That's a very twisted view of taxation and its purpose. States/cities/counties/etc. don't complete based on tax rates. Studies have demonstrated that over and over. The competition is based on infrastructure and quality of life -- is there adequate transportation for employees and goods, is there a well-educated workforce, is there a thriving arts culture (yes, smart businesses look for this!).
And guess what? A state provides none of this without taxes.
So Amazon is acting like a typical "me first" corporation and ignoring its responsibility to the community. What else is new?
The fact is, online retailers have been leeching off communities for far too long. They make use of the infrastructure these communities provide but use tax evasion to make sure they don't contribute to its upkeep. Moreover, local businesses fairly, rightfully and morally pay their taxes and get undercut by unfair competition.
This isn't an issue of some ridiculous "internet freedom" entitlement. This is about making sure that everyone shares the responsibility for keeping our society working. Bravo to North Carolina for calling these online retailers to be responsible.
This isn't a tax in the traditional sense. This is an attempt to quantify some of the externalities in our energy consumption and reflect the true cost of what we're doing. Free marketers should be all for this because it increases the amount and accuracy of pricing information to the market. If Republicans and Blue Dogs really care about market competition, they should be overjoyed about this bill because it starts to level the playing field.
But of course this isn't about markets. This is about ideology and some people will never give up the idea that they've got a God-given right to consume everything and pollute as much as they wish and damn the consequences to anyone else.
I think GM's approach (with bringing out a less exotic Chevy Volt) or yours (an even more austere Aveo) is backwards. Say it costs $5000 to $10000 to implement the advanced battery and drivetrain at this point in development. This means the cost of an electric Aveo would be 50+ percent higher than for a gas one, which is "cheap enough" to run in the first place.
Don't confuse the Volt and an electric Aveo. The Volt has a range extender. That makes it very different than an electric Aveo or a Tesla and will probably be more attractive to the avereage person due to the range anxiety issue. The Volt is not any less advanced technologically than the Tesla (who cares about "exotic?"). It's just different. An arguably a better investment.
GM in particular has been the ABSOLUTE LEAST INNOVATIVE auto company on the entire planet for decades--even its best products are dependable but very boring and un-innovative, and they've invested the least into new technologies in their plants out of EVERY SINGLE company that builds cars in N America.
Depends on what you look at. For production vehicles, that may be true. But GM did do the EV1 and has a done a ton of research into electric, hybrid and range-extended vehicles. The Tesla's downfall is the huge battery pack. It seems a poor engineering tradeoff to spend $$$ to get 200 miles on a change when 40 miles will do just fine for the average commuter (i.e. 80% or more of us) and a simple ICE running at its most efficient RPM can supply extra range when necessary. This is borne out in the price difference between the Model S and the Volt.
San Francisco is nice, but it's not exactly someplace you live in you're straight and have a family.
Excuse me? Have you actually been to San Francisco?
Most cities are NOT like NYC, LA, SF, etc. They're usually more like Atlanta. The suburbs are fairly safe, and the downtown is a big ghetto full of crime and drugs.
You don't have a clue what you're talking about. Most cities have neighborhoods with various income levels. Some of the poorer areas may also have higher crime but even that is not always true. I feel quite comfortable walking around North Minneapolis, even though people in the suburban Twin Cities area are too afraid to ever go there. Too bad, they're missing some nice restaurants and coffee shops.
There's a reason white people (or rather, middle-class people) have been moving out of cities: they want to get away from all the crime
Wrong again. People initially left because after WWII the government subsidized new housing construction in the suburbs and penalized infill development. Segregated neighborhoods were official policy and people were led to believe their cities were crime-infested when in fact the numbers demonstrate the opposite. It was only after the whites (and the wealth and tax base) left the cities that crime began to become a real problem. And today, those impressions persist even though, for example, crime has droppepd nearly 70% in Minneapolis over the last decade or so.
U.S. residents have a distorted and paranoid view on crime. It's incredibly unlikely that you or I will be killed or harmed by some random act of violence. Almost all violent crime is targeted toward specific individuals by people they know.
People don't usually move to cities because they really want to live around so many other people; they go there because there's employment, and they're hungry. Niceties like walkability are luxuries in selecting a city to move to.
People go to the cities because of the community (yes, most people actually like being around other people) and the amenities. We're seeing a trend of people moving back into our cities. Downtowns are becoming more residential. It's certainly happening here in the Twin Cities. In fact the company I work for is about to move from the suburbs to downtown because it's already lost recruits due to its isolated location. The younger generation wants to live and work in the city.
It was quite clear from my original wording that I was talking about Seymour Cray.
No, it wasn't. But let's not continue a useless argument. Obviously I misunderstood what you were saying. But I'll note that Seymour didn't invent vector computing, so the statement is a little misleading. To his credit, Cray was the first ot use vector registers, which was an important innovation.
If I were talking about a company, what sense would it make to exclude the kind of computers the company made after it was defunct? Name one company that not only manufactures but designs new computers after it has gone out of business.
Cray the company did not go out of business. It got acquired a spun off and acquired again. That's not quite the same thing as going under. And as I noted earlier, it's the same people.
You clearly have a bug up your ass about Cray and your entire contribution to this thread has been to show off that bug, to the point of making an ass of yourself.
Well, that's your opinion. I simply think it's unwise to ignore the lessons history teaches.
You're making the mistake of equating a company's products with one person. It doesn't work that way. Seymour Cray did not single-handedly produce any machine. It takes a team of dedicated people to do that. Therefore, this quote is a non-sequitur if "Cray" means an individual:
A GPU is very much like the vector processor of Cray-style supercomputing (when Cray was still alive that is)
The quote only makes sense if "Cray" refers to the company. "Cray-style supercomputing" only exists because of a group of people, not one man. Otherwise you're going to be in serious trouble when the one man gets hit by a bus.
And let's be a little more accurate with our history (this is directed to all of us). Seymour Cray had many more failures than successes. He had a tendancy to go for the perfect instead of the good and that resulted in numerous over-budget, cancelled projects. We need to be much more realistic about how we treat key figures in our industry, lest we fail to learn the lessons they've provided us.
Seymour Cray was killed by a speeding redneck in a trans-am in 1996.
So? Cray != Seymour. In fact the most successful Cray machines were not designed by Seymour.
The company currently known as Cray as formerly known as TERA, which bought the assets of Cray Research from SGI who acquired Cray Research after Seymour had left to form Cray Computer which is also defunct.
So? Many of the engineers there have been there for a long time. Even if they've been bounced around between companies, it's a good number of the same people. And who's to say that SGI and Tera didn't provide some good brainpower to the current Cray, Inc.? No one has a monopoly on good design.
Seymour was never significantly involved in multi-core or multi-threaded processors or NUMA. In fact, he specifically avoided designs even hinting of that sort of complexity because he felt that simplicity in design made it easier to fully utilize the maximum performance of the hardware.
So? Seymour was wrong. It worked in the early days of CDC and Cray Research but it doesn't work any more. The microprocessor vendors made sure of that. Honestly, the man wasn't a god.
It's going to be interesting to see how Larrabee and AMD's Fusion battle it out. With Larrabee, Intel is taking a tightly integrated approach. One can easily imagine that LRBni will be integrated into mainstream CPUs in the not-so-distant future, at which point Intel will argue that no one needs a GPU.
AMD, on the other hand, is taking he approach of (relatively) loosely-coupled specialized processors. One, the CPU, for general-purpose/integer/branchy code and the GPU for graphics (and HPC?).
Currently my bet is on Intel because of the much simpler Larrabee programming model. But if the performance isn't there, things could get heated.
No, it isn't. It's a myth that the only places dense enough for public transit are the big cities. Places like Atlanta, Dallas, Charlotte, etc. are investing heavily in rail transit. These are places not exactly known for their urban density.
The real problem is lack of political will. We could build some fantastic public transit systems today if we'd stop listening to the nutjobs who think the answer is to give everyone a car and let 'em fight it out on the (hugely expanded) freeways.
Most Urban areas in the US are cesspools of crime and poverty.
This is quite possibly the most ignorant statement I've ever read on Slashdot.
Have you even been to a major urban area? There aren't bullets flying everywhere, even in the worst parts of Detroit. Please get out a little bit and walk around a big city. Yes, even in the "bad" parts.
To believe that governments are able to supply them is to think that a monopoly is better at serving customers than a competitive market.
You're putting words in my mouth. Government is not a monopoly. Government is accountable to the public because we make up the government. A private corporation is not accountable to the public. A heck of a lot more scrutiny goes into government programs than happens in the private sector. It's all that "red tape" people are always complaining about.
You're splitting hairs but it would be very easy to satisfy your definition. Make all public transit fare-free and ban toll roads. It'd actually be a very good idea.
You don't understand what a public good is? Then I'm afraid you will never understand the value of great community.
Public goods like infrastructure should and must be paid with public dollars. Because once the private sector gets involved, you can't have public accountability anymore.
This is exactly why the health care system in the U.S. has utterly failed. We have the worst health outcomes of all industrialized nations and spend the most on health care because we've decided that the health of the public is not a public good. It's been a huge mistake that's costing us trillions of dollars.
From my perspective, Amtrak is doing a great job given the crap they've had to deal with. I ride the Empire Builder semi-regularly and it's always a comfortable ride with great service. Some of the things Amtrak has to deal with include:
Constant threats of massive budget cuts
Massive budget cuts
Being at the mercy of freight rail comapnies for right-of-way (this is the primary cause of delayed trains)
Paying large fines to said freight rail companies when trains get delayed due to said freight rail companies' policies
It's a wonder Amtrak works as well as it does. We need to remind the freight rail companies that we gave them their trackage in the first place and they'd better play nice with passenger rail or we'll take it away from them.
Do you not think that roads enjoy the same subsidt transit does? ALL transportation is subsidized and that's a necessary thing because it's a public good.
This system won't take people from NYC to LA. It's for going from Minneapolis or Madison to Chicago. These are routes where air travel is wasteful (2 hours in the airport waiting for a one hour flight) and rail competes very well. Even with it's relatively slow speed and frequent stops, Amtrak's Empire Builder from the Twin Cities to Chicago is almost always packed. You usually can't get a ticket within a month of travel.
Yes, we need to invest in commuter rail and light rail. Many cities are doing just that. But there is most definitely a place for intercity rail in this country.
Clock speeds are probably not going to get significantly higher unless we have a radical new fabrication technology. I'm not ruling that out because it's happened before, but we're really getting to some physical limits with circuitry.
Given that, there are really two ways to speed up a single thread of execution and they go hand-in-hand. First you do some architecting to provide a better platform to program on. You do that with microarchitecture and ISA. Larrabee delivers the ISA part. IF compilers can use it well, it could provide significant speedup on some codes. Which brings us to the second way to speed up single thread performance: use the hardware more efficiently. Compilers and other tools are going to be key to future performance and programmers are going to have to get smarter about what they're doing.
I think it's a rather exciting time because the game is changing.
OpenCL is a big ugly hack meant to provide a standard API for legacy GPUs. It's totally inappropriate for something like Larrabee, which is much more general purpose. A good vectorizing compiler will be able to make use of most of Larrabee's features directly. You'll be able to write code in standard languages, with an eye toward writing in a way that makes vectorization possible. While this does require that programmers get trained to understand things like loop dependence, it doesn't require learning a whole new language and API.
Slashdot Mirror
Taxes are how states compete for business.
That's a very twisted view of taxation and its purpose. States/cities/counties/etc. don't complete based on tax rates. Studies have demonstrated that over and over. The competition is based on infrastructure and quality of life -- is there adequate transportation for employees and goods, is there a well-educated workforce, is there a thriving arts culture (yes, smart businesses look for this!).
And guess what? A state provides none of this without taxes.
It's not. They should pay their taxes as well.
So Amazon is acting like a typical "me first" corporation and ignoring its responsibility to the community. What else is new?
The fact is, online retailers have been leeching off communities for far too long. They make use of the infrastructure these communities provide but use tax evasion to make sure they don't contribute to its upkeep. Moreover, local businesses fairly, rightfully and morally pay their taxes and get undercut by unfair competition.
This isn't an issue of some ridiculous "internet freedom" entitlement. This is about making sure that everyone shares the responsibility for keeping our society working. Bravo to North Carolina for calling these online retailers to be responsible.
This isn't a tax in the traditional sense. This is an attempt to quantify some of the externalities in our energy consumption and reflect the true cost of what we're doing. Free marketers should be all for this because it increases the amount and accuracy of pricing information to the market. If Republicans and Blue Dogs really care about market competition, they should be overjoyed about this bill because it starts to level the playing field.
But of course this isn't about markets. This is about ideology and some people will never give up the idea that they've got a God-given right to consume everything and pollute as much as they wish and damn the consequences to anyone else.
I think GM's approach (with bringing out a less exotic Chevy Volt) or yours (an even more austere Aveo) is backwards. Say it costs $5000 to $10000 to implement the advanced battery and drivetrain at this point in development. This means the cost of an electric Aveo would be 50+ percent higher than for a gas one, which is "cheap enough" to run in the first place.
Don't confuse the Volt and an electric Aveo. The Volt has a range extender. That makes it very different than an electric Aveo or a Tesla and will probably be more attractive to the avereage person due to the range anxiety issue. The Volt is not any less advanced technologically than the Tesla (who cares about "exotic?"). It's just different. An arguably a better investment.
GM in particular has been the ABSOLUTE LEAST INNOVATIVE auto company on the entire planet for decades--even its best products are dependable but very boring and un-innovative, and they've invested the least into new technologies in their plants out of EVERY SINGLE company that builds cars in N America.
Depends on what you look at. For production vehicles, that may be true. But GM did do the EV1 and has a done a ton of research into electric, hybrid and range-extended vehicles. The Tesla's downfall is the huge battery pack. It seems a poor engineering tradeoff to spend $$$ to get 200 miles on a change when 40 miles will do just fine for the average commuter (i.e. 80% or more of us) and a simple ICE running at its most efficient RPM can supply extra range when necessary. This is borne out in the price difference between the Model S and the Volt.
San Francisco is nice, but it's not exactly someplace you live in you're straight and have a family.
Excuse me? Have you actually been to San Francisco?
Most cities are NOT like NYC, LA, SF, etc. They're usually more like Atlanta. The suburbs are fairly safe, and the downtown is a big ghetto full of crime and drugs.
You don't have a clue what you're talking about. Most cities have neighborhoods with various income levels. Some of the poorer areas may also have higher crime but even that is not always true. I feel quite comfortable walking around North Minneapolis, even though people in the suburban Twin Cities area are too afraid to ever go there. Too bad, they're missing some nice restaurants and coffee shops.
There's a reason white people (or rather, middle-class people) have been moving out of cities: they want to get away from all the crime
Wrong again. People initially left because after WWII the government subsidized new housing construction in the suburbs and penalized infill development. Segregated neighborhoods were official policy and people were led to believe their cities were crime-infested when in fact the numbers demonstrate the opposite. It was only after the whites (and the wealth and tax base) left the cities that crime began to become a real problem. And today, those impressions persist even though, for example, crime has droppepd nearly 70% in Minneapolis over the last decade or so.
U.S. residents have a distorted and paranoid view on crime. It's incredibly unlikely that you or I will be killed or harmed by some random act of violence. Almost all violent crime is targeted toward specific individuals by people they know.
People don't usually move to cities because they really want to live around so many other people; they go there because there's employment, and they're hungry. Niceties like walkability are luxuries in selecting a city to move to.
People go to the cities because of the community (yes, most people actually like being around other people) and the amenities. We're seeing a trend of people moving back into our cities. Downtowns are becoming more residential. It's certainly happening here in the Twin Cities. In fact the company I work for is about to move from the suburbs to downtown because it's already lost recruits due to its isolated location. The younger generation wants to live and work in the city.
It was quite clear from my original wording that I was talking about Seymour Cray.
No, it wasn't. But let's not continue a useless argument. Obviously I misunderstood what you were saying. But I'll note that Seymour didn't invent vector computing, so the statement is a little misleading. To his credit, Cray was the first ot use vector registers, which was an important innovation.
If I were talking about a company, what sense would it make to exclude the kind of computers the company made after it was defunct? Name one company that not only manufactures but designs new computers after it has gone out of business.
Cray the company did not go out of business. It got acquired a spun off and acquired again. That's not quite the same thing as going under. And as I noted earlier, it's the same people.
You clearly have a bug up your ass about Cray and your entire contribution to this thread has been to show off that bug, to the point of making an ass of yourself.
Well, that's your opinion. I simply think it's unwise to ignore the lessons history teaches.
You're making the mistake of equating a company's products with one person. It doesn't work that way. Seymour Cray did not single-handedly produce any machine. It takes a team of dedicated people to do that. Therefore, this quote is a non-sequitur if "Cray" means an individual:
A GPU is very much like the vector processor of Cray-style supercomputing (when Cray was still alive that is)
The quote only makes sense if "Cray" refers to the company. "Cray-style supercomputing" only exists because of a group of people, not one man. Otherwise you're going to be in serious trouble when the one man gets hit by a bus.
And let's be a little more accurate with our history (this is directed to all of us). Seymour Cray had many more failures than successes. He had a tendancy to go for the perfect instead of the good and that resulted in numerous over-budget, cancelled projects. We need to be much more realistic about how we treat key figures in our industry, lest we fail to learn the lessons they've provided us.
Seymour Cray was killed by a speeding redneck in a trans-am in 1996.
So? Cray != Seymour. In fact the most successful Cray machines were not designed by Seymour.
The company currently known as Cray as formerly known as TERA, which bought the assets of Cray Research from SGI who acquired Cray Research after Seymour had left to form Cray Computer which is also defunct.
So? Many of the engineers there have been there for a long time. Even if they've been bounced around between companies, it's a good number of the same people. And who's to say that SGI and Tera didn't provide some good brainpower to the current Cray, Inc.? No one has a monopoly on good design.
Seymour was never significantly involved in multi-core or multi-threaded processors or NUMA. In fact, he specifically avoided designs even hinting of that sort of complexity because he felt that simplicity in design made it easier to fully utilize the maximum performance of the hardware.
So? Seymour was wrong. It worked in the early days of CDC and Cray Research but it doesn't work any more. The microprocessor vendors made sure of that. Honestly, the man wasn't a god.
It's going to be interesting to see how Larrabee and AMD's Fusion battle it out. With Larrabee, Intel is taking a tightly integrated approach. One can easily imagine that LRBni will be integrated into mainstream CPUs in the not-so-distant future, at which point Intel will argue that no one needs a GPU.
AMD, on the other hand, is taking he approach of (relatively) loosely-coupled specialized processors. One, the CPU, for general-purpose/integer/branchy code and the GPU for graphics (and HPC?).
Currently my bet is on Intel because of the much simpler Larrabee programming model. But if the performance isn't there, things could get heated.
Uhh...Cray is still very much alive. And doing vectors. And threads. And multicore. All long before Intel/AMD.
The real problem is population density.
No, it isn't. It's a myth that the only places dense enough for public transit are the big cities. Places like Atlanta, Dallas, Charlotte, etc. are investing heavily in rail transit. These are places not exactly known for their urban density.
The real problem is lack of political will. We could build some fantastic public transit systems today if we'd stop listening to the nutjobs who think the answer is to give everyone a car and let 'em fight it out on the (hugely expanded) freeways.
Most Urban areas in the US are cesspools of crime and poverty.
This is quite possibly the most ignorant statement I've ever read on Slashdot.
Have you even been to a major urban area? There aren't bullets flying everywhere, even in the worst parts of Detroit. Please get out a little bit and walk around a big city. Yes, even in the "bad" parts.
To believe that governments are able to supply them is to think that a monopoly is better at serving customers than a competitive market.
You're putting words in my mouth. Government is not a monopoly. Government is accountable to the public because we make up the government. A private corporation is not accountable to the public. A heck of a lot more scrutiny goes into government programs than happens in the private sector. It's all that "red tape" people are always complaining about.
Cheaper for who? It would very much NOT be cheaper to give someone a car and then expect them to pay for gas, maintenance, etc.
You're splitting hairs but it would be very easy to satisfy your definition. Make all public transit fare-free and ban toll roads. It'd actually be a very good idea.
You don't understand what a public good is? Then I'm afraid you will never understand the value of great community.
Public goods like infrastructure should and must be paid with public dollars. Because once the private sector gets involved, you can't have public accountability anymore.
This is exactly why the health care system in the U.S. has utterly failed. We have the worst health outcomes of all industrialized nations and spend the most on health care because we've decided that the health of the public is not a public good. It's been a huge mistake that's costing us trillions of dollars.
There's more than out-of-pocket costs. Externalities are at play here. Read the report for a full explanation.
Where is there mass transit that works in the U. S.?
Hmm...
Places like the Twin Cities are still behind but are starting to catch up.
From my perspective, Amtrak is doing a great job given the crap they've had to deal with. I ride the Empire Builder semi-regularly and it's always a comfortable ride with great service. Some of the things Amtrak has to deal with include:
It's a wonder Amtrak works as well as it does. We need to remind the freight rail companies that we gave them their trackage in the first place and they'd better play nice with passenger rail or we'll take it away from them.
Do you not think that roads enjoy the same subsidt transit does? ALL transportation is subsidized and that's a necessary thing because it's a public good.
This system won't take people from NYC to LA. It's for going from Minneapolis or Madison to Chicago. These are routes where air travel is wasteful (2 hours in the airport waiting for a one hour flight) and rail competes very well. Even with it's relatively slow speed and frequent stops, Amtrak's Empire Builder from the Twin Cities to Chicago is almost always packed. You usually can't get a ticket within a month of travel.
Yes, we need to invest in commuter rail and light rail. Many cities are doing just that. But there is most definitely a place for intercity rail in this country.
Not to mention that "free" parking is not free. Read this report for an eye-opener: The Myth of Free Parking
Clock speeds are probably not going to get significantly higher unless we have a radical new fabrication technology. I'm not ruling that out because it's happened before, but we're really getting to some physical limits with circuitry.
Given that, there are really two ways to speed up a single thread of execution and they go hand-in-hand. First you do some architecting to provide a better platform to program on. You do that with microarchitecture and ISA. Larrabee delivers the ISA part. IF compilers can use it well, it could provide significant speedup on some codes. Which brings us to the second way to speed up single thread performance: use the hardware more efficiently. Compilers and other tools are going to be key to future performance and programmers are going to have to get smarter about what they're doing.
I think it's a rather exciting time because the game is changing.
OpenCL is a big ugly hack meant to provide a standard API for legacy GPUs. It's totally inappropriate for something like Larrabee, which is much more general purpose. A good vectorizing compiler will be able to make use of most of Larrabee's features directly. You'll be able to write code in standard languages, with an eye toward writing in a way that makes vectorization possible. While this does require that programmers get trained to understand things like loop dependence, it doesn't require learning a whole new language and API.