The US Government just changed daylight slavings time again
That's right, they changed it again. It has changed many times over the past century, and sometimes the changes depend on your exact locality. Therefore, it would be rather foolish to buy something that would need to be replaced if the rules change again. (And they will change again; I predict that the fall date will move back as soon as the first kid gets run over walking to school in the dark.)
As for patching, my computer was automatically updated with new timezone info about a week after this was finalized, and well over a year before anything is going to happen. Once again, it would be foolish to buy anything that attempts to automatically track DST that isn't trivial to patch.
The Qt Commercial License is the correct license to use for the construction of proprietary, commercial software. The license allows you to:
Build commercial software and software whose source code you wish to keep private.
...
It doesn't say here that the GPL doesn't allow you to do these things, only that the QT commercial license does allow them. What the GPL allows and doesn't allow is in the GPL. The GPL is one of the licenses included in the software; that is the place to look for what you can and can't do with the software, not some website summary.
It's true that the website's tricky wording is probably bordering on FUD in an effort to encourage license sales. However, it doesn't actually say that you can't develop commercial open-source software without the pricey QT license.
The earth receives over 14,000 times as much solar energy as the entire human race currently generates. Any conceivable amount of power that we would generate, even if we increased power usage by 50X, would be a tiny rounding error compared to natural solar heating.
Doubling the natural amount of CO2, which is what we're working towards now, is far more significant than increasing our heat production. It's just the same as the situation where significantly increasing the amount of insulation in your house would raise the temperature much more than just burning 1% more fuel in your furnace would.
That, my Gawd, with cheap clean energy we would just end up with more people using up even more of the planet even faster.
I don't know why he would think that. A sufficient amount of cheap clean energy would give us the ability to economically reprocess any and all waste back into raw materials. In theory, with enough energy we would hardly need to use any fresh resources at all, and we would no longer need landfills.
All that is cancelled out by the fact that the x86 instructions are much more compact than RISC instructions, so the instruction cache is far more effective. It comes out as a wash.
The translation logic is a tiny fraction of modern CPU real estate. Recent RISC CPUs like the Power architecture have a lot of tricky scheduling logic of their own. They're all trying to find instruction-level parallelism, this requires a lot of instruction munging. Nobody is going to get by without pipelining. (The Itanium tried, but it didn't turn out so well.)
However, the web doesn't need any central control at all to work, so every web server is it's own independent implementation. The web is totally independent of its creators. DNS doesn't work like that.
I don't disagree with that. It just happens that x86 currently beats others speedwise because it's gotten the most investment of engineering resources.
There might be faster x86 processors available, but there's also fast IA64 processors, PPC processors, x86_64 processors, etc.
There might be a few exotic processors that eke out an edge in certain benchmarks over the top x86 and x86/64 CPUs, but that's solely due to huge, ultra-expensive caches or crazy numbers of cores on chip. X86 chips with similar amounts of cache and core count would perform similarly.
The fastest chips that anybody would consider putting in a personal computer today happen to be x86. Like you said, the instruction set doesn't have anything to do with the performance. Complaining about having to use x86s like the original GP post is pointless.
At best, x86 is slower than RISC based processors since, without fancy scheduling, and other nifty tricks, RISC processors do more per cycle than their x86 counterparts.
x86 processors have been RISC under the hood for almost a decade. All modern "RISC" chips also use fancy scheduling and other nifty tricks just about as much as x86 CPUs do. (The only exception is Itanium, which tries to shift the nifty tricks to the compiler, with less-than-stellar results.)
Why? Unless you write your code in assembler (or you have some kind of irrational preference for a particular endianness), you'll never tell the difference between instruction set architectures. The only user-observable or programmer-observable difference between CPUs is speed, and x86 is faster.
Hey, airlines spend billions of dollars per year on their infrastructure, too.
I think that they need to get on this bandwagon. Their executives should be saying: "Thousands of businessmen are flying on MY airplanes every day to go make deals for SERIOUSLY big money. Well, I want a cut of that! It's not fair that someone gets to use MY airplanes to get to a multimillion-dollar deal, and I only get $599! This has got to stop!!"
I'm tired of hearing about all these advances that we will NEVER see.
What do you mean? When I was a kid, fiber cable was just a novelty you read about in Popular Science. They claimed that it had the potential capacity to transmit things like War and Peace in just a few seconds.
Well, guess what: Today, in the comfort of your own home, you can downloadWar and Peace in just a few seconds.
This should have been a non-issue. If copyright laws were still in compliance with the US constitution, these old shows would have entered the public domain years ago.
just look at their name: PriceWaterhouseCoopersWhatALongFuckingNameTooCheap ToBuyAFuckingHyphen
Actually, I think It's Can'tAffordToBuyAHyphenAfterPayingTooMuchForTypese tting.
It isn't possible to do justice to the CamelCaseOnAcid layout of their actual logo here in plain HTML.
You are absolutely right of course, in my parent this is exactly what I meant. MS-Office runs on BSDish MacOSX, if demand were sufficiently high it would be available for Linux.
Apple is Microsoft's designated antitrust deflector. Demand for office on other platforms has little or nothing to do with it. They put up with this one token 3rd-party platform as a necessary evil to ward off accusations of monopoly status.
For any other case, their shrewd strategy is to keep people herded inside their OS ecosystem. That means they will write few or no user apps for other platforms, even if the ported product would be profitible in and of itself.
That was a good thing. They had ballistic mode as a backup option if the guidance system failed. That's a robust design.
What would happen if the shuttle's guidance system failed? It would end up ballistic as well, but not before breaking up into thousands of white-hot pieces.
But the shuttle doesn't use expendable engines, so it can only lift 24 tons in the real world. And since they've decided that routinely lifting heavy loads in the shuttle is too dangerous for the crew, much of what capacity they do have goes wasted.
If it's okay for me to download music from russian sites at a nickel a track or watch movies from bit torrent trackers (and I do both of these readily and happily), who am I (and most of us) to criticise a little GPL violation here and there?
If you, an insignificant individual, do those things, at least you're behaving consistently (if not necessarilly legally).
If a commercial software vendor does those things, while at the same time demanding that people pay for *their* goods and abide by *their* licenses, then they are being big hypocrites.
The point is they've borrowed from Unix - but rather than pipe text that needs to be formatted and parsed between commands they're passing.NET objects which don't.
So now you need to be sure that the target program calls the appropriate methods on the supplied objects with correct parameters. In most real-world cases, that's going to be no easier than parsing the data.
Generally, data is just data, but objects are data+behavior. Instead of just worrying about the behavior of the application, you'll be trying to mix the behavior of the target application with the behavior of the data objects and making sure that they're compatible. This is a whole new layer of complexity that will likely come with obscure interactions and unexpected pitfalls.
Most every remote object technology in current use sounds great on paper, but they're almost invariably a PITA to use, not to mention slow and flaky. I'd be surprised if this doesn't end up in the same boat.
The last cosmonaut killed was > 30 years ago. All of their deaths were due to isolated failures in small components (parachute in one case, air pressurization valve in another). These have long ago been fixed. The shuttle incidents occurred because of fundamental design flaws that can't be corrected, but only partially mitigated at huge additional cost.
In the 1980s, a Soyuz booster did explode (just like the Challenger), but since they didn't commit the fundamental design flaw of omitting an escape system, the cosmonauts walked away from the incident.
Their launch cost = 1/20th of shuttle launch cost.
Which country's taxpayers are getting a better deal for their money?
Slashdot Mirror
That's right, they changed it again. It has changed many times over the past century, and sometimes the changes depend on your exact locality. Therefore, it would be rather foolish to buy something that would need to be replaced if the rules change again. (And they will change again; I predict that the fall date will move back as soon as the first kid gets run over walking to school in the dark.)
As for patching, my computer was automatically updated with new timezone info about a week after this was finalized, and well over a year before anything is going to happen. Once again, it would be foolish to buy anything that attempts to automatically track DST that isn't trivial to patch.
It doesn't say here that the GPL doesn't allow you to do these things, only that the QT commercial license does allow them. What the GPL allows and doesn't allow is in the GPL. The GPL is one of the licenses included in the software; that is the place to look for what you can and can't do with the software, not some website summary.
It's true that the website's tricky wording is probably bordering on FUD in an effort to encourage license sales. However, it doesn't actually say that you can't develop commercial open-source software without the pricey QT license.
Probably because the moon rocket can only hold 3 people at a time.
Doubling the natural amount of CO2, which is what we're working towards now, is far more significant than increasing our heat production. It's just the same as the situation where significantly increasing the amount of insulation in your house would raise the temperature much more than just burning 1% more fuel in your furnace would.
I don't know why he would think that. A sufficient amount of cheap clean energy would give us the ability to economically reprocess any and all waste back into raw materials. In theory, with enough energy we would hardly need to use any fresh resources at all, and we would no longer need landfills.
The translation logic is a tiny fraction of modern CPU real estate. Recent RISC CPUs like the Power architecture have a lot of tricky scheduling logic of their own. They're all trying to find instruction-level parallelism, this requires a lot of instruction munging. Nobody is going to get by without pipelining. (The Itanium tried, but it didn't turn out so well.)
However, the web doesn't need any central control at all to work, so every web server is it's own independent implementation. The web is totally independent of its creators. DNS doesn't work like that.
I don't disagree with that. It just happens that x86 currently beats others speedwise because it's gotten the most investment of engineering resources.
There might be faster x86 processors available, but there's also fast IA64 processors, PPC processors, x86_64 processors, etc.
There might be a few exotic processors that eke out an edge in certain benchmarks over the top x86 and x86/64 CPUs, but that's solely due to huge, ultra-expensive caches or crazy numbers of cores on chip. X86 chips with similar amounts of cache and core count would perform similarly.
The fastest chips that anybody would consider putting in a personal computer today happen to be x86. Like you said, the instruction set doesn't have anything to do with the performance. Complaining about having to use x86s like the original GP post is pointless.
At best, x86 is slower than RISC based processors since, without fancy scheduling, and other nifty tricks, RISC processors do more per cycle than their x86 counterparts.
x86 processors have been RISC under the hood for almost a decade. All modern "RISC" chips also use fancy scheduling and other nifty tricks just about as much as x86 CPUs do. (The only exception is Itanium, which tries to shift the nifty tricks to the compiler, with less-than-stellar results.)
Why? Unless you write your code in assembler (or you have some kind of irrational preference for a particular endianness), you'll never tell the difference between instruction set architectures. The only user-observable or programmer-observable difference between CPUs is speed, and x86 is faster.
That comes out around a cool $1 Billion per visitor. And so much has been accomplished. Such a deal.
The answer is right there in your local header file. From <vector>:
The Schrodinger Corp. makes special PC cases that can handle those requirements.
I think that they need to get on this bandwagon. Their executives should be saying: "Thousands of businessmen are flying on MY airplanes every day to go make deals for SERIOUSLY big money. Well, I want a cut of that! It's not fair that someone gets to use MY airplanes to get to a multimillion-dollar deal, and I only get $599! This has got to stop!!"
What do you mean? When I was a kid, fiber cable was just a novelty you read about in Popular Science. They claimed that it had the potential capacity to transmit things like War and Peace in just a few seconds.
Well, guess what: Today, in the comfort of your own home, you can download War and Peace in just a few seconds.
8 gigs of dual-channel PC60000 RAM ought to be able to handle the job.
This should have been a non-issue. If copyright laws were still in compliance with the US constitution, these old shows would have entered the public domain years ago.
The only problem with your plan: No such set exists.
Actually, I think It's Can'tAffordToBuyAHyphenAfterPayingTooMuchForTypese tting.
It isn't possible to do justice to the CamelCaseOnAcid layout of their actual logo here in plain HTML.
Apple is Microsoft's designated antitrust deflector. Demand for office on other platforms has little or nothing to do with it. They put up with this one token 3rd-party platform as a necessary evil to ward off accusations of monopoly status.
For any other case, their shrewd strategy is to keep people herded inside their OS ecosystem. That means they will write few or no user apps for other platforms, even if the ported product would be profitible in and of itself.
What would happen if the shuttle's guidance system failed? It would end up ballistic as well, but not before breaking up into thousands of white-hot pieces.
But the shuttle doesn't use expendable engines, so it can only lift 24 tons in the real world. And since they've decided that routinely lifting heavy loads in the shuttle is too dangerous for the crew, much of what capacity they do have goes wasted.
If you, an insignificant individual, do those things, at least you're behaving consistently (if not necessarilly legally).
If a commercial software vendor does those things, while at the same time demanding that people pay for *their* goods and abide by *their* licenses, then they are being big hypocrites.
So now you need to be sure that the target program calls the appropriate methods on the supplied objects with correct parameters. In most real-world cases, that's going to be no easier than parsing the data.
Generally, data is just data, but objects are data+behavior. Instead of just worrying about the behavior of the application, you'll be trying to mix the behavior of the target application with the behavior of the data objects and making sure that they're compatible. This is a whole new layer of complexity that will likely come with obscure interactions and unexpected pitfalls.
Most every remote object technology in current use sounds great on paper, but they're almost invariably a PITA to use, not to mention slow and flaky. I'd be surprised if this doesn't end up in the same boat.
In the 1980s, a Soyuz booster did explode (just like the Challenger), but since they didn't commit the fundamental design flaw of omitting an escape system, the cosmonauts walked away from the incident.
Their launch cost = 1/20th of shuttle launch cost.
Which country's taxpayers are getting a better deal for their money?
They would be a valuable resource today if it weren't for centuries of gross mismanagement.