Sure, it would be better if the computer could magically figure out what your brain means, but it would be a lot harder than letting the brain adapt naturally. We could have this technology now instead of in 10 years after we figure out how the brain's signals work. If I was a paraplegic, I'd happily spend years learning to work a brain implant that would allow me to affect the world again, even if it was hard. I mean, what else would I do?
Well, thanks for the STL tip. Unfortunately, I can find no reference to the at member in any documentation except MSDN's STL docs, which I tend to avoid. Did I mention that the available STL documentation sucks? It reads like a specification instead of a usage guide. While a specification is nice to have, a usage guide is what coders need, preferably with examples. GOOD documentation would note at the top of the page for vector that operator[] is not bounds checked, but at is. Anyway, yes, that is a HUGE GIANT design flaw. In addition, you can still iterate iterators past the end of their collections, no bounds checking there. Partial bounds checking that many (I'd say most) people are unaware of is almost as bad as no bounds checking at all.
I think getting the brain to control limbs through this process will be much easier than anticipated. I don't think you actually need to figure out which part of the brain is used to move the legs. All you need to do is connect this chip to some random patch of neurons in the general vicinity of your motor control area. Then, instead of making the computer figure out what patterns of neuronal activity mean "move leg", the person with the implant could train their brain to produce specific patterns for the computer. The brain is not well understood but it is much more flexible than the computer, so making it adapt to the computer would probably be much easier than the opposite. I'll bet that over a period of years, a person could learn to use artificial limbs in this way and be just as adept with them as with their natural limbs (if the artificial limbs were as physically capable).
Are you saying that vectors have bounds checking? When was the last time you used a vector? Vectors are completely unsafe. If you make a mistake (like going off the end, or deleting/inserting an element in a vector that's being iterated over), you don't get nice compile-time or even nice run-time errors. You get big fat segfaults, or even worse, it gives you trash data and your program crashes a minute later doing something unrelated. That's my biggest problem with the C++ standard library. Sure, they've got some nice concepts in there, but you better not mess up or you're screwed. C++ doesn't need more ways to screw up, there are plenty of those already. What C++ needs is ways to *avoid* screwing up, and that's not what the standard library provides. (My second biggest problem with the standard library is that the way they do things is so damn "elegant," and the documentation is so full of jargon, you need a degree in CS and at least a year of stl coding experience to fully understand it. These things can be done in a way that's easier to understand.)
What C++ really needs is a supplementary "standard library for dummies" which provides bounds-checked dynamic arrays, iterators that work (or at least fail gracefully) when the object changes, ways to make your objects garbage collected, and lots and lots of functions that are useful in practice, such as conversions from various number types to strings and vice versa. (I hate to praise anything related to MFC, but CString's AppendFormat function is brilliant.) Standard math objects with overloaded operators like complex numbers, 2D and 3D vectors, and matrices would be nice too. Oh, and more concise syntax for common things like declaring an iterator.
There is an up-and-coming x86 (and PowerPC and ARM) emulator named QEMU that is over 60 times faster than Bochs (resulting in being ~4x slower than native code). It'll be a while before running Windows on QEMU is reliable, but it can be installed and booted on the virtual machine right now.
It's perfectly valid to list Mudd and Pomona separately, because Mudd has way more wireless access points than Pomona. They pretty much cover the entire Mudd campus, from the dorms to the dining hall to the academic end. It's really nice. The campuses don't overlap at all; they are quite well-defined and much farther apart than the range of any access point. AFAIK the other three colleges don't have very many WAPs, but I might be mistaken about that. It is easier for Mudd to cover its whole campus than many other schools because it's so small, but it still takes a lot of access points.
That's also why you do not need to accept the terms of the GPL to use any GPLed software.
This message really needs to get out more. Recently I've been seeing a lot of GPL programs with click-through license agreements (mostly windows programs using one of the major windows installers). Making people check a checkbox saying they agree to the GPL is dumb, because they don't have to agree just to use the software. It's fine to note that the software is licensed under the GPL, and even to show the GPL during the install process, but requiring agreement just to install the software is not right.
Well, it is obvious in that benchmark something *besides* the graphics card is limiting performance, since increasing resolution hardly even decreases the framerate. If you look at only the benchmarks where it appears that both cards are stressed to the max, NVidia's card does seem to be about twice as fast as ATI's when the new features of DirectX 9 are used. Of course that doesn't make the submitter's statement correct, but it is quite impressive for a one-generation improvement in card performance. (... at a cost of $500 + a beefy new power supply to feed it? yikes)
The problem with backwards compatibility is: if you never break compatibility to fix things that are broken, you end up with a monstrosity like the Win32 API, MFC, COM, and ATL mess that Microsoft has. Backwards compatibility is convenient, but it can be a problem too. The key is breaking compatibility in such a way that applications doing things the old way and the new way can run at the same time. That way users don't have to worry about it, only programmers do.
But you can always tell whether or not you're in a rotating frame. Centrifugal force is never necessary to explain the motion of objects because you can always just use a non-rotating frame. There is no analagous way to get rid of gravity; it is necessary to explain the movements of objects in space.
Here's the short explanation of why the centrifugal force is "imaginary": The centrifugal force is used only by people who are in rotating frames (for example: standing on the edge of a turntable) to explain why balls they drop tend to fall away from them instead of straight down. An observer who is not rotating needs no centrifugal force to explain this: the ball simply travels in a straight line after being released, while the person on the turntable moves in a curve. The centrifugal force is a convenient shortcut for people on the turntable because by using it they can assume they are at rest and still use Newtonian physics even though they are really rotating. (As the site explains, this also requires the use of the coriolis force to completely explain the motion of objects from the turntable perspective.)
The centripetal force is the force inward keeping the person on the turntable moving in a circle around the center instead of moving off in a straight line due to inertia. Both the person on the turntable and the observer can agree that this force exists; the person on the turntable can feel the acceleration. Therefore the centripetal force is "real" because everyone agrees it exists no matter what frame they're in.
At least here at Slashdot we can have some educating "goin on".
Let me provide a counterpoint. No, increasing usability does not decrease security. Here's why:
A lot of viruses exploit flaws in OS/application code, usually C/C++. These flaws are completely unrelated to usability issues, so increasing usability does not require these flaws to become more common.
All other viruses are actually *caused* by usability *flaws*. This includes those viruses that come as.pif or.zip files, and spyware that installs itself by instructing users to click "yes" on IE warning dialogs. The problem is not that users are stupid, the problem is that usability is bad. Truly usable software would always inform the user of the consequences of their actions in a way the user can understand, and not allow various ways of "tricking" the user into running something dangerous. Fixing these flaws increases usability and security.
Bad usability can cause security breaches in other ways: users can be unaware that they just shared their entire hard drive to the Internet with write access, or that there is hidden information inside Word documents, or things like that. Usable software always informs the user of the consequences of their actions. Increasing usability increases user awareness of security issues.
Increasing usability can increase code complexity, which means that there will be more bugs. However, the security problems fixed by the increased usability outweigh this, especially when safe languages are used so that code execution bugs aren't a possibility.
Usability is not the problem. Bloated, complex code in unsafe languages is the problem. The two are not necessarily associated. Increased usability increases security due to increased user awareness of security issues.
If it's simply weight that they guy is adding to the camera, to make sudden changes in momentum more difficult
That's not at all what he's doing. The key to the steadycam is that the center of gravity of the apparatus is inside the handle (which is why you need a weight on the end of a pole to counterbalance the camcorder). This means that as you yank the camcorder around by the handle, *only* the position of the camera changes, not the orientation. This removes the much of the "jerkiness" of handheld shots that otherwise screams "low-budget amateur video!" Even without a gimbal mount for the handle, this device can still reduce handheld video jerkiness by a significant amount. Of course a gimbal-mounted handle would be better and would allow easier smooth panning, but it would be hard to do for $14 with commonly-available parts and easy assembly.
This problem is solved, by Fontconfig and Xft. Modern linux distros have a directory you can drop fonts in where they will be automatically detected and immediately available.
With a decent package manager (APT, for example) the user doesn't have to know whether or not a new version of glibc is involved, so who cares? It Just Works. Your suggestion of static linking for everything shows you haven't put much thought into this. The problem with static linking isn't disk space, it's memory usage.
Once again, who cares about libraries? Multiple versions of the same library can be installed, and the package manager takes care of it all for you.
This makes no sense. Obviously you're going to have to at least upgrade for things to change. You don't have to switch distributions though, you can just upgrade to the latest version of your favorite and odds are these things will be fixed. If they're not then that's your distro's fault, so go complain to them, not Slashdot.
This is a great idea. We were just thinking the other day about a video capture based stat tracking system for our Halo LAN matches. Sometimes large portions of our potential Halo-playing time are taken up by arguments about team balancing (AKA "team wanking"), but if a computer could analyze the stats and make balanced teams for us we wouldn't have to worry about that anymore. Stats on who's best/worst at killing who, and with what weapons, would be really interesting. Plus the automated trash talking would be great fun! Having a robotic voice affirm your greatness after you get a triple killtacular would just be that much more awesome:-)
Those are KDE's standard icons. They have been made by the great Everaldo specifically for the KDE project, and have been subsequently adopted by many projects (even some on Windows) because they are one of the best-looking, most-complete, and most-free icon sets out there. Everaldo was an employee of Conectiva when he started making the icon set, but Conectiva made the icons freely available.
I believe the red team was using extremely high-resolution elevation data and pictures of the race area for path planning. So they might have computed more in-between waypoints based on this data and used GPS to travel between them, with cameras and rangefinders and stuff on-board mostly for random obstacle avoidance. That would explain the team member's comments. Also it's possible that the course started out easy with many GPS waypoints and got harder later on with more path planning necessary. Since no bot traveled more than 7 miles I doubt any would have gotten to the hard section.
You have to be really really careful about putting it label side down, too.
Um, that's what he just said. You shouldn't lay CDs down upside-down, because the side that's most vulnerable to damage is the label side, not the clear side (counterintuitively). Also, writing on a cd label with a ballpoint pen or hard pencil makes an instant coaster because you just crush the metal layer that has all the tiny pits (try it on a junk CDR and look at the other side, you can see the damage). Damage to the clear side is mostly harmless unless it's really extreme, so it really isn't a problem to put CDs on a surface with the clear side down (unless you put something abrasive on top of them).
Like I said, even if the CEO reads it and approves it personally (as opposed to his/her secretary), it's still deceptive to credit the statement to the CEO. Saying something is quite different from approving what someone else said.
You're right, I haven't been exposed to any large "enterprise" in the last 50 or 60 years. Is that so surprising? I'm not a veteran of the business world, I'm in college! I'm about to graduate and get my first real job. I don't know the tricks of the PR trade, and I'm sure there are many other Slashdotters who don't. Is that so hard to believe? Sheesh, give me a break!
Did you actually think the pr people were interviewing the ceo for a press release?
Um... yes? The entire point of this practice is to decieve people into thinking that the CEO/VP/whoever actually said that stuff (even if they did sign off on it personally, after reading it personally, which is probably not how it happens, it's still not the same as saying it). Now that I'm aware of it, I won't be fooled anymore, but people who don't work in PR are not aware that PR people do this.
You don't need to learn assembly, you can use C with TI-GCC. Much easier, but still just as fast. Actually, to stay on topic here, a 48-hour game competition for the TI-89 would be a fun thing to do. Since calculator games don't need to be as complex as computer games, the resulting games might actually see widespread play after the competition too.
You know what? As a consumer, what I really want is Half-Life 2. I want Doom 3. I want Grand Theft Auto 5. I'm not afraid to admit it. I don't think there's anything wrong with producing game sequels. I think incremental improvments are still improvements. I like an original, completely different game just as much as anyone else, but all these people bitching about the number of sequels/clones are missing the point. The game industry is huge now, and still growing. There's more than enough room for sequels, clones, *and* original games. If you can't find any original games out there, you're just not looking hard enough (hint: there's more to the game industry than what you'll find at your local Wal-Mart).
Slashdot Mirror
Sure, it would be better if the computer could magically figure out what your brain means, but it would be a lot harder than letting the brain adapt naturally. We could have this technology now instead of in 10 years after we figure out how the brain's signals work. If I was a paraplegic, I'd happily spend years learning to work a brain implant that would allow me to affect the world again, even if it was hard. I mean, what else would I do?
In closing, I hate the STL ;-)
# find /proc | xargs cat
I think getting the brain to control limbs through this process will be much easier than anticipated. I don't think you actually need to figure out which part of the brain is used to move the legs. All you need to do is connect this chip to some random patch of neurons in the general vicinity of your motor control area. Then, instead of making the computer figure out what patterns of neuronal activity mean "move leg", the person with the implant could train their brain to produce specific patterns for the computer. The brain is not well understood but it is much more flexible than the computer, so making it adapt to the computer would probably be much easier than the opposite. I'll bet that over a period of years, a person could learn to use artificial limbs in this way and be just as adept with them as with their natural limbs (if the artificial limbs were as physically capable).
What C++ really needs is a supplementary "standard library for dummies" which provides bounds-checked dynamic arrays, iterators that work (or at least fail gracefully) when the object changes, ways to make your objects garbage collected, and lots and lots of functions that are useful in practice, such as conversions from various number types to strings and vice versa. (I hate to praise anything related to MFC, but CString's AppendFormat function is brilliant.) Standard math objects with overloaded operators like complex numbers, 2D and 3D vectors, and matrices would be nice too. Oh, and more concise syntax for common things like declaring an iterator.
There is an up-and-coming x86 (and PowerPC and ARM) emulator named QEMU that is over 60 times faster than Bochs (resulting in being ~4x slower than native code). It'll be a while before running Windows on QEMU is reliable, but it can be installed and booted on the virtual machine right now.
It's perfectly valid to list Mudd and Pomona separately, because Mudd has way more wireless access points than Pomona. They pretty much cover the entire Mudd campus, from the dorms to the dining hall to the academic end. It's really nice. The campuses don't overlap at all; they are quite well-defined and much farther apart than the range of any access point. AFAIK the other three colleges don't have very many WAPs, but I might be mistaken about that. It is easier for Mudd to cover its whole campus than many other schools because it's so small, but it still takes a lot of access points.
This message really needs to get out more. Recently I've been seeing a lot of GPL programs with click-through license agreements (mostly windows programs using one of the major windows installers). Making people check a checkbox saying they agree to the GPL is dumb, because they don't have to agree just to use the software. It's fine to note that the software is licensed under the GPL, and even to show the GPL during the install process, but requiring agreement just to install the software is not right.
Well, it is obvious in that benchmark something *besides* the graphics card is limiting performance, since increasing resolution hardly even decreases the framerate. If you look at only the benchmarks where it appears that both cards are stressed to the max, NVidia's card does seem to be about twice as fast as ATI's when the new features of DirectX 9 are used. Of course that doesn't make the submitter's statement correct, but it is quite impressive for a one-generation improvement in card performance. (... at a cost of $500 + a beefy new power supply to feed it? yikes)
The problem with backwards compatibility is: if you never break compatibility to fix things that are broken, you end up with a monstrosity like the Win32 API, MFC, COM, and ATL mess that Microsoft has. Backwards compatibility is convenient, but it can be a problem too. The key is breaking compatibility in such a way that applications doing things the old way and the new way can run at the same time. That way users don't have to worry about it, only programmers do.
But you can always tell whether or not you're in a rotating frame. Centrifugal force is never necessary to explain the motion of objects because you can always just use a non-rotating frame. There is no analagous way to get rid of gravity; it is necessary to explain the movements of objects in space.
Here's the short explanation of why the centrifugal force is "imaginary": The centrifugal force is used only by people who are in rotating frames (for example: standing on the edge of a turntable) to explain why balls they drop tend to fall away from them instead of straight down. An observer who is not rotating needs no centrifugal force to explain this: the ball simply travels in a straight line after being released, while the person on the turntable moves in a curve. The centrifugal force is a convenient shortcut for people on the turntable because by using it they can assume they are at rest and still use Newtonian physics even though they are really rotating. (As the site explains, this also requires the use of the coriolis force to completely explain the motion of objects from the turntable perspective.)
The centripetal force is the force inward keeping the person on the turntable moving in a circle around the center instead of moving off in a straight line due to inertia. Both the person on the turntable and the observer can agree that this force exists; the person on the turntable can feel the acceleration. Therefore the centripetal force is "real" because everyone agrees it exists no matter what frame they're in.
At least here at Slashdot we can have some educating "goin on".
- A lot of viruses exploit flaws in OS/application code, usually C/C++. These flaws are completely unrelated to usability issues, so increasing usability does not require these flaws to become more common.
- All other viruses are actually *caused* by usability *flaws*. This includes those viruses that come as
.pif or .zip files, and spyware that installs itself by instructing users to click "yes" on IE warning dialogs. The problem is not that users are stupid, the problem is that usability is bad. Truly usable software would always inform the user of the consequences of their actions in a way the user can understand, and not allow various ways of "tricking" the user into running something dangerous. Fixing these flaws increases usability and security.
- Bad usability can cause security breaches in other ways: users can be unaware that they just shared their entire hard drive to the Internet with write access, or that there is hidden information inside Word documents, or things like that. Usable software always informs the user of the consequences of their actions. Increasing usability increases user awareness of security issues.
- Increasing usability can increase code complexity, which means that there will be more bugs. However, the security problems fixed by the increased usability outweigh this, especially when safe languages are used so that code execution bugs aren't a possibility.
Usability is not the problem. Bloated, complex code in unsafe languages is the problem. The two are not necessarily associated. Increased usability increases security due to increased user awareness of security issues.It's a Sierpinski triangle, or at least an approximation thereof.
That's not at all what he's doing. The key to the steadycam is that the center of gravity of the apparatus is inside the handle (which is why you need a weight on the end of a pole to counterbalance the camcorder). This means that as you yank the camcorder around by the handle, *only* the position of the camera changes, not the orientation. This removes the much of the "jerkiness" of handheld shots that otherwise screams "low-budget amateur video!" Even without a gimbal mount for the handle, this device can still reduce handheld video jerkiness by a significant amount. Of course a gimbal-mounted handle would be better and would allow easier smooth panning, but it would be hard to do for $14 with commonly-available parts and easy assembly.
This is a great idea. We were just thinking the other day about a video capture based stat tracking system for our Halo LAN matches. Sometimes large portions of our potential Halo-playing time are taken up by arguments about team balancing (AKA "team wanking"), but if a computer could analyze the stats and make balanced teams for us we wouldn't have to worry about that anymore. Stats on who's best/worst at killing who, and with what weapons, would be really interesting. Plus the automated trash talking would be great fun! Having a robotic voice affirm your greatness after you get a triple killtacular would just be that much more awesome :-)
Those are KDE's standard icons. They have been made by the great Everaldo specifically for the KDE project, and have been subsequently adopted by many projects (even some on Windows) because they are one of the best-looking, most-complete, and most-free icon sets out there. Everaldo was an employee of Conectiva when he started making the icon set, but Conectiva made the icons freely available.
I believe the red team was using extremely high-resolution elevation data and pictures of the race area for path planning. So they might have computed more in-between waypoints based on this data and used GPS to travel between them, with cameras and rangefinders and stuff on-board mostly for random obstacle avoidance. That would explain the team member's comments. Also it's possible that the course started out easy with many GPS waypoints and got harder later on with more path planning necessary. Since no bot traveled more than 7 miles I doubt any would have gotten to the hard section.
Um, that's what he just said. You shouldn't lay CDs down upside-down, because the side that's most vulnerable to damage is the label side, not the clear side (counterintuitively). Also, writing on a cd label with a ballpoint pen or hard pencil makes an instant coaster because you just crush the metal layer that has all the tiny pits (try it on a junk CDR and look at the other side, you can see the damage). Damage to the clear side is mostly harmless unless it's really extreme, so it really isn't a problem to put CDs on a surface with the clear side down (unless you put something abrasive on top of them).
Like I said, even if the CEO reads it and approves it personally (as opposed to his/her secretary), it's still deceptive to credit the statement to the CEO. Saying something is quite different from approving what someone else said.
You're right, I haven't been exposed to any large "enterprise" in the last 50 or 60 years. Is that so surprising? I'm not a veteran of the business world, I'm in college! I'm about to graduate and get my first real job. I don't know the tricks of the PR trade, and I'm sure there are many other Slashdotters who don't. Is that so hard to believe? Sheesh, give me a break!
Um... yes? The entire point of this practice is to decieve people into thinking that the CEO/VP/whoever actually said that stuff (even if they did sign off on it personally, after reading it personally, which is probably not how it happens, it's still not the same as saying it). Now that I'm aware of it, I won't be fooled anymore, but people who don't work in PR are not aware that PR people do this.
You don't need to learn assembly, you can use C with TI-GCC. Much easier, but still just as fast. Actually, to stay on topic here, a 48-hour game competition for the TI-89 would be a fun thing to do. Since calculator games don't need to be as complex as computer games, the resulting games might actually see widespread play after the competition too.
You know what? As a consumer, what I really want is Half-Life 2. I want Doom 3. I want Grand Theft Auto 5. I'm not afraid to admit it. I don't think there's anything wrong with producing game sequels. I think incremental improvments are still improvements. I like an original, completely different game just as much as anyone else, but all these people bitching about the number of sequels/clones are missing the point. The game industry is huge now, and still growing. There's more than enough room for sequels, clones, *and* original games. If you can't find any original games out there, you're just not looking hard enough (hint: there's more to the game industry than what you'll find at your local Wal-Mart).