What about archived snapshots of my Windows VM? Or for that matter, my entire Time Machine back-end? Block-level deduplication is basically a prerequisite for being able to have Time Machine back up a virtual hard drive image.
Yes. The antenna on my GPS receiver wouldn't fit into any smartphone, even if you took out the phone bits. I suspect if you took a smartphone into a deep valley with thick tree cover, you would find that it couldn't track satellites as well as even the cheapest dedicated receivers, let alone as well as a high-end unit. Also, no smartphone that I've yet heard of has a barometric altimeter, which means the altitude readings off a smartphone are pretty much a novelty. (The last time I went sea kayaking with a GPS, the altitude reported was -6ft the whole time, despite the receiver clearly being 8inches above the water.) An accurate altimeter can be very useful when navigating and geocaching in the mountains.
However, all the above reasons pale in comparison the the fact that my heart doesn't skip a few beats when I drop my GPS onto a rock in a river. I know it will be fine as long as I fetch it out before it gets washed downstream. If the same happened with an iPhone (which is far more likely to be dropped, due to the smooth, slick surfaces), I'd have good reason to panic.
And what is the actual thing ISPs provide access to? The Internet. But what is "the internet"? It certainly isn't a physical object that resides in a specific place. It's a communication system. Packet based, even. Which makes it very much an electronic embodiment of the idea of a system of post offices and post roads. All that internet service providers really do is deliver packets of information for their customers. At a high level, the service that ISPs provide is fundamentally the same as kind of service as what the USPS provides.
I think it's more that nobody is taking seriously the fundamental differences between hard drives and flash. Nobody has really stopped to do a comprehensive assessment of what existing assumptions embodied in our software and users will be broken by flash memory that is asymmetric in both access speed and access granularity. As a result, the pre-Intel flash SSD controllers made really stupid trade-offs, and they ended up with drives that were less suitable for the consumer market than ordinary hard drives. Once Intel made everybody realize that latency and IOPS mattered a lot more to consumers than throughput, people moved on to the next difference, and started complaining about the lower write performance of a nearly full SSD. Even today, I still see people referring to it as a "bug", when it is nothing more than an inherent difference from the spinning platters of hard drives. Smart garbage collection (which requires smart OS support) is a way of hiding the limitation, but the lack of it isn't a bug any more than a hard drive with a small cache is faulty. It just has obvious room for improvement.
OS X has the only OpenCL implementation that allows you to use CPUs and GPUs to run compute kernels from the same context. NVidia's implementation is GPU-only, and ATI's seems to still be CPU-only, and you can't use them simultaneously.
To be fair, it doesn't really matter what version of Flash you're running. It still sucks, and is very insecure. The embarrassing part is more that it downgrades the version than that it exposes users to an extra security risk.
Apple's been using the UNIX trademark in relation to Snow Leopard for quite a while. Either 10.6 is certified (as Apple's website seems to imply), or The Open Group is in danger of losing their trademark.
What do you do when a security vulnerability or other serious bug is found in the version of the DLL? Do you trust the app vendor to be around to release an update in a timely fashion? Or would you prefer Microsoft releases an update that looks for any outdated DLLs anywhere on your hard drive and overwrites them? What if there are games that, as an anti-cheating measure, check the hash of their version of the DLL to detect tampering?
I think the point of the article is that new computers must be 64-bit capable in order to be advertised as Win7-ready. This is quit different from saying that computers being upgraded need 64-bit capabilities. In fact, Microsoft would be in huge trouble if they made Win7 refuse to install on non-64-bit capable machines, because the "release candidate" runs on machines as old as my 1.5Ghz Athlon XP, and such a drastic change in specs from something called a release candidate might not go over well with the FTC or the EU.
At worst, a phone in repeater mode would last as long as the normal talk time. However, if it's acting as a repeater in a dense mesh, it probably wouldn't need to (and shouldn't) transmit at as high a power as it would to reach a tower a mile away.
You haven't watched the hour+ long tech demo, have you? You seem to be completely unaware of it's capabilities for collaboratively building a document, or it's extension systems that mean people will be adding new capabilities all the time. It's a lot more than just an integration of email and IM.
At my university, we have a the VCL, a pool of blade servers accessible by RDP or SSH that get imaged on the fly when a user requests a machine with certain apps. These blades get wiped on log-out. (Home directories are of course stored elsewhere, and accessed over AFS.) This is very secure, but it lets students get admin access to their machine, and it also helps keep software licensing costs down, because it is trivial to limit the number of concurrent users of a package that isn't volume licensed. Performance when accessing the VCL on-campus is great, and in a corporate environment it could work great with thin clients.
I don't think that EEE can only apply to an open standard. It can also apply to an idea or a specific product (here, IE). In this case, Google's pretty much combined the steps: They've gone from despising and barely tolerating IE to embracing it, precisely because they've extended it (to properly support web standards and to support new standards), and for each convert, they're pretty close to extinguishing IE. Once Chrome Frame becomes popular and popular sites learn to detect Chrome Frame and disable their IE hacks, then Google will be able to effectively extinguish IE from their user's machines by releasing an update that sets Chrome as the default renderer for all pages. There probably isn't anything legal that Microsoft could do to stop them, because Chrome Frame is an opt-in piece of software.
Eyefinity seems pretty cool, but one thing I haven't read about is how it works for gaming spread across monitors of different sizes and resolutions. If I were to start using Eyefinity, I would want to buy a pair of 19-20" monitors to put in portrait mode to flank my 24" monitor. However, the two new monitors might have the same vertical size (in portrait mode) as the big one, they would have a slightly different resolution: 1280 vs. 1200. It makes things more complex, but it would be great if ATI could make this work well. What would be particularly useful is to allow the monitors on the periphery to have lower DPI, because peripheral vision is lower resolution anyways.
You're just trolling. Yes, fglrx might be hard to configure in some cases, but it definitely works with large monitors, and AMD/ATI has been releasing detailed specs for all their recent GPUs, so the open source drivers are the best out there, and can use pretty much all the features of the hardware (albeit with lower performance in some cases).
Touch screens don't work for calculators. Tactile feedback is too important, because most data entry into a calculator comes when the user is looking at stuff written on paper. If you can't touch type even a few digits, then the device is basically too unusable to be worth carrying around for calculating purposes.
And while high resolution displays would be nice, I don't think it needs to be anywhere as big as on a netbook. A 4-5" display would probably be the largest that could be effectively used by a calculator with capabilities comparable to what's on the market today.
I'm betting that the 50g cost about a third of what your phone would cost without subsidies.
Also, the technology under the hood is completely different. Instead of a 4Mhz custom CPU with a 4 bit data bus and half a megabyte of address space, you've now got a 75Mhz ARM SoC that could be running in the 200-266Mhz range if the batteries were bigger. The Linux kernel would run on this machine if it had more RAM. The tech industry has evolved so much that running an emulator on the new hardware is cheaper than fabricating more of the old hardware.
Why would I choose a netbook over my HP 50g for math? A netbook would be over twice the price, twice the size, have worse battery life, no number pad, and wouldn't come with any math software on it. My HP, on the other hand, comes with a specialized keyboard and a user interface to match that make it very efficient to input equations. In fact, the user interface is so much more efficient that for simple problems, the HP is quicker overall than a desktop with a full size keyboard and a fast CPU. And I suspect that the kinds of problems that are slow to run on my HP would also be slow running on an Intel Atom that had Windows and a Java UI (I'm looking at you, Maple!) getting in the way.
Graphing calculators are very useful tools in spite of the preponderance of general purpose computers, precisely because they are specialized. Just because a tool is specialized enough that it doesn't include functionality relevant to you doesn't mean that it is useless.
I know that TI doesn't do a lot of direct communication with teachers, but every high school math teacher I've met or otherwise dealt with has been able to decide which calculators, if any, may be used on their assignments and tests. On the other hand, I've never heard of a specific instance of a school-wide or higher level policy on calculator use, excepting the ones from ACT and the College Board.
Speaking as a mathematician with Mathematica, Maple, Matlab, and an HP 50g at my fingertips, I always reach for the HP first for any problem I don't feel like doing on paper. Partly because I am in applied math, basically any symbolic problem I need to do can be handled by my HP, and the user interface is far superior. (Not to mention the startup time!) For plotting and numerical problems, MATLAB gets almost as much use as the HP due to the size of the problems.
The 35 is the only true HP style scientific calculator they are making right now, but the 50g is most definitely an HP on the inside, even though the keyboard is barely above TI quality. From my experience, the only advantage the TI 89 has over the HP 50g for a college student or professional is a higher resolution screen.
Slashdot Mirror
What about archived snapshots of my Windows VM? Or for that matter, my entire Time Machine back-end? Block-level deduplication is basically a prerequisite for being able to have Time Machine back up a virtual hard drive image.
A stream of alpha particles would have a well-defined current, despite the lack of electrons.
Yes. The antenna on my GPS receiver wouldn't fit into any smartphone, even if you took out the phone bits. I suspect if you took a smartphone into a deep valley with thick tree cover, you would find that it couldn't track satellites as well as even the cheapest dedicated receivers, let alone as well as a high-end unit. Also, no smartphone that I've yet heard of has a barometric altimeter, which means the altitude readings off a smartphone are pretty much a novelty. (The last time I went sea kayaking with a GPS, the altitude reported was -6ft the whole time, despite the receiver clearly being 8inches above the water.) An accurate altimeter can be very useful when navigating and geocaching in the mountains.
However, all the above reasons pale in comparison the the fact that my heart doesn't skip a few beats when I drop my GPS onto a rock in a river. I know it will be fine as long as I fetch it out before it gets washed downstream. If the same happened with an iPhone (which is far more likely to be dropped, due to the smooth, slick surfaces), I'd have good reason to panic.
And what is the actual thing ISPs provide access to? The Internet. But what is "the internet"? It certainly isn't a physical object that resides in a specific place. It's a communication system. Packet based, even. Which makes it very much an electronic embodiment of the idea of a system of post offices and post roads. All that internet service providers really do is deliver packets of information for their customers. At a high level, the service that ISPs provide is fundamentally the same as kind of service as what the USPS provides.
I think it's more that nobody is taking seriously the fundamental differences between hard drives and flash. Nobody has really stopped to do a comprehensive assessment of what existing assumptions embodied in our software and users will be broken by flash memory that is asymmetric in both access speed and access granularity. As a result, the pre-Intel flash SSD controllers made really stupid trade-offs, and they ended up with drives that were less suitable for the consumer market than ordinary hard drives. Once Intel made everybody realize that latency and IOPS mattered a lot more to consumers than throughput, people moved on to the next difference, and started complaining about the lower write performance of a nearly full SSD. Even today, I still see people referring to it as a "bug", when it is nothing more than an inherent difference from the spinning platters of hard drives. Smart garbage collection (which requires smart OS support) is a way of hiding the limitation, but the lack of it isn't a bug any more than a hard drive with a small cache is faulty. It just has obvious room for improvement.
OS X has the only OpenCL implementation that allows you to use CPUs and GPUs to run compute kernels from the same context. NVidia's implementation is GPU-only, and ATI's seems to still be CPU-only, and you can't use them simultaneously.
The difference between one ISO terabyte and 1 TiB is relatively smaller than the variance among normal fingernails.
To be fair, it doesn't really matter what version of Flash you're running. It still sucks, and is very insecure. The embarrassing part is more that it downgrades the version than that it exposes users to an extra security risk.
Apple's been using the UNIX trademark in relation to Snow Leopard for quite a while. Either 10.6 is certified (as Apple's website seems to imply), or The Open Group is in danger of losing their trademark.
What do you do when a security vulnerability or other serious bug is found in the version of the DLL? Do you trust the app vendor to be around to release an update in a timely fashion? Or would you prefer Microsoft releases an update that looks for any outdated DLLs anywhere on your hard drive and overwrites them? What if there are games that, as an anti-cheating measure, check the hash of their version of the DLL to detect tampering?
I think the point of the article is that new computers must be 64-bit capable in order to be advertised as Win7-ready. This is quit different from saying that computers being upgraded need 64-bit capabilities. In fact, Microsoft would be in huge trouble if they made Win7 refuse to install on non-64-bit capable machines, because the "release candidate" runs on machines as old as my 1.5Ghz Athlon XP, and such a drastic change in specs from something called a release candidate might not go over well with the FTC or the EU.
At worst, a phone in repeater mode would last as long as the normal talk time. However, if it's acting as a repeater in a dense mesh, it probably wouldn't need to (and shouldn't) transmit at as high a power as it would to reach a tower a mile away.
You haven't watched the hour+ long tech demo, have you? You seem to be completely unaware of it's capabilities for collaboratively building a document, or it's extension systems that mean people will be adding new capabilities all the time. It's a lot more than just an integration of email and IM.
They've had Super User for a while now. It's not linux specific, but linux questions are very much welcome.
At my university, we have a the VCL, a pool of blade servers accessible by RDP or SSH that get imaged on the fly when a user requests a machine with certain apps. These blades get wiped on log-out. (Home directories are of course stored elsewhere, and accessed over AFS.) This is very secure, but it lets students get admin access to their machine, and it also helps keep software licensing costs down, because it is trivial to limit the number of concurrent users of a package that isn't volume licensed. Performance when accessing the VCL on-campus is great, and in a corporate environment it could work great with thin clients.
I don't think that EEE can only apply to an open standard. It can also apply to an idea or a specific product (here, IE). In this case, Google's pretty much combined the steps: They've gone from despising and barely tolerating IE to embracing it, precisely because they've extended it (to properly support web standards and to support new standards), and for each convert, they're pretty close to extinguishing IE. Once Chrome Frame becomes popular and popular sites learn to detect Chrome Frame and disable their IE hacks, then Google will be able to effectively extinguish IE from their user's machines by releasing an update that sets Chrome as the default renderer for all pages. There probably isn't anything legal that Microsoft could do to stop them, because Chrome Frame is an opt-in piece of software.
Eyefinity seems pretty cool, but one thing I haven't read about is how it works for gaming spread across monitors of different sizes and resolutions. If I were to start using Eyefinity, I would want to buy a pair of 19-20" monitors to put in portrait mode to flank my 24" monitor. However, the two new monitors might have the same vertical size (in portrait mode) as the big one, they would have a slightly different resolution: 1280 vs. 1200. It makes things more complex, but it would be great if ATI could make this work well. What would be particularly useful is to allow the monitors on the periphery to have lower DPI, because peripheral vision is lower resolution anyways.
You're just trolling. Yes, fglrx might be hard to configure in some cases, but it definitely works with large monitors, and AMD/ATI has been releasing detailed specs for all their recent GPUs, so the open source drivers are the best out there, and can use pretty much all the features of the hardware (albeit with lower performance in some cases).
Touch screens don't work for calculators. Tactile feedback is too important, because most data entry into a calculator comes when the user is looking at stuff written on paper. If you can't touch type even a few digits, then the device is basically too unusable to be worth carrying around for calculating purposes.
And while high resolution displays would be nice, I don't think it needs to be anywhere as big as on a netbook. A 4-5" display would probably be the largest that could be effectively used by a calculator with capabilities comparable to what's on the market today.
I'm betting that the 50g cost about a third of what your phone would cost without subsidies.
Also, the technology under the hood is completely different. Instead of a 4Mhz custom CPU with a 4 bit data bus and half a megabyte of address space, you've now got a 75Mhz ARM SoC that could be running in the 200-266Mhz range if the batteries were bigger. The Linux kernel would run on this machine if it had more RAM. The tech industry has evolved so much that running an emulator on the new hardware is cheaper than fabricating more of the old hardware.
Why would I choose a netbook over my HP 50g for math? A netbook would be over twice the price, twice the size, have worse battery life, no number pad, and wouldn't come with any math software on it. My HP, on the other hand, comes with a specialized keyboard and a user interface to match that make it very efficient to input equations. In fact, the user interface is so much more efficient that for simple problems, the HP is quicker overall than a desktop with a full size keyboard and a fast CPU. And I suspect that the kinds of problems that are slow to run on my HP would also be slow running on an Intel Atom that had Windows and a Java UI (I'm looking at you, Maple!) getting in the way.
Graphing calculators are very useful tools in spite of the preponderance of general purpose computers, precisely because they are specialized. Just because a tool is specialized enough that it doesn't include functionality relevant to you doesn't mean that it is useless.
Ah. I escaped the public school system before that was around (or at least before it made it to my schools).
I know that TI doesn't do a lot of direct communication with teachers, but every high school math teacher I've met or otherwise dealt with has been able to decide which calculators, if any, may be used on their assignments and tests. On the other hand, I've never heard of a specific instance of a school-wide or higher level policy on calculator use, excepting the ones from ACT and the College Board.
Speaking as a mathematician with Mathematica, Maple, Matlab, and an HP 50g at my fingertips, I always reach for the HP first for any problem I don't feel like doing on paper. Partly because I am in applied math, basically any symbolic problem I need to do can be handled by my HP, and the user interface is far superior. (Not to mention the startup time!) For plotting and numerical problems, MATLAB gets almost as much use as the HP due to the size of the problems.
The 35 is the only true HP style scientific calculator they are making right now, but the 50g is most definitely an HP on the inside, even though the keyboard is barely above TI quality. From my experience, the only advantage the TI 89 has over the HP 50g for a college student or professional is a higher resolution screen.