That's a non-argument. Common mice already have far higher resolution/smaller movement steps than any common screen resolution. Translated: you already have sub-pixel accuracy.
Far more important (for gaming, for instance) is how well the mouse tracks it's movement. This is where optical mice get their advantage: there isn't a rubber ball and dirty, unregular running sensors anymore, but direct optical sensing. Also, how many times/second (and with what delay) movement info is passed to the computer, can make a difference. This latter being a connection type/OS/driver issue.
So 'diminishing returns' certainly applies here. I like a good FPS game, and mechanical -> optical mouse makes a small, but very noticable difference for an experienced gamer. Optical mouse -> newer, better model optical mouse makes much less of a difference.
Why are there slashdotters that believe memory chips will replace hard drives on a large scale any time soon?
Ultimately, market price also results from how hard it is to produce the core technology (harddrives: mechanics, thin-layer magnetic materials, read/write head construction etc., memory chips: IC fabrication technology, advances in semiconductor technology)
So far, harddrive-based technology clearly had the upper hand in storage/price ratio. But who says that will stay the same in years to come?
Maybe next-gen harddrive technologies will prove to be very hard/expensive to produce, while increasingly dense/cheap memory chips might be relatively easy to produce. Who knows what will be easy to mass-produce in 5 or 10 years from now?
Some months ago, I donated an old videocard to an art school student, who was using them for a similar project. A bigger version had been done before. It involved something like 6 videocards stuck in an ordinary PC, and a customised Linux system to control it.
I ran into a website once, where they sell (surplus) super-strong magnets and related items. They warn that the stronger magnets should not be handled without decent gloves and eye protection. With good reason!
That got me to tear up an old hard drive, magnets used in there are likely the most powerful ones you'll find at home. They are usually placed in some metal assembly, keep that with the magnet! It helps contain the magnetic field mostly inside the assembly. If you remove it, the magnetic field spreads way further, can erase magnetic media fairly quick, and distort CRT screens (permanently, if you're not careful). Also, watch your fingers! I kid you not...
AFAIK, with the toroid-shaped magnetic field/plasma reactors, it is believed that effectiveness could be increased with scale/size. So something that researchers would like to do, is simply build LARGER reactors.
Ofcourse such high tech, power-draining facilities cost enormous amounts of money to construct/operate.
Still I think some money should be kept flowing into this research: it may be terribly expensive, but when succesful, rewards for society could be astronomical too.
There is no substitute for live music, but it isn't the only way for musicians to get rewarded:
For the biggest popstars, the number of fans outnumber the number of people that can visit a concert
Concerts don't cover all situations where one likes to listen to music
Some artists make fantastic music, but a lousy live act (and vice versa)
Some people don't WANT to visit a concert, but really appreciate the music anyway
I sometimes download different copies of a song, and compare those to find out what remix or encoding quality I like best. A pay-per-download model doesn't fit that at all.
Almost never I would buy a full CD with that song, regardless of having free evaluation available. On the other hand, I don't mind buying a full album, as long as I like that album as a whole.
IMHO a pay-by-donation model a la PayPal button on artist-supported site would suit both today's music artists and listeners quite well. The biggest problem here is the non-existence of a universal, easy to use, possibly anonymous, internet micro-payment system.
Maybe a good candidate for using a tiny ethanol-powered fuel cell? Low empty weight, direct fuel -> electricity conversion, and ethanol has fairly good energy-weight ratio (better than batteries anyway). You could even use a lightweight plastic bag as fuel tank.
Fill up just a bit for short flights, or fill up to max. for longer fun.
I don't know exactly who this Paul Graham is, or what he does for a living, but I ran into some very interesting essays written by him, on several occasions. They're listed on his homepage under "Essays" (duhhh..) - check those out
He writes the kind of insightful programmer's stuff, a la Joel on Software.
The average person treats a PC like a VCR, as an appliance.
I suppose that is true for, say, 90% (pick your number) of users? You can try to change that, or accept it.
Changing that means: educating users. For some limited groups that might work, but I'd say experience shows that for Joe average, it doesn't. Average users, for the most part, aren't gonna change their behaviour, they're just gonna keep on browsing random websites, clicking on random e-mail attachments, pop in random disks, and run random binaries.
Accepting that, means: consider a PC an appliance. Let maintenance be done in ways that are possible to do for a normal user, but might just as well be done by third parties (automatic updates is one way). So a possible solution would to be to create OS'es that make a PC behave such that it's safe to run random binaries, pop in random disks, click on random e-mail attachments, without worries about screwing up the system.
Current PC security ultimately depends on trusting the user, assuming that he/she knows what he's doing. I'd say, experience shows that for those 90% of Joe average, that trust is misplaced. For a user sitting behind a PC, it's just too easy to change essential aspects of the OS. Maybe some other security model would make more sense?
No, they don't. They just don't (and/or don't want to) understand all the inner workings of technology they use every day. That's true for computers, cars, kitchen appliances, VCR's, whatever.
So in terms of computer security, an average user behaves like a dummie. The book should have been named "Computer security for Dummies" or something like that, to appeal more to the target audience. Isn't this "... for dummies" series of books very popular anyway?
Linux nor Open Source is Microsoft's enemy. If Microsoft would jump on the Free/OS train today, make a 180 degree turn, pour billions into Free/OS projects, they'd be welcome to. Linux, Open Source would profit from that (even when huge amounts of crappy code would be poured in OS; natural selection works quite well), and it wouldn't hurt their own customers either. Customers are always willing to pay for something that suits their needs, even if it's built using freely available stuff.
But Microsoft would have to understand Open Source. Not from a marketing point of view, but in their gut, kidney's & toes. And to make a 180 degree turn, they'd have to totally change their company culture, views, and convince their own customers. And that is the hard thing.
So Microsoft's real enemy is the one within: themselves.
I think it's more on the availability of code or not
For reliable systems, you need proper testing. Having the code helps here. But for mission critical systems, you need FOOLPROOF VERIFICATION. That probably implies having access to any code used, but isn't enough. When the stakes are high, it becomes just as much a people/management issue.
If safety/health of patients depends on it, you MUST use equipment that you know to be reliable at all times. That includes hardware and software. Period.
That obviously excludes use of OS'es like Windows, Linux, or any other common, general-purpose OS. If the application isn't safety-critical, use whatever is most practical.
If I would be in a hospital and find out they use W***s or L**x for running stuff like a breathing machine or hart monitor, I'd sue them, go to the press, or both.
For applications where it matters (latest 3D games), the performance bottlenecks are CPU, memory and video card. Faster disk I/O helps load a game faster, but does nothing to make it run faster.
So, twice the cost, twice the hardware hassle (cabling, power, physical space), twice the noise, twice the power consumption, and for practical purposes less than twice the performance, twice the chance of drive failure, and when a drive fails, twice the amount of data lost.
Add to that the more difficult configuration, and extra hassle when re-installing an OS: is it worth the trouble?
For all but a few home users: probably not. Maybe that's why many PC's only have a single hdd?
Ofcourse you have...
Interesting question: could you ever be truly happy with a 'copy-cat' human-like robot (or dog, cat) as a partner/friend, that looks like, smells like, behaves like a real human?
Personally, I will always prefer the real thing, flesh and blood, but a good copy could be fun company...
proximity power is only really used in electric toothbrush rechargers
Not really, it's also used as a way to charge electric cars, and uses inductive coupling to transfer power. You can find a description and some pictures here.
There's likely other applications on the market.
And weren't there some research projects using microwaves to power high-altitude planes or low-orbit satellites?
proximity power is only really used in electric toothbrush rechargers
Not really, it's also used as a way to charge electric cars, and uses inductive coupling to transfer power. You can find a description and some pictures here.
There's likely other applications on the market.
And weren't there some research projects using microwaves to power high-altitude planes or low-orbit satellites?
Why not use lightbeams or optical fibers for interconnects?
Lightbeams can cross without disturbing each other, on/off switch times are very high (see modern telecommunication), physical distance doesn't matter much, connections can be electrically isolated, and the dataflow can't be tapped into by placing an antenna nearby.
Integrating optic/electronic conversion shouldn't be too hard, I think.
You could stack building blocks in 3 dimensions, and leave some space in between for air or cooling liquids...
Slashdot Mirror
Far more important (for gaming, for instance) is how well the mouse tracks it's movement. This is where optical mice get their advantage: there isn't a rubber ball and dirty, unregular running sensors anymore, but direct optical sensing. Also, how many times/second (and with what delay) movement info is passed to the computer, can make a difference. This latter being a connection type/OS/driver issue.
So 'diminishing returns' certainly applies here. I like a good FPS game, and mechanical -> optical mouse makes a small, but very noticable difference for an experienced gamer. Optical mouse -> newer, better model optical mouse makes much less of a difference.
Ultimately, market price also results from how hard it is to produce the core technology (harddrives: mechanics, thin-layer magnetic materials, read/write head construction etc., memory chips: IC fabrication technology, advances in semiconductor technology)
So far, harddrive-based technology clearly had the upper hand in storage/price ratio. But who says that will stay the same in years to come?
Maybe next-gen harddrive technologies will prove to be very hard/expensive to produce, while increasingly dense/cheap memory chips might be relatively easy to produce. Who knows what will be easy to mass-produce in 5 or 10 years from now?
Some months ago, I donated an old videocard to an art school student, who was using them for a similar project. A bigger version had been done before. It involved something like 6 videocards stuck in an ordinary PC, and a customised Linux system to control it.
That got me to tear up an old hard drive, magnets used in there are likely the most powerful ones you'll find at home. They are usually placed in some metal assembly, keep that with the magnet! It helps contain the magnetic field mostly inside the assembly. If you remove it, the magnetic field spreads way further, can erase magnetic media fairly quick, and distort CRT screens (permanently, if you're not careful). Also, watch your fingers! I kid you not...
It irritates me that every single technology advance appears to be used first to increase the "faster, bigger", and make ergonomics acceptible later.
Why not use such advances to make current hard drives more silent & cool running, and then ramp up the speed?
Ofcourse such high tech, power-draining facilities cost enormous amounts of money to construct/operate.
Still I think some money should be kept flowing into this research: it may be terribly expensive, but when succesful, rewards for society could be astronomical too.
Linus accepting a job at SCO: 1,000,000:1
- For the biggest popstars, the number of fans outnumber the number of people that can visit a concert
- Concerts don't cover all situations where one likes to listen to music
- Some artists make fantastic music, but a lousy live act (and vice versa)
- Some people don't WANT to visit a concert, but really appreciate the music anyway
I sometimes download different copies of a song, and compare those to find out what remix or encoding quality I like best. A pay-per-download model doesn't fit that at all.Almost never I would buy a full CD with that song, regardless of having free evaluation available. On the other hand, I don't mind buying a full album, as long as I like that album as a whole.
IMHO a pay-by-donation model a la PayPal button on artist-supported site would suit both today's music artists and listeners quite well. The biggest problem here is the non-existence of a universal, easy to use, possibly anonymous, internet micro-payment system.
Please rename to "Greed". That seems more appropriate for many of the stories covered.
Fill up just a bit for short flights, or fill up to max. for longer fun.
He writes the kind of insightful programmer's stuff, a la Joel on Software.
I suppose that is true for, say, 90% (pick your number) of users? You can try to change that, or accept it.
Changing that means: educating users. For some limited groups that might work, but I'd say experience shows that for Joe average, it doesn't. Average users, for the most part, aren't gonna change their behaviour, they're just gonna keep on browsing random websites, clicking on random e-mail attachments, pop in random disks, and run random binaries.
Accepting that, means: consider a PC an appliance. Let maintenance be done in ways that are possible to do for a normal user, but might just as well be done by third parties (automatic updates is one way). So a possible solution would to be to create OS'es that make a PC behave such that it's safe to run random binaries, pop in random disks, click on random e-mail attachments, without worries about screwing up the system.
Current PC security ultimately depends on trusting the user, assuming that he/she knows what he's doing. I'd say, experience shows that for those 90% of Joe average, that trust is misplaced. For a user sitting behind a PC, it's just too easy to change essential aspects of the OS. Maybe some other security model would make more sense?
No, they don't. They just don't (and/or don't want to) understand all the inner workings of technology they use every day. That's true for computers, cars, kitchen appliances, VCR's, whatever.
So in terms of computer security, an average user behaves like a dummie. The book should have been named "Computer security for Dummies" or something like that, to appeal more to the target audience. Isn't this "... for dummies" series of books very popular anyway?
But Microsoft would have to understand Open Source. Not from a marketing point of view, but in their gut, kidney's & toes. And to make a 180 degree turn, they'd have to totally change their company culture, views, and convince their own customers. And that is the hard thing.
So Microsoft's real enemy is the one within: themselves.
For reliable systems, you need proper testing. Having the code helps here. But for mission critical systems, you need FOOLPROOF VERIFICATION. That probably implies having access to any code used, but isn't enough. When the stakes are high, it becomes just as much a people/management issue.
That obviously excludes use of OS'es like Windows, Linux, or any other common, general-purpose OS. If the application isn't safety-critical, use whatever is most practical.
If I would be in a hospital and find out they use W***s or L**x for running stuff like a breathing machine or hart monitor, I'd sue them, go to the press, or both.
So, twice the cost, twice the hardware hassle (cabling, power, physical space), twice the noise, twice the power consumption, and for practical purposes less than twice the performance, twice the chance of drive failure, and when a drive fails, twice the amount of data lost.
Add to that the more difficult configuration, and extra hassle when re-installing an OS: is it worth the trouble?
For all but a few home users: probably not. Maybe that's why many PC's only have a single hdd?
Interesting question: could you ever be truly happy with a 'copy-cat' human-like robot (or dog, cat) as a partner/friend, that looks like, smells like, behaves like a real human?
Personally, I will always prefer the real thing, flesh and blood, but a good copy could be fun company...
Not really, it's also used as a way to charge electric cars, and uses inductive coupling to transfer power. You can find a description and some pictures here. There's likely other applications on the market.
And weren't there some research projects using microwaves to power high-altitude planes or low-orbit satellites?
Not really, it's also used as a way to charge electric cars, and uses inductive coupling to transfer power. You can find a description and some pictures here. There's likely other applications on the market.
And weren't there some research projects using microwaves to power high-altitude planes or low-orbit satellites?
Lightbeams can cross without disturbing each other, on/off switch times are very high (see modern telecommunication), physical distance doesn't matter much, connections can be electrically isolated, and the dataflow can't be tapped into by placing an antenna nearby.
Integrating optic/electronic conversion shouldn't be too hard, I think.
You could stack building blocks in 3 dimensions, and leave some space in between for air or cooling liquids...
If you're into language/OS design, then go RTFA. It gives a good view of the why/how to do things, and some historic perspective.
This Stephen Galton isn't a lawyer himself, by accident?
You'all might not want to skip this site: bluejackQ.com