware of that new dimension and put it into implementation.
Not really true. Population density of a state has little to do with how urbanized a state is. A western "county" is as large as an east coast "state", and unlike the east coast, many western counties are largely owned by the federal government (read: cannot be populated). So the county will look sparse in terms of persons/square mile, but in reality everyone lives in one or two cities and towns -- totally urban. For example, the mountain West is one of the most urban (>80% in many of these states) and demographically youngest regions of the country and also quite conservative. The states only have average populations, but most of the people live in a few large cities (in no small part because the Federal government continues to take all the land in those States).
So I don't think it is a rural versus urban issue. States like Arizona and Nevada are very young and urban, moreso than most of the traditional "liberal" States, but are also notoriously conservative in many ways, though in the West it is a very libertarian kind of conservatism.
I have only limited sympathy for people who don't save any money and the expect everyone to break out the violins when it bites them in the ass. It isn't the the fault of the government, Bush, or anyone else that you are fiscally irresponsible. I'm not exactly the definition of fiscally responsible, but I managed to scrape by for two years while unemployed and another year and a half working at a startup where I make so little I might as well be unemployed.
So quit yer whinin'. I don't want to hear it. You pissed all your money away on frivolous shit for years on end (and yes, without knowing anything about you I'll bet you spent at least half your money on frivolous shit -- I used to be the same way) and now its time to pay the piper. What a pathetic lot.
I don't know what issues you are having, but the Opterons are friggin' golden; we run them on the Arima/Rioworks MBs and have had zero problems. A 1.6GHz part runs just over $200 and schools just about everything else for heavy duty work. It feels like high priced iron but comes at a fraction of the price. Everything mission critical runs on these puppies.
You may doubt AMD, but I don't and we use them for mission critical roles. The Opterons are ridiculously good for the price, a real professional product.
Does it really matter? Most people with a 4-year CS degree have at best mediocre knowledge and abilities in these things as well out here in the real world. Most 4-year CS folks may have had to take a bunch of classes on more esoteric aspects of CS, but most don't leave school with useful knowledge beyond that of any other competent code monkey -- they largely forget all the esoteric stuff after 6 months in the real world.
The problem with CS is that you don't actually use 90% of the esoteric stuff ever once you leave school and get a real job, making them indistinguishable from the 2-year folks (or 0-year folks) in relatively short order. Above a certain baseline skill level, motivation and natural ability are pretty much everything. I work with CS PhDs who actually do real work, and even these folks are pretty clueless about theoretical computer science. They may have dabbled in it at one time, but working in the real world rotted their brains.
So you spend 3-5 years on the ultimate piece of software with a whole collection of spectacular algorithms never seen before in computer science, and put it on the web at your little website that no one has heard of. Among the trickle of people that buy your product is BigCorp, who make their own implementation of your secret algorithm sauce and sell it through their billion dollar marketing machine to every PHB on the planet. You barely cover the cost of your website, and BigCorp makes hundreds of millions on your research and development effort. You disappear into obscurity, not making a dime.
In short, without a patent the algorithm inventor would be at a HUGE disadvantage in the market in virtually every single aspect of doing business. Copyright doesn't help with this, only a fool thinks "bringing something to market" is practical in the complete absence of design protection and very limited capital, particularly if the design is genuinely worth something in the open market.
So its either trade secrets or patents. Both have their upside and downside. The "copyright is protection enough" argument is a sham. Its virtually no protection at all.
1.) The argument that software patents are bad because most software patents are frivolous is a strawman. Most non-software patents are also frivolous. This is not unique to software, people here only notice software patents because they work in software. If you don't like the frivolousness of patents granted then deal with the general frivolousness of the patent system, because "software" patents and "frivolous" patents only intersect and neither contains the other in its entirety.
2.) Software is hardware is software. If you can't patent software, then you shouldn't be able patent hardware either. This was settled the better part of a century ago, in case anyone was paying attention. Pretending that software is a special case that is different from hardware creates a distinction where none exists.
3.) A minority of software patents, just like a minority of normal hardware patents, cover inventions that took substantial research and development effort that no one could reasonable claim to be "obvious". This would seem to be precisely what is supposed to be protected by patents under ideal circumstances and I can't see a reasonable argument that says these inventions should be treated differently than all other patents. Otherwise we would be in the position of allowing frivolous non-software patents and disallowing heavy-duty substantive "software" patents.
That said, it seems to me that the biggest reason a lot of geeks don't like software patents is that it is inconvenient for them. Every rationalization that disallows "software" patents but allows other types of patents has been uniformly weak and inconsistent. If you think patents are bad, then ALL patents are bad, not just the ones you wish you didn't have to abide by.
Not just no, but hell no. One of THE major points of having shared memory is that such systems have an order of magnitude (or two) lower memory latency. Seeing as how memory latency defines the performance bound for most apps that just happen to run well on shared memory systems, there is no way you can reasonably use gigabit network fabrics as a substitute. Unless you can afford greatly reduced performance to save a few bucks.
When gigabit networks have typical latencies like that of a ccNUMA system, this will be worth considering. But we are still pretty far from that today, and it isn't a trivial engineering problem to get there.
The most common uses for this: long-haul video transfer, massive network file systems, and extremely high-volume internet servers. Massive distributed private file systems and applications are hot, and will be a major driver of high-performance fiber ethernet fabrics.
For most purposes, a properly architected network will add almost no real latency beyond what is required due to physical limits i.e. the speed of light. Regional ethernet fabrics should be easily sub-millisecond between any two points in the region, which is quite adequate for most applications that might be amenable to being networked to begin with. Most people still think of networks in terms of routing bottlenecks and other topology biasing issues like that, but large-scale ethernet fabrics, an increasingly popular choice for network providers big and small, make most of these issues go away for a lot of purposes and mitigate the performance hit for most others. 10G will just make ubiquitous ethernet fabrics that much more scalable.