Although Carpel Tunnel Syndrome makes the news the most, the most common hand problems that computer users experience is tendonitis. Tendonitis is painful, but it doesn't damage nerves.
What evidence there is points to the mouse as the source of computer-related hand problems, not the keyboard. I remember hearing a lot about people getting computer-related hand problems in the early 90's, around the time Windows 3.1 came out and the mouse became mainstream.
I tend to have hand trouble in the spring when I start riding my bike, and that demands a hand position different from anything I do all winter.
I've had times when my tendonitis was so bad that I thought I'd have to change careers, but I found that push-ups were an effective treatment for me. Push-ups are great because they are a functional exercise that works the whole upper body -- they build up the big muscles in your chest as well as the little muscles in your wrist. I love lifting weights, but you can do push-ups at home without any equipment or gym memberships.
Note that every motion you do involves a complete 'chain' that leads back to your center of gravity -- if you push a key on a computer, it's only an ounce or two of force, but it's ultimately backed up by your whole mass. Your big muscles help your little muscles do their work and vice versa...
He's three and a half years old now. I started in the winter when he was about
one and a half. Mostly Nintendo 64 and SNES under emulation. His favorite game
is "Super Mario 64" and he always says "Wanna play mario in the castle".
He's starting to understand something about the games. One day we were playing
Zelda and there was a room with a crack in the wall he said, "Use a bomb
here!"
I've felt guilty about this, but I had a talk about this with the janitor in my building (who is as insightful about games as anyone) and his take was: "Didn't your parents play cards with you when you were a kid? Is this any different?"
There's a lot more to the genotype of a somatic cell than DNA: cells accumulate a whole bunch of 'markers' such as methylated bases and proteins stuck to the DNA, and repeat units that fall off the ends of the chromosome with every cell division.
Nature has good mechanism for making sure germ line (reproductive cells) stay in a good state, but manipulated cells never seem to be 100% right. Clones often end up with poor health and life expectancy because of this, and I'm afraid stem cell therapy will end in poor results, maybe even cancer.
It's bizzare that stem cells have become such an issue for the left and the right. I see Democrats screaming at the top of their lungs so we can have more research into medical treatments that we can't afford, while Republicans are blowing the ethical issues entirely out of proportion.
"Open DRM" at first sounds like a contradiction, yet, the modern approach in cryptographic systems is to design systems so that security depends on secret key material, not secret algorithms. It's a rule of nature that any piece of hardware that falls in the hands of the enemy will give up its secrets, and algorithm secrecy didn't stop Jon from cracking DVD encryption.
In an open DRM system, anybody could create their own DRM "universe" by generating their own set of keys to initialize the system -- this opens the possibility of using DRM to do different things than today's systems, such as protecting privacy: Sun is quite interested in providing storage records for medical records and such, and some kind of DRM would help with HIPPA compliance. (But when I look at the privacy policy I get from my Doc, there are so many people that can see my records that she could save money and just leave them on the curb.)
It's hard to picture media companies getting behind Sun, but other companies that want to build their own systems for protecting information might get on board -- Sun hopes that this will help them sell storage systems.
I live in a twisty little valley, so my Verizon phone works everywhere I go but home. It wouldn't be worth it for Verizon to put a tower in to serve my valley, because only a handful of people live there.
European phone carriers have researched the possibility of a dual mode phone that switch from GSM to DECT (digital cordless) depending on what's available, but they never commercialized it. I think they don't want to lose control of their network.
For people like me, the best answer would be a micro-cell that covers just a few hundred feet around my house. However, business and politics get in the way -- even if Verizon wanted to do it, I have a different carrier for my landline phone.
There are two markets for codesign: (1) very expensive speciality products that need high performance (ex. military), and (2) high-volume systems (game consoles, cell phones, network routers, set-top boxes.) In both of these scenarios, you can afford high development costs.
Codesign isn't as bad as you might thing -- you don't need to design the hardware at the gate level, but you can snap together pre-existing modules. Simulation tools let you check out the design space before you burn any silicon.
Yes, the programming involves concurrency, and that's hard, but I've got news for you -- future improvements in computer power are going to come from parallel operation -- we're not going to see big clock rate bumps in the immediate future.
One exciting frontier will be processors that hit a good balance between specialization and generalization: network processors, software radio processors, 3d accelerators, stream processors and chips like Cell and the Xbox processor will all be produced in large volumes, and people are going to program them, like it or not.
Athlon wins the prize for brute CPU power, but the real strength of PowerPC is that IBM can design custom chips based on combining PowerPC cores with additional processing elements. This technology is behind Deep Blue, Blue Gene, the PS3, and the Xbox 360.
This kind of chip is hard to program for, but can deliver unbeatable performance per dollar, square centimeter and watt when software is codesigned with the hardware.
AMD and Intel are going in this direction with dual-core, but IBM is already way ahead. For instance, BlueGene is based on a special chip that has two PowerPC cores with an incoherent cache (tricky to program but cheap and fast) and adds an enhanced vector processing unit. IBM is a leader in higher-end SoC solutions (really, anything that gets power from the wallplug instead of a battery.) Lower-power applications are using MIPS and ARM cores instead...
The Opteron, high-end cousin of the Athlon 64, is a great chip for servers. We have a Sun V40z, and the guys I work with are always amazed at how fast it is, and we've only got single core processors -- with dual cores, it'll smoke just about anything:
The gradual adoption of IE6 has made cookies less useful. The default policies are pretty tough. You'll lose a lot of cookies if you don't specify a P3P policy and you'll probably lose some if you specify an honest P3P policy: I've noticed the decline in cookie acceptance for quite a long time.
One big problem is that P3P doesn't have sufficient granularity to explain that a site's policies aren't terribly privacy violating. For instance, I work on a largely academic site that has interactive features that require user log in, valid emails, all that. We share psuedonymous logs with computer science researchers. IE6 would like our cookies if we did one or the other, but there's no way to convince it that we'll never share personally identifiable information with other parties.
If there's anything wrong with Linux, it's that Linux wants to be everything for everybody. On one hand, HP wants to run it on Superdome servers with 64 processors, but other people want to run it in little embedded boxes. Some people want to use it to run servers, while other people want to play 3-d games with the latest graphics cards.
(Sounds a bit like Windows, doesn't it?)
Solaris 10, on the other hand, is solidly aimed at server applications. You certainly can run it as a desktop, but you're not going to get the kind of hardware support that Linux has. On the other hand, a lot of hardware is junk: for instance, Apache has to disable sendfile() on Linux because sendfile() is unreliable in many configurations of Linux -- if you've got the wrong network card and you're serving out of an NFS mount, sendfile() will send corrupted data.
A vendor like Sun or Apple that can control the hardware and software stack from bottom to top can offer better quality and reliability than more open systems where you can plug any piece of trash into your system and expect it to run, sort-of.
One concern I have with Sun is that they'll be a battle between the SPARC and Opteron divisions within Sun. Right now, Sun has better control of the SPARC products, which are really solid. Sun's Opteron products give you a lot more computer for the dollar, but I suspect they'll come with more headaches. (We've had our share of headaches with a Xeon-based Sun box.)
Linux will continue to the be the OS of choice for people who want Unix on the desktop or who want to run small, informal servers. On the other hand, we've had the worst kind of stability problems with Red Hat 'Enterprise' Linux 3... We solved them by going to an (unsupported) 2.6 kernel, but we're thinking of running Solaris 10 on the next machine.
Slashdot Mirror
Although Carpel Tunnel Syndrome makes the news the most, the most common hand problems that computer users experience is tendonitis. Tendonitis is painful, but it doesn't damage nerves.
What evidence there is points to the mouse as the source of computer-related hand problems, not the keyboard. I remember hearing a lot about people getting computer-related hand problems in the early 90's, around the time Windows 3.1 came out and the mouse became mainstream.
I tend to have hand trouble in the spring when I start riding my bike, and that demands a hand position different from anything I do all winter.
I've had times when my tendonitis was so bad that I thought I'd have to change careers, but I found that push-ups were an effective treatment for me. Push-ups are great because they are a functional exercise that works the whole upper body -- they build up the big muscles in your chest as well as the little muscles in your wrist. I love lifting weights, but you can do push-ups at home without any equipment or gym memberships.
Note that every motion you do involves a complete 'chain' that leads back to your center of gravity -- if you push a key on a computer, it's only an ounce or two of force, but it's ultimately backed up by your whole mass. Your big muscles help your little muscles do their work and vice versa...
He's three and a half years old now. I started in the winter when he was about one and a half. Mostly Nintendo 64 and SNES under emulation. His favorite game is "Super Mario 64" and he always says "Wanna play mario in the castle". He's starting to understand something about the games. One day we were playing Zelda and there was a room with a crack in the wall he said, "Use a bomb here!" I've felt guilty about this, but I had a talk about this with the janitor in my building (who is as insightful about games as anyone) and his take was: "Didn't your parents play cards with you when you were a kid? Is this any different?"
Don't forget... "You make a dead man come".
(Not a G-rated comment, but neither was the song.)
There's a lot more to the genotype of a somatic cell than DNA: cells accumulate a whole bunch of 'markers' such as methylated bases and proteins stuck to the DNA, and repeat units that fall off the ends of the chromosome with every cell division.
Nature has good mechanism for making sure germ line (reproductive cells) stay in a good state, but manipulated cells never seem to be 100% right. Clones often end up with poor health and life expectancy because of this, and I'm afraid stem cell therapy will end in poor results, maybe even cancer.
It's bizzare that stem cells have become such an issue for the left and the right. I see Democrats screaming at the top of their lungs so we can have more research into medical treatments that we can't afford, while Republicans are blowing the ethical issues entirely out of proportion.
"Open DRM" at first sounds like a contradiction, yet, the modern approach in cryptographic systems is to design systems so that security depends on secret key material, not secret algorithms. It's a rule of nature that any piece of hardware that falls in the hands of the enemy will give up its secrets, and algorithm secrecy didn't stop Jon from cracking DVD encryption.
In an open DRM system, anybody could create their own DRM "universe" by generating their own set of keys to initialize the system -- this opens the possibility of using DRM to do different things than today's systems, such as protecting privacy: Sun is quite interested in providing storage records for medical records and such, and some kind of DRM would help with HIPPA compliance. (But when I look at the privacy policy I get from my Doc, there are so many people that can see my records that she could save money and just leave them on the curb.)
It's hard to picture media companies getting behind Sun, but other companies that want to build their own systems for protecting information might get on board -- Sun hopes that this will help them sell storage systems.
I live in a twisty little valley, so my Verizon phone works everywhere I go but home. It wouldn't be worth it for Verizon to put a tower in to serve my valley, because only a handful of people live there.
European phone carriers have researched the possibility of a dual mode phone that switch from GSM to DECT (digital cordless) depending on what's available, but they never commercialized it. I think they don't want to lose control of their network.
For people like me, the best answer would be a micro-cell that covers just a few hundred feet around my house. However, business and politics get in the way -- even if Verizon wanted to do it, I have a different carrier for my landline phone.
There are two markets for codesign: (1) very expensive speciality products that need high performance (ex. military), and (2) high-volume systems (game consoles, cell phones, network routers, set-top boxes.) In both of these scenarios, you can afford high development costs.
Codesign isn't as bad as you might thing -- you don't need to design the hardware at the gate level, but you can snap together pre-existing modules. Simulation tools let you check out the design space before you burn any silicon.
Yes, the programming involves concurrency, and that's hard, but I've got news for you -- future improvements in computer power are going to come from parallel operation -- we're not going to see big clock rate bumps in the immediate future.
One exciting frontier will be processors that hit a good balance between specialization and generalization: network processors, software radio processors, 3d accelerators, stream processors and chips like Cell and the Xbox processor will all be produced in large volumes, and people are going to program them, like it or not.
You'll need the dual core, then. Personally, Old English is my brand.
Athlon wins the prize for brute CPU power, but the real strength of PowerPC is that IBM can design custom chips based on combining PowerPC cores with additional processing elements. This technology is behind Deep Blue, Blue Gene, the PS3, and the Xbox 360.
This kind of chip is hard to program for, but can deliver unbeatable performance per dollar, square centimeter and watt when software is codesigned with the hardware.
AMD and Intel are going in this direction with dual-core, but IBM is already way ahead. For instance, BlueGene is based on a special chip that has two PowerPC cores with an incoherent cache (tricky to program but cheap and fast) and adds an enhanced vector processing unit. IBM is a leader in higher-end SoC solutions (really, anything that gets power from the wallplug instead of a battery.) Lower-power applications are using MIPS and ARM cores instead...
It depends on the game, but the Athlon 64 usually beats the Pentium 4.
x 57-06.html#opengl
http://www.tomshardware.com/cpu/20050627/athlon_f
The Opteron, high-end cousin of the Athlon 64, is a great chip for servers. We have a Sun V40z, and the guys I work with are always amazed at how fast it is, and we've only got single core processors -- with dual cores, it'll smoke just about anything:
http://www.sun.com/servers/entry/v40z/index.jsp
The gradual adoption of IE6 has made cookies less useful. The default policies are pretty tough. You'll lose a lot of cookies if you don't specify a P3P policy and you'll probably lose some if you specify an honest P3P policy: I've noticed the decline in cookie acceptance for quite a long time. One big problem is that P3P doesn't have sufficient granularity to explain that a site's policies aren't terribly privacy violating. For instance, I work on a largely academic site that has interactive features that require user log in, valid emails, all that. We share psuedonymous logs with computer science researchers. IE6 would like our cookies if we did one or the other, but there's no way to convince it that we'll never share personally identifiable information with other parties.
If there's anything wrong with Linux, it's that Linux wants to be everything for everybody. On one hand, HP wants to run it on Superdome servers with 64 processors, but other people want to run it in little embedded boxes. Some people want to use it to run servers, while other people want to play 3-d games with the latest graphics cards. (Sounds a bit like Windows, doesn't it?) Solaris 10, on the other hand, is solidly aimed at server applications. You certainly can run it as a desktop, but you're not going to get the kind of hardware support that Linux has. On the other hand, a lot of hardware is junk: for instance, Apache has to disable sendfile() on Linux because sendfile() is unreliable in many configurations of Linux -- if you've got the wrong network card and you're serving out of an NFS mount, sendfile() will send corrupted data. A vendor like Sun or Apple that can control the hardware and software stack from bottom to top can offer better quality and reliability than more open systems where you can plug any piece of trash into your system and expect it to run, sort-of. One concern I have with Sun is that they'll be a battle between the SPARC and Opteron divisions within Sun. Right now, Sun has better control of the SPARC products, which are really solid. Sun's Opteron products give you a lot more computer for the dollar, but I suspect they'll come with more headaches. (We've had our share of headaches with a Xeon-based Sun box.) Linux will continue to the be the OS of choice for people who want Unix on the desktop or who want to run small, informal servers. On the other hand, we've had the worst kind of stability problems with Red Hat 'Enterprise' Linux 3... We solved them by going to an (unsupported) 2.6 kernel, but we're thinking of running Solaris 10 on the next machine.
http://arxiv.org/abs/astro-ph/0502312