"...and all you need to do is move in"... and then freeze your ass off out in orbit when they won't let you land, you obsolete old biological. Next thing you know, they'll be winging their way back to Earth to set us up the bomb and declare that all our base are belong to them.
HAL-15, desktop model (the one NASA is testing)
http://www.starbridgesystems.com/prod-hal1.html
HA-300, the rack-mounted, 12.8 TeraOp version
http://www.starbridgesystems.com/prod-hal3.html
The Star Bridge website seems strangely non-Slashdotted, considering how much trouble I had getting the NASA sites to load.
When I saw this one, I was sure it had to be an early April Fool's joke, but it looks like they're for real. The company's hype still sounds pretty pie in the sky, but if they can deliver even 10% of what they're promising, a hell of a lot of computational power could be available in a few years.
They cite cost savings in chip design (simpler, lower power, etc.) and chipfab retooling as a point in their favor (a single type of chip, customized for different applications). They cite it for speed of implementation, rather than reduced cost, but presumably that would come later. The HAL-300 is priced somewhere around $26 million, so don't bother to check E.bay for a few months yet.
Speaking as a grown-up wunderkind, the biggest thing you can do to overcome the age thing is to keep your mouth shut. Knowing when to do that is the surest sign of maturity.
I finished my doctorate at 26 and got a professional position, but had a hard time getting the respect I felt I deserved. People kept assuming that I was a grad student or a postdoc. What turned the trick is all of the peripheral bullshit that shouldn't matter, but really does.
I've got a sense of humor, and for a long time couldn't resist the opportunity to make a joke during professional conversations because I found the pomposity of other scientists so damn stifling. What I evenutally realized was that the other scientists truly didn't know as much as I did, but the "professional demeanor" made them seem more knowledgable than they were, and a large part of it was just being quiet and nodding sagely, or saying things like, "Yes, that's one of the things to consider." Admitting ignorance in pursuit of knowledge, or trying to leaven the conversation with actual interpersonal interaction is a mistake. You don't have to act like a prick, as one poster said, but adopting the professional demeanor, however irritating, is a big key. Having the patience and maturity to go that route, to be able to bite your f*cking tounge, is one of the things that separates the successful.
Of course, wearing a tie adds 5 years to your appearance (-5 for fish ties), sporting a moustache/string of pearls adds another 5 (-5 for odd piercings), good quality leather shoes adds another 5 (-5 for tennis shoes), etc., etc.
Sorry if this is a restatement of other's posts, but it is just so very, very true, I had to testify.
I was discussing SciAm's article about optical switching technology (which I read weeks ago in dead-tree format) with a friend who did his M.Sc. in laser crystalography. He pooh-poohed the bubble refraction/reflection switches, the nano-mirror switches, the delay loop/phase change switches... the only thing that will work, he insisted, was a set of specialized, pre-pumped lasing crystals that would boost signal power as they change the signal direction from one light path to another by refraction.
I observed that the crystals he was talking about would only work for one wavelength (so you couldn't stack signals) and in any case don't exist now, and won't exist for some time. "Doesn't matter," he said, "that's still the only technology that'll work."
He doesn't work on lasers and optics anymore, but if that's the kind of attitude that the telcom companies have, I'm glad there's more than one group trying to solve the problem.
I get tired of people insisting that we built robots that look and act like humans, upright posture on a bilaterally symmetric torso, with mobility provided by a precarious active-balance system built around two extensible, very power-intensive, mechanically complex struts. Four legged animals get around just fine, and they are inherintly more stable when at rest.
I accept that the sensors for light, sounds, ambient/atmospheric chemicals should be up high to get more range, so some kind of a head is useful, but a low center of gravity aids stability, reducing power requirements for balance and fine position control.
If you want to use these spherical motors for propulsion, they should be used as feet at the end of four (or more) legs. On level terrain, they can be powered so the robot rolls along on them, like castors on a chair; on rough terrain, they would be locked onto their supports and the robot would climb like a goat or an ant. Dinging the rotor out of true with an especially hard step would be a problem, but perhaps that could be corrected by changing the strength/firing pattern of the magnets.
This little robot is only big enough to roll around, carrying its own weight and not much more. Why limit it to that, though? This thing navigates with IR sensors, right? Why not scale it up to skateboard or lunch-cart sized, and have it deliver objects (forms to be signed, sandwiches, etc.) from office to office? Tell it to take 2 lefts, go 6 offices down and turn right, etc. It avoids the walls on its own, and to augment the internal navigation system, it could pick up info from IR beacons/signposts at intersections in the hallways. It could also get new instructions via the Palm's IR port.
"Rover! Take this to Bob on the fourth floor!"
"Thanks, Rover. Now go back to Helen."
Implant a receiver, have clip-on alternate sensors
on
End To Blindness?
·
· Score: 1
The end of the article mentions that the researchers are thinking about implanting a chip in the tissue of the visual cortex. Imagine, a Geordi LaForge-like sensor/visor prosthesis that can be removed, upgraded, customized, etc., because it sends its information to implanted receivers that are hard-wired to the brain. This would bypass the nerves that connect the retina to the visual cortex, possibly opening up this technology for use with a much wider range of degenerative diseases. As long as the visual cortex is intact, the technology could be applied.
Or, for the more paranoid among you, it could also be something The Man would implant into navy SEALs and Mossad agents... sure, they look like normal humans, but when they put on their special visors, they get UV, IR, RF vision, and with the built-in wireless WAN, what Sgt. Smith sees, everybody in the unit (and back at the base) sees in a little popup window.
My physics-major college roommate used fluorescing glass fibers in a project at Fermilab, and he brought some back home to play with. They were maybe 25cm (~10 inches) long, and very narrow, similar to fishing line. When exposed to light perpendicular to its length, a surprisingly bright purple light came out of both ends; the brighter the incident light, the brighter the flashlight-type dot of light that came out of the end. If you covered up part of the fiber, light still came out the end, just dimmer. The longer the fiber exposed to the light, the brighter the dot. We tried this with incandecent and fluorescent bulbs, bright sunlight, dim sunlight, candlelight, flashlights of various brightness, any light source we could lay our hands on. The only thing that changed was the intensity of the glowing dot.
A similar technique is used in light-gathering spotting sights. These are popular with bowhunters, and are essentially a rod of fluorescing plastic (~5cm long, ~0.5cm in diam) that you mount on your bow to help you sight on the target. The end is tapered and set at a right angle such that the incident light that hits the side of the rod makes the tapered little point (~0.2cm in diam) glow really brightly. Based on my rough estimates of the dimensions, I'd say that the area of the side of the rod (capturing incident light) is ~100X the size of the glowing tip. As with the narrow fibers, it works under any light level except complete blackness, and remember, under very dim conditions, even a faintly glowing dot looks bright.
I thought of glueing blocks or a frame to reduce the amount of epoxy required to make the final shape. Without actually having made one, I think the problems with flexing and stressing that you referred to would be a problem. If you go ahead and use a bunch of epoxy, then you can remove enough solidified material after the fact to reduce the weight, but leave enough to prevent the mirror from losing it's shape. The epoxy itself is its own reinforcing structure, integral with the reflective surface. It means discarding a fair amount of hardened epoxy. Ideally, you wouldn't use it in the first place, but at least you maintain the quality of your reflecting surface during casting. Having the mirror be a solid block of epoxy makes it easy to drill mounting holes for a frame to orient it, too.
I had actually thought of making several large off-center parabolic mirrors with this method, coating them so as to get a diffuse focal point and orienting them to overlap the focal points. Total reflective area would approximatly equal a much larger mirror, allowing you to create a pretty powerful solar furnace. The diffuse focal point system works nicely for solar heating applications.
The mercury mirror is pretty old news. Scientific American had an article about them a couple of years ago. One thing that was proposed for the readership, though, was to use the same idea to make big mirrors for homebrew telecopes. You place a circular container on a turntable (the kind to get sound out of those flat, black vinyl things your parents have in the basement), fill it with a slow-curing epoxy/resin mix, cover it to prevent air currents from rippling the finish, and turn the turntable on. A few hours later, you've got a parabolic dish. Pour in enough additional epoxy/resin mix, formulated to be even slower curing, will improve the quality of the surface by making a thin top coat. Silver coat the surface (or have it done professionally) and you've got as big a primary reflector as you want. SciAm reported results that were as good or better than most home-use telescopes. Place the container off center and your focal point will be off center, too, if you want to get creative.
For really big epoxy mirrors (a meter or so in diameter), I would imagine you would want to drill out some of the material from the backside, or it would get pretty heavy and cumbersome.
I've seen less extreme examples of supercooled MoBo using refrigerator coils and mineral oil. These got down to -20C or so. Is Fluorinert so much better that it justifies the staggeringly huge capital outlay? I overclock because I'm too cheap to buy a faster/newer CPU, but maybe that's just me.
I heard the NPR story. Steve Ballmer (Little Big Brother) said, (rough quote): "People are very concerned about privacy on the internet. The best way to ensure complete privacy is to have all of your information, your credit cards, your birthday, your friend's phone numbers, whatever, in one centralized location that is completely secure, so you can set the restrictions on access to your data. This is exactly what we're proposing."
He seemed so ardent and sincere, it was like he didn't think anyone could possibly question the fox's motives in volunteering to guard the henhouse (for a modest fee, of course!). What a pompous, sanctimonious jerk! A quote from Bill G. (Big Big Brother) was something like, "We're poised to take over everyone's business transactions on the internet, and to make this a part of future versions of Windows." Is there anyone, anywhere who doesn't see this as blatant disregard for the antitrust ruling that was just handed down? Maybe they think that by the time it all goes through the appeals process, they'll own the world so the future ruling won't matter... do they think the courts are cut off from the daily news, and won't take this into consideration? Just the *proposal* of.NET should result in unfavorable rulings in the appeals, even if.NET doesn't make one dime.
The depth of their delusions of godhood is just staggering... we want what's best for you, now do it our way or we'll kill you. Talk about being divorced from reality!
An aritcle in Scientific American (March 2000, sorry, not on the web) discussed GA for use in sovling complex problems like engine design, routing, etc. They pointed out that this technique yields good solutions, and tends to yield better solutions the longer you let them run. However, it isn't really intended to find the absolute optimal solution. To do that, you need a crafstman/artisan to optimize it by hand. GA may be for wimps (X-p), but it will produce good (or very good) results reasonably quickly.
Also, I agree that the product of the evolutionary process will probably yield designs (or software) that will work well, but which may be pretty incomprehensible to the people who produced it. Refer to the earlier comment by "roman_mir":
>>>At the end a computer program was generated that sorted the entire string of characters. Interestly enough, the programmer could not figure out exactly how the string was sorted, the software was just too complex to understand (I supposed he did not want to waste time trying).>>>
Each organism that biologists study incorporates (and builds on) legacy functionality from its evolutionary predecessors to solve new problems, or move into new niches. Some basic cellular functions are unchanged from bacteria to people, but obviously lots of others have changed a lot! Figuring out exactly how these critters do what they do isn't easy. Although I don't think it would be a problem with engines or other real-world structures, GA-designed software may have to be studied and analyzed like DNA from a newly identified species.
Of course, if you find/evolve a piece of code that works really well, it can be snipped up, rearranged and combined with other code to make new programs (analagous to transposons, retroviral exchange and recombination) which might work even better.
*****Microsoft's hardware division actually has to compete with other companies for business, so it has to provide products that are good enough to convince people to buy them instead of the competetion.*******
Same goes for many of the games they make. Flight Simulator, Age of Empires are quite good, because they have a lot of competition in the game market. I'm delighted that the company will be broken up, and that a thousand flowers will bloom in the OS market.
Perhaps I'm naive, but when they say that this computer is exclusively for calculating gravitational interactions, why could you not make some data substitutions and use it for different calculations?
Step 1) Acquire data on the purchasing behavior and demographic info of a couple of million consumers from some unscrupulous web retail site.
Step 2) Get a few scaling variables on the front- and back-end, replace stellar mass with income, replace stellar velocity with purchasing habits, replace stallar cluster density with population density (or proximity to retail outlets), etc., etc.
Step 3) Run the system to model consumer purchasing decisions for a product you're planning to introduce into the marketplace.
Surveys measure economic activity on a large scale and make broad predictions. Could this be used to more accurately model and predict economic behavior on a more precise scale? The data would be constantly updated, and the models would be constantly rerun to get the most accurate picture possible of how you and I will spend our $$$. Just make sure the the observed isn't aware of the observation, or your models lose their viability.
>I wonder if there is a law against things like that.
If it becomes identified as a real threat, a new law will probably be passed and the BATF would probably enforce it. Right now, though, the FCC has existing laws against devices that interfere with other devices' functioning. The laws are meant to address unitended interference, not intentional terrorist activity, but I think they would apply.
However, just because it's illegal doesn't mean it won't get built/used, especially if it's so easy to do. This actually sounds easier to make (and safer to operate) than a homemade zip gun or ammonium nitrate/fuel oil bomb. I think one of the big dangers is people using these things as a prank. I forsee the pranks costing $10,000 in damage (data loss, down time, etc.) and $100,000,000 in reactionary home Faraday cages/PC shielding for banks, hospitals, etc.
I don't think decisions to adopt or avoid Linux will be made solely on the basis of this MSNBC report, the original WSJ story, the free executive summary, or on the $995 report. My impression is that word-of-mouth experience drives Linux adoption as much as any media reports. Besides, for every report that says you can't use a Linux box to replace Big Iron, there's another one reporting some well-known.com that's adopting it for mission-critical service (Salon comes to mind). Good news offsets the bad, or at least muddies the water enough that the decicion makers will have to consider all the evidence.
I realize that I'm opening myself up to flames galore for introducing a non-tech reference here, but I seem to recall from my Liberal Arts Education (i.e. misspent youth) that prior to Europe's Age of Exploration in the 15th-17th centuries, the Chinese empire used gift giving as a way to impress and intimidate their neighbors. "We are so powerful and impressive," the message went, "that we can give you lavish gifts and not think anything of it." After Marco Polo and other explorers increased the competetion in the region for trade and prestige, the gifts had to get more and more lavish, until China couldn't keep it up. They then withdrew and wanted to have nothing to do with anyone else, writing the rest of the world off as illiterate barbarians who wouldn't know a great culture if it bit them in the ass.
When the prime desirable commodity for hackers and workers in Open Source community is the respect and admiration of the rest of the community, what will happen when the community gets a) much larger or b) even harder to impress? With more and more "illiterate barbarians" getting involved (or at least hanging around), people who wouldn't recognize clever and beautiful code if you showed it to them and wouldn't be impressed anyway, how does one get the prestige needed to make writing Open Source worthwhile?
One way is just to reduce the effective size of the community again. So, will the Open Source community continue to fracture along lines of skill level? Wizards who only write for, and value the opinion of, other Wizards, followed by Apprentice Wizards, who will explain stuff to the Wizard wanna-be's, who will, in turn explain stuff to Anonymous Cowards who will lord their wisdom over newbies like me?
With the increased press coverage, a lot of people are thinking about Linux. How many of them will now say, "I'll wait until MS Office is ported, then I'll give Linux a try."? Personally, I think MS will port Office to OS/2 before they port it to Linux, and it'll be a cold day in hell in either case.
Slashdot Mirror
"...and all you need to do is move in" ... and then freeze your ass off out in orbit when they won't let you land, you obsolete old biological. Next thing you know, they'll be winging their way back to Earth to set us up the bomb and declare that all our base are belong to them.
... and pictures, too.
HAL-15, desktop model (the one NASA is testing)
http://www.starbridgesystems.com/prod-hal1.html
HA-300, the rack-mounted, 12.8 TeraOp version
http://www.starbridgesystems.com/prod-hal3.html
The Star Bridge website seems strangely non-Slashdotted, considering how much trouble I had getting the NASA sites to load.
When I saw this one, I was sure it had to be an early April Fool's joke, but it looks like they're for real. The company's hype still sounds pretty pie in the sky, but if they can deliver even 10% of what they're promising, a hell of a lot of computational power could be available in a few years.
They cite cost savings in chip design (simpler, lower power, etc.) and chipfab retooling as a point in their favor (a single type of chip, customized for different applications). They cite it for speed of implementation, rather than reduced cost, but presumably that would come later. The HAL-300 is priced somewhere around $26 million, so don't bother to check E.bay for a few months yet.
Speaking as a grown-up wunderkind, the biggest thing you can do to overcome the age thing is to keep your mouth shut. Knowing when to do that is the surest sign of maturity.
I finished my doctorate at 26 and got a professional position, but had a hard time getting the respect I felt I deserved. People kept assuming that I was a grad student or a postdoc. What turned the trick is all of the peripheral bullshit that shouldn't matter, but really does.
I've got a sense of humor, and for a long time couldn't resist the opportunity to make a joke during professional conversations because I found the pomposity of other scientists so damn stifling. What I evenutally realized was that the other scientists truly didn't know as much as I did, but the "professional demeanor" made them seem more knowledgable than they were, and a large part of it was just being quiet and nodding sagely, or saying things like, "Yes, that's one of the things to consider." Admitting ignorance in pursuit of knowledge, or trying to leaven the conversation with actual interpersonal interaction is a mistake. You don't have to act like a prick, as one poster said, but adopting the professional demeanor, however irritating, is a big key. Having the patience and maturity to go that route, to be able to bite your f*cking tounge, is one of the things that separates the successful.
Of course, wearing a tie adds 5 years to your appearance (-5 for fish ties), sporting a moustache/string of pearls adds another 5 (-5 for odd piercings), good quality leather shoes adds another 5 (-5 for tennis shoes), etc., etc.
Sorry if this is a restatement of other's posts, but it is just so very, very true, I had to testify.
... all the world looks like a nail.
I was discussing SciAm's article about optical switching technology (which I read weeks ago in dead-tree format) with a friend who did his M.Sc. in laser crystalography. He pooh-poohed the bubble refraction/reflection switches, the nano-mirror switches, the delay loop/phase change switches... the only thing that will work, he insisted, was a set of specialized, pre-pumped lasing crystals that would boost signal power as they change the signal direction from one light path to another by refraction.
I observed that the crystals he was talking about would only work for one wavelength (so you couldn't stack signals) and in any case don't exist now, and won't exist for some time. "Doesn't matter," he said, "that's still the only technology that'll work."
He doesn't work on lasers and optics anymore, but if that's the kind of attitude that the telcom companies have, I'm glad there's more than one group trying to solve the problem.
I get tired of people insisting that we built robots that look and act like humans, upright posture on a bilaterally symmetric torso, with mobility provided by a precarious active-balance system built around two extensible, very power-intensive, mechanically complex struts. Four legged animals get around just fine, and they are inherintly more stable when at rest.
I accept that the sensors for light, sounds, ambient/atmospheric chemicals should be up high to get more range, so some kind of a head is useful, but a low center of gravity aids stability, reducing power requirements for balance and fine position control.
If you want to use these spherical motors for propulsion, they should be used as feet at the end of four (or more) legs. On level terrain, they can be powered so the robot rolls along on them, like castors on a chair; on rough terrain, they would be locked onto their supports and the robot would climb like a goat or an ant. Dinging the rotor out of true with an especially hard step would be a problem, but perhaps that could be corrected by changing the strength/firing pattern of the magnets.
This little robot is only big enough to roll around, carrying its own weight and not much more. Why limit it to that, though? This thing navigates with IR sensors, right? Why not scale it up to skateboard or lunch-cart sized, and have it deliver objects (forms to be signed, sandwiches, etc.) from office to office? Tell it to take 2 lefts, go 6 offices down and turn right, etc. It avoids the walls on its own, and to augment the internal navigation system, it could pick up info from IR beacons/signposts at intersections in the hallways. It could also get new instructions via the Palm's IR port.
"Rover! Take this to Bob on the fourth floor!"
"Thanks, Rover. Now go back to Helen."
The end of the article mentions that the researchers are thinking about implanting a chip in the tissue of the visual cortex. Imagine, a Geordi LaForge-like sensor/visor prosthesis that can be removed, upgraded, customized, etc., because it sends its information to implanted receivers that are hard-wired to the brain. This would bypass the nerves that connect the retina to the visual cortex, possibly opening up this technology for use with a much wider range of degenerative diseases. As long as the visual cortex is intact, the technology could be applied.
Or, for the more paranoid among you, it could also be something The Man would implant into navy SEALs and Mossad agents... sure, they look like normal humans, but when they put on their special visors, they get UV, IR, RF vision, and with the built-in wireless WAN, what Sgt. Smith sees, everybody in the unit (and back at the base) sees in a little popup window.
My physics-major college roommate used fluorescing glass fibers in a project at Fermilab, and he brought some back home to play with. They were maybe 25cm (~10 inches) long, and very narrow, similar to fishing line. When exposed to light perpendicular to its length, a surprisingly bright purple light came out of both ends; the brighter the incident light, the brighter the flashlight-type dot of light that came out of the end. If you covered up part of the fiber, light still came out the end, just dimmer. The longer the fiber exposed to the light, the brighter the dot. We tried this with incandecent and fluorescent bulbs, bright sunlight, dim sunlight, candlelight, flashlights of various brightness, any light source we could lay our hands on. The only thing that changed was the intensity of the glowing dot.
A similar technique is used in light-gathering spotting sights. These are popular with bowhunters, and are essentially a rod of fluorescing plastic (~5cm long, ~0.5cm in diam) that you mount on your bow to help you sight on the target. The end is tapered and set at a right angle such that the incident light that hits the side of the rod makes the tapered little point (~0.2cm in diam) glow really brightly. Based on my rough estimates of the dimensions, I'd say that the area of the side of the rod (capturing incident light) is ~100X the size of the glowing tip. As with the narrow fibers, it works under any light level except complete blackness, and remember, under very dim conditions, even a faintly glowing dot looks bright.
I thought of glueing blocks or a frame to reduce the amount of epoxy required to make the final shape. Without actually having made one, I think the problems with flexing and stressing that you referred to would be a problem. If you go ahead and use a bunch of epoxy, then you can remove enough solidified material after the fact to reduce the weight, but leave enough to prevent the mirror from losing it's shape. The epoxy itself is its own reinforcing structure, integral with the reflective surface. It means discarding a fair amount of hardened epoxy. Ideally, you wouldn't use it in the first place, but at least you maintain the quality of your reflecting surface during casting. Having the mirror be a solid block of epoxy makes it easy to drill mounting holes for a frame to orient it, too.
I had actually thought of making several large off-center parabolic mirrors with this method, coating them so as to get a diffuse focal point and orienting them to overlap the focal points. Total reflective area would approximatly equal a much larger mirror, allowing you to create a pretty powerful solar furnace. The diffuse focal point system works nicely for solar heating applications.
The mercury mirror is pretty old news. Scientific American had an article about them a couple of years ago. One thing that was proposed for the readership, though, was to use the same idea to make big mirrors for homebrew telecopes. You place a circular container on a turntable (the kind to get sound out of those flat, black vinyl things your parents have in the basement), fill it with a slow-curing epoxy/resin mix, cover it to prevent air currents from rippling the finish, and turn the turntable on. A few hours later, you've got a parabolic dish. Pour in enough additional epoxy/resin mix, formulated to be even slower curing, will improve the quality of the surface by making a thin top coat. Silver coat the surface (or have it done professionally) and you've got as big a primary reflector as you want. SciAm reported results that were as good or better than most home-use telescopes. Place the container off center and your focal point will be off center, too, if you want to get creative.
For really big epoxy mirrors (a meter or so in diameter), I would imagine you would want to drill out some of the material from the backside, or it would get pretty heavy and cumbersome.
I've seen less extreme examples of supercooled MoBo using refrigerator coils and mineral oil. These got down to -20C or so. Is Fluorinert so much better that it justifies the staggeringly huge capital outlay? I overclock because I'm too cheap to buy a faster/newer CPU, but maybe that's just me.
I heard the NPR story. Steve Ballmer (Little Big Brother) said, (rough quote): "People are very concerned about privacy on the internet. The best way to ensure complete privacy is to have all of your information, your credit cards, your birthday, your friend's phone numbers, whatever, in one centralized location that is completely secure, so you can set the restrictions on access to your data. This is exactly what we're proposing."
.NET should result in unfavorable rulings in the appeals, even if .NET doesn't make one dime.
He seemed so ardent and sincere, it was like he didn't think anyone could possibly question the fox's motives in volunteering to guard the henhouse (for a modest fee, of course!). What a pompous, sanctimonious jerk! A quote from Bill G. (Big Big Brother) was something like, "We're poised to take over everyone's business transactions on the internet, and to make this a part of future versions of Windows." Is there anyone, anywhere who doesn't see this as blatant disregard for the antitrust ruling that was just handed down? Maybe they think that by the time it all goes through the appeals process, they'll own the world so the future ruling won't matter... do they think the courts are cut off from the daily news, and won't take this into consideration? Just the *proposal* of
The depth of their delusions of godhood is just staggering... we want what's best for you, now do it our way or we'll kill you. Talk about being divorced from reality!
An aritcle in Scientific American (March 2000, sorry, not on the web) discussed GA for use in sovling complex problems like engine design, routing, etc. They pointed out that this technique yields good solutions, and tends to yield better solutions the longer you let them run. However, it isn't really intended to find the absolute optimal solution. To do that, you need a crafstman/artisan to optimize it by hand. GA may be for wimps (X-p), but it will produce good (or very good) results reasonably quickly.
Also, I agree that the product of the evolutionary process will probably yield designs (or software) that will work well, but which may be pretty incomprehensible to the people who produced it. Refer to the earlier comment by "roman_mir":
>>>At the end a computer program was generated that sorted the entire string of characters. Interestly enough, the programmer could not figure out exactly how the string was sorted, the software was just too complex to understand (I supposed he did not want to waste time trying).>>>
Each organism that biologists study incorporates (and builds on) legacy functionality from its evolutionary predecessors to solve new problems, or move into new niches. Some basic cellular functions are unchanged from bacteria to people, but obviously lots of others have changed a lot! Figuring out exactly how these critters do what they do isn't easy. Although I don't think it would be a problem with engines or other real-world structures, GA-designed software may have to be studied and analyzed like DNA from a newly identified species.
Of course, if you find/evolve a piece of code that works really well, it can be snipped up, rearranged and combined with other code to make new programs (analagous to transposons, retroviral exchange and recombination) which might work even better.
*****Microsoft's
hardware division actually has to compete with other companies for business, so it has to provide products that are good enough to convince
people to buy them instead of the competetion.*******
Same goes for many of the games they make. Flight Simulator, Age of Empires are quite good, because they have a lot of competition in the game market. I'm delighted that the company will be broken up, and that a thousand flowers will bloom in the OS market.
Perhaps I'm naive, but when they say that this computer is exclusively for calculating gravitational interactions, why could you not make some data substitutions and use it for different calculations?
Step 1) Acquire data on the purchasing behavior and demographic info of a couple of million consumers from some unscrupulous web retail site.
Step 2) Get a few scaling variables on the front- and back-end, replace stellar mass with income, replace stellar velocity with purchasing habits, replace stallar cluster density with population density (or proximity to retail outlets), etc., etc.
Step 3) Run the system to model consumer purchasing decisions for a product you're planning to introduce into the marketplace.
Surveys measure economic activity on a large scale and make broad predictions. Could this be used to more accurately model and predict economic behavior on a more precise scale? The data would be constantly updated, and the models would be constantly rerun to get the most accurate picture possible of how you and I will spend our $$$. Just make sure the the observed isn't aware of the observation, or your models lose their viability.
>I wonder if there is a law against things like that.
If it becomes identified as a real threat, a new law will probably be passed and the BATF would probably enforce it. Right now, though, the FCC has existing laws against devices that interfere with other devices' functioning. The laws are meant to address unitended interference, not intentional terrorist activity, but I think they would apply.
However, just because it's illegal doesn't mean it won't get built/used, especially if it's so easy to do. This actually sounds easier to make (and safer to operate) than a homemade zip gun or ammonium nitrate/fuel oil bomb. I think one of the big dangers is people using these things as a prank. I forsee the pranks costing $10,000 in damage (data loss, down time, etc.) and $100,000,000 in reactionary home Faraday cages/PC shielding for banks, hospitals, etc.
Gray lady? Isn't that the NY Times?
.com that's adopting it for mission-critical service (Salon comes to mind). Good news offsets the bad, or at least muddies the water enough that the decicion makers will have to consider all the evidence.
I don't think decisions to adopt or avoid Linux will be made solely on the basis of this MSNBC report, the original WSJ story, the free executive summary, or on the $995 report. My impression is that word-of-mouth experience drives Linux adoption as much as any media reports. Besides, for every report that says you can't use a Linux box to replace Big Iron, there's another one reporting some well-known
I realize that I'm opening myself up to flames galore for introducing a non-tech reference here, but I seem to recall from my Liberal Arts Education (i.e. misspent youth) that prior to Europe's Age of Exploration in the 15th-17th centuries, the Chinese empire used gift giving as a way to impress and intimidate their neighbors. "We are so powerful and impressive," the message went, "that we can give you lavish gifts and not think anything of it." After Marco Polo and other explorers increased the competetion in the region for trade and prestige, the gifts had to get more and more lavish, until China couldn't keep it up. They then withdrew and wanted to have nothing to do with anyone else, writing the rest of the world off as illiterate barbarians who wouldn't know a great culture if it bit them in the ass.
When the prime desirable commodity for hackers and workers in Open Source community is the respect and admiration of the rest of the community, what will happen when the community gets a) much larger or b) even harder to impress? With more and more "illiterate barbarians" getting involved (or at least hanging around), people who wouldn't recognize clever and beautiful code if you showed it to them and wouldn't be impressed anyway, how does one get the prestige needed to make writing Open Source worthwhile?
One way is just to reduce the effective size of the community again. So, will the Open Source community continue to fracture along lines of skill level? Wizards who only write for, and value the opinion of, other Wizards, followed by Apprentice Wizards, who will explain stuff to the Wizard wanna-be's, who will, in turn explain stuff to Anonymous Cowards who will lord their wisdom over newbies like me?
With the increased press coverage, a lot of people are thinking about Linux. How many of them will now say, "I'll wait until MS Office is ported, then I'll give Linux a try."? Personally, I think MS will port Office to OS/2 before they port it to Linux, and it'll be a cold day in hell in either case.