And how much of that is because the user interface is designed around having important elements at or near the edge of the screen so that they are easy to reach? And do your kiosks have the touch screen at eye level, with the screen vertical? That's the way most kiosks seem to work, and it is pretty obvious that such an arrangement doesn't demonstrate the advantages of a touch screen. To really asses the usability of a touch screen, you need to use a posture similar to that used for typing (but with more incline for the screen).
Assuming you have a good ergonomic set-up for your virtual tours, I don't see why a touch screen couldn't be easier than, or at least as easy as using a mouse. Panning could be done with as little as one finger, zooming could be accomplished with two fingers in a manner similar to the iPhone (saving you a round-trip to the edge of the screen to select a zooming tool), and changing location is the same finger tap as with the mouse. What's more is that you are much less at risk of RSI.
I can understand the webmail being hard to use with a touch screen, as those things are seldom easy to use under any conditions.
Wouldn't it be easier if they simply sold service, supported open standards and reduced operating costs by not stocking a giant cache of crappy phones they cover under replacement? No. That comes too close to the competitive landscape that Google and others are trying to create. If cellular providers end up as commodities, they won't be able to earn the huge* profits they currently enjoy, and they'll have to spend more money upgrading their infrastructure. It's the same situation as with ISPs. (Ever notice how both our cellular networks and our broadband are lagging way behind European countries, Korea, and Japan?)
*Huge compared to, eg. Dell, which sells products in the most commoditized high-tech markets.
I'm not sure whether I should rejoice that more games will be getting competent writers, or weep that gaming is going to be degraded to sitcom quality.
That's a good point that too many people in the computer industry have yet to grasp, but there are some old, simple technologies that are really past their prime and survive on inertia alone. The example given above of a mouse and cursor is a pretty good one. I'm quite sure that, given a well designed user interface, I could be far more productive with a multi-touch screen as a pointing device than with a mouse. The problem is that that would completely change the ergonomics of computer workstations and user interfaces (ie. the screen would have to be closer to horizontal than vertical, and buttons would have to be bigger and more round on average.) Those factors have done a pretty good job of keeping tablets off the desktops of non-artists.
This could actually be interpreted as partly Google's fault, for raising expectations of "beta" software. Which is exactly what Boot Camp on 10.4 is: a public beta that expired quite a while ago. In particular, when the beta software also involves your boot sector and the Windows bootloader, you should consider yourself lucky to have anything recoverable. (Of course, it doesn't sound like Windows was at fault here, but nobody should be surprised when something like this breaks.)
In the case of the OP on the Apple forums, it sounds like the biggest problem was that the person had less than 1GB free space on the OS X partition. Obviously, this is only indirectly due to BootCamp, but it did stop the OP from doing an "archive and re-install" of the OS. It is interesting that one person reported that running the 10.4.11 updater under 10.5 but applied to the 10.4.10 partition works, so it isn't a completely reliable bug.
It is also worth noting that nobody has reported an actual filesystem corruption requiring a reformat, so the linked article is just plain wrong. Using the "archive and install" option to roll back the OS seems to be a reliable workaround. (With the one exception noted above.)
If memory isn't in the working set, is it really hurting performance? For the vast majority of users, I think not. And the difference between browsers is definitely too small to care about. Besides, it isn't like anybody is limited on swap space.
Go take a class in basic microeconomics. When taxes are increased, suppliers always bear at least some of the extra tax burden, and sometimes all of it. This comes from a straightforward application of the laws of supply and demand.
Hmmm... How long ago was that? I wouldn't be surprised if things have been watered down a bit. That said, the current course listings for physics first mention GR in a graduate-level special topics class, and the undergrad Modern Physics class doesn't require any math beyond multivariable calculus, so it's coverage of GR can't include any of the math (PDEs and tensors, for starters).
Some workloads benefit from vector processors, and some don't. For now, it is best economically to keep vector co-processors separate from CPUs, and use the advances in chip tech to lower power consumption and add more cores to the CPU.
For example, many server workloads are handled best by a chip like Sun's UltraSparc T1, which doesn't have any floating point capabilities worth mentioning. People running that kind of server wouldn't buy a Xeon or Opteron that had a 600M-transistor vector processor. It's a huge waste of money. Similarly, people with low-end PCs would probably never use such an integrated vector processor fully, so competition would keep that kind of CPU out of that market.
That leaves pretty much just the gaming and scientific computation markets. Of the two workloads, the former is occasionally CPU-bound rather than GPU bound, but most of the time, the vector processor is the biggest bottleneck by far for both workloads. In that case, it is much more economical if you can upgrade the vector processor without throwing away a perfectly good CPU.
You need a at least a graduate degree in physics to really understand this stuff. Things like General Relativity generally aren't covered in undergrad physics curriculums due to a lack of higher-level math. Ditto for the nuances of the Standard Model.
So unless you already understand what Lie algebras are and you've got the Schrödinger equation memorized, you're talking about spending the next several years sifting through Wolfram MathWorld, Wikipedia, and various other websites.
The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no warrants shall issue, but upon probable cause, supported by oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized. Clear enough? No warrant, no searches or seizures of my stuff. They are particularly prohibited from searching through all of my correspondence without a warrant.
Unlike e-mail, social networking sites are built from the ground up to have features like whitelisting and identity verification. They are also controlled by real entities that have the power to ban spammers and others that abuse the system. Quite surprisingly, these sites actually turn out to have real technical advantages over e-mail.
Of course, with things like PGP, e-mail can be extended to be a far better messaging system than social networking sites, and you get privacy, too. If only it were "cool" to use cryptographic signing and encryption.
More ad-hominem. Any science on the menu? If you have any well-reasoned alternative theories, do please post them (or better yet, publish them where more scientists will be watching). And if you would care to respond directly to any of my points, I would probably read it.
By the way, when did you last attend a university science class?
Whoops. I seem to have given a knee-jerk response to a troll.
By the looks of his sig, his hobby is to say "but what if your most basic, obvious principles are completely wrong?" without offering a better explanation. And without even realizing that people before him (and smarter than him) have asked the same question, and always verified the conventional wisdom. I realize now that he is wholly unconcerned with science, and merely dabbles in pseudo-science and demagoguery pandering towards those even less educated than him.
Do you really expect language like that to cause somebody to re-evaluate some of the most well-verified physical laws ever postulated?
"Einstein fanatics"?! "demigod"?! You sound like a crackpot UFO conspiracy theorist. If you think there are flaws with the current models, the only acceptable way to address those concerns is with science. Not ad-hominem attacks against people who are demonstrably smarter and more polite than you.
Please. It really isn't hard to show that the dependency on r can only take on a few values and still yield a universe that comes at all close to what we observe. For example, the only halfway-plausible power of r that allows closed orbits (such as planets around stars) in classical mechanics is exactly 2. All other values either don't allow closed orbits in general, or are trivially shown by experiment to be absurdly wrong. Now, we have observed that orbits aren't exactly closed (the most famous example being the precession of the perihelion of Mercury), but these were explained astoundingly well by relativity.
Astrophysics is way beyond getting the growth rate of a fundamental force wrong.
You seem to have missed my major point: it isn't hard to make a game engine run at different resolutions. It doesn't make memory requirements exorbitant. Hell, even DOOM can render to a dozen different screen sizes. With a console like the Wii using OpenGL, it is pretty much trivial for the game developers to enable higher resolutions when running on the more powerful device. It certainly doesn't approach double the effort. After all, it isn't like the resolution is hard-coded with assembly language. Even N64 games were written in C. And practically every 3d engine already includes a level-of-detail scaling system to use for rendering distant objects. It's been well established that 3d graphics is one of the most trivially scalable computing tasks, and you have yet to offer any reason why that can't apply to consoles.
And while is may not be such a good idea for Nintendo right now, the situation will probably be very different two years from now. Certainly when the HD-DVD/Blu-ray thing gets settled, Nintendo will have incentive to release a compatible, HD-capable Wii. But even now, Nintendo is preparing to release a new revision of the Wii that supports DVD playback. Would anybody be surprised if it included a faster GPU with better decoding features? Or if they added 802.11n support next year?
The days are long gone where games magically break if the clock speed of the CPU is too fast or too slow. Nobody in their right mind complains that their quad-core gaming rig is wasted on Half-life 2 just because the engine also runs on the original Xbox.
I see no reason why the gaming market must contain only the extremes: perfectly uniform consoles and infinitely many permutations of PC hardware. The example I gave, of improved hardware enabling higher resolutions, seems quite reasonable to me. It really isn't hard to modify a game engine to run at a different resolution, or to run at different resolutions on different hardware. Nintendo could offer several different speed grades of Wii without making it at all hard for games to scale completely across the lineup. It also wouldn't seem to be hard to market. There would be "HD capable" games and "HD capable" consoles, but with full compatibility with the non-HD counterparts. There wouldn't have to be any hassling with upgradability or expandability.
Do you have any justification for your assertion that anything less than perfect homogeneity in the console market would fail?
PC games can handle a fairly broad range of hardware capabilities without squandering the extra power on the high-end machines. There's no reason that can't also be done to a lesser extent on consoles. There's nothing outlandish about the idea of Nintendo releasing an HD-capable Wii a few years from now, and game devs writing games that run on both versions.
But the IBM PC came out in 1981, well before the ARM was created. And ARM was never a really good architecture for desktops anyways. By the time really good RISC chips were coming out (PPC, Alpha, etc.), the Wintel monopoly was cemented. And x86 chips have long been RISC under the hood, so the only real downside of x86 affects only low-level programming.
Slashdot Mirror
And how much of that is because the user interface is designed around having important elements at or near the edge of the screen so that they are easy to reach? And do your kiosks have the touch screen at eye level, with the screen vertical? That's the way most kiosks seem to work, and it is pretty obvious that such an arrangement doesn't demonstrate the advantages of a touch screen. To really asses the usability of a touch screen, you need to use a posture similar to that used for typing (but with more incline for the screen).
Assuming you have a good ergonomic set-up for your virtual tours, I don't see why a touch screen couldn't be easier than, or at least as easy as using a mouse. Panning could be done with as little as one finger, zooming could be accomplished with two fingers in a manner similar to the iPhone (saving you a round-trip to the edge of the screen to select a zooming tool), and changing location is the same finger tap as with the mouse. What's more is that you are much less at risk of RSI.
I can understand the webmail being hard to use with a touch screen, as those things are seldom easy to use under any conditions.
*Huge compared to, eg. Dell, which sells products in the most commoditized high-tech markets.
I'm not sure whether I should rejoice that more games will be getting competent writers, or weep that gaming is going to be degraded to sitcom quality.
That's a good point that too many people in the computer industry have yet to grasp, but there are some old, simple technologies that are really past their prime and survive on inertia alone. The example given above of a mouse and cursor is a pretty good one. I'm quite sure that, given a well designed user interface, I could be far more productive with a multi-touch screen as a pointing device than with a mouse. The problem is that that would completely change the ergonomics of computer workstations and user interfaces (ie. the screen would have to be closer to horizontal than vertical, and buttons would have to be bigger and more round on average.) Those factors have done a pretty good job of keeping tablets off the desktops of non-artists.
This could actually be interpreted as partly Google's fault, for raising expectations of "beta" software. Which is exactly what Boot Camp on 10.4 is: a public beta that expired quite a while ago. In particular, when the beta software also involves your boot sector and the Windows bootloader, you should consider yourself lucky to have anything recoverable. (Of course, it doesn't sound like Windows was at fault here, but nobody should be surprised when something like this breaks.)
In the case of the OP on the Apple forums, it sounds like the biggest problem was that the person had less than 1GB free space on the OS X partition. Obviously, this is only indirectly due to BootCamp, but it did stop the OP from doing an "archive and re-install" of the OS. It is interesting that one person reported that running the 10.4.11 updater under 10.5 but applied to the 10.4.10 partition works, so it isn't a completely reliable bug.
It is also worth noting that nobody has reported an actual filesystem corruption requiring a reformat, so the linked article is just plain wrong. Using the "archive and install" option to roll back the OS seems to be a reliable workaround. (With the one exception noted above.)
If you can come up with a way for the web browser to know how much memory is reasonable for a plugin to use, please tell us.
If memory isn't in the working set, is it really hurting performance? For the vast majority of users, I think not. And the difference between browsers is definitely too small to care about. Besides, it isn't like anybody is limited on swap space.
Go take a class in basic microeconomics. When taxes are increased, suppliers always bear at least some of the extra tax burden, and sometimes all of it. This comes from a straightforward application of the laws of supply and demand.
The iMac uses mobile chips so that it can stay thin and quiet. Apple wouldn't put a power-hungry desktop GPU in the iMac.
Hmmm... How long ago was that? I wouldn't be surprised if things have been watered down a bit. That said, the current course listings for physics first mention GR in a graduate-level special topics class, and the undergrad Modern Physics class doesn't require any math beyond multivariable calculus, so it's coverage of GR can't include any of the math (PDEs and tensors, for starters).
Some workloads benefit from vector processors, and some don't. For now, it is best economically to keep vector co-processors separate from CPUs, and use the advances in chip tech to lower power consumption and add more cores to the CPU.
For example, many server workloads are handled best by a chip like Sun's UltraSparc T1, which doesn't have any floating point capabilities worth mentioning. People running that kind of server wouldn't buy a Xeon or Opteron that had a 600M-transistor vector processor. It's a huge waste of money. Similarly, people with low-end PCs would probably never use such an integrated vector processor fully, so competition would keep that kind of CPU out of that market.
That leaves pretty much just the gaming and scientific computation markets. Of the two workloads, the former is occasionally CPU-bound rather than GPU bound, but most of the time, the vector processor is the biggest bottleneck by far for both workloads. In that case, it is much more economical if you can upgrade the vector processor without throwing away a perfectly good CPU.
You need a at least a graduate degree in physics to really understand this stuff. Things like General Relativity generally aren't covered in undergrad physics curriculums due to a lack of higher-level math. Ditto for the nuances of the Standard Model.
So unless you already understand what Lie algebras are and you've got the Schrödinger equation memorized, you're talking about spending the next several years sifting through Wolfram MathWorld, Wikipedia, and various other websites.
The place for hardware decoders is on the graphics card. Hence the reason why Linux needs to use the CPU.
"Reasonable" doesn't even come into play until you're trying to get the warrant.
IE has to open an external program just to show you the html source!? You'd think they could include some kind of mechanism to display text...
Unlike e-mail, social networking sites are built from the ground up to have features like whitelisting and identity verification. They are also controlled by real entities that have the power to ban spammers and others that abuse the system. Quite surprisingly, these sites actually turn out to have real technical advantages over e-mail.
Of course, with things like PGP, e-mail can be extended to be a far better messaging system than social networking sites, and you get privacy, too. If only it were "cool" to use cryptographic signing and encryption.
More ad-hominem. Any science on the menu? If you have any well-reasoned alternative theories, do please post them (or better yet, publish them where more scientists will be watching). And if you would care to respond directly to any of my points, I would probably read it.
By the way, when did you last attend a university science class?
Whoops. I seem to have given a knee-jerk response to a troll.
By the looks of his sig, his hobby is to say "but what if your most basic, obvious principles are completely wrong?" without offering a better explanation. And without even realizing that people before him (and smarter than him) have asked the same question, and always verified the conventional wisdom. I realize now that he is wholly unconcerned with science, and merely dabbles in pseudo-science and demagoguery pandering towards those even less educated than him.
Do you really expect language like that to cause somebody to re-evaluate some of the most well-verified physical laws ever postulated?
"Einstein fanatics"?! "demigod"?! You sound like a crackpot UFO conspiracy theorist. If you think there are flaws with the current models, the only acceptable way to address those concerns is with science. Not ad-hominem attacks against people who are demonstrably smarter and more polite than you.
Please. It really isn't hard to show that the dependency on r can only take on a few values and still yield a universe that comes at all close to what we observe. For example, the only halfway-plausible power of r that allows closed orbits (such as planets around stars) in classical mechanics is exactly 2. All other values either don't allow closed orbits in general, or are trivially shown by experiment to be absurdly wrong. Now, we have observed that orbits aren't exactly closed (the most famous example being the precession of the perihelion of Mercury), but these were explained astoundingly well by relativity.
Astrophysics is way beyond getting the growth rate of a fundamental force wrong.
You seem to have missed my major point: it isn't hard to make a game engine run at different resolutions. It doesn't make memory requirements exorbitant. Hell, even DOOM can render to a dozen different screen sizes. With a console like the Wii using OpenGL, it is pretty much trivial for the game developers to enable higher resolutions when running on the more powerful device. It certainly doesn't approach double the effort. After all, it isn't like the resolution is hard-coded with assembly language. Even N64 games were written in C. And practically every 3d engine already includes a level-of-detail scaling system to use for rendering distant objects. It's been well established that 3d graphics is one of the most trivially scalable computing tasks, and you have yet to offer any reason why that can't apply to consoles.
And while is may not be such a good idea for Nintendo right now, the situation will probably be very different two years from now. Certainly when the HD-DVD/Blu-ray thing gets settled, Nintendo will have incentive to release a compatible, HD-capable Wii. But even now, Nintendo is preparing to release a new revision of the Wii that supports DVD playback. Would anybody be surprised if it included a faster GPU with better decoding features? Or if they added 802.11n support next year?
The days are long gone where games magically break if the clock speed of the CPU is too fast or too slow. Nobody in their right mind complains that their quad-core gaming rig is wasted on Half-life 2 just because the engine also runs on the original Xbox.
I see no reason why the gaming market must contain only the extremes: perfectly uniform consoles and infinitely many permutations of PC hardware. The example I gave, of improved hardware enabling higher resolutions, seems quite reasonable to me. It really isn't hard to modify a game engine to run at a different resolution, or to run at different resolutions on different hardware. Nintendo could offer several different speed grades of Wii without making it at all hard for games to scale completely across the lineup. It also wouldn't seem to be hard to market. There would be "HD capable" games and "HD capable" consoles, but with full compatibility with the non-HD counterparts. There wouldn't have to be any hassling with upgradability or expandability.
Do you have any justification for your assertion that anything less than perfect homogeneity in the console market would fail?
PC games can handle a fairly broad range of hardware capabilities without squandering the extra power on the high-end machines. There's no reason that can't also be done to a lesser extent on consoles. There's nothing outlandish about the idea of Nintendo releasing an HD-capable Wii a few years from now, and game devs writing games that run on both versions.
Hey, don't blame me! I'm part of your generation!
But the IBM PC came out in 1981, well before the ARM was created. And ARM was never a really good architecture for desktops anyways. By the time really good RISC chips were coming out (PPC, Alpha, etc.), the Wintel monopoly was cemented. And x86 chips have long been RISC under the hood, so the only real downside of x86 affects only low-level programming.