Will this look any better than a movie broadcast in 1080i and played back on an HDTV with 3:2 pulldown detection? I was under the impression that the pull-up done to convert 24fps film into interlaced video is completely reversible, and so 1080 progressive quality is already available through 1080i broadcasts.
It will be nice to have discs of HD content eventually, but I don't see what is so impressive that makes this worthy of coverage.
No no, you've got it all wrong. How many of those 120,000 will have brown skin and a beard? That's the dirty little secret that'll make this system viable.
Not exactly. It's kind of like an algorithm, you can rewrite it under some circumstances using different constructs. You can't do this for more than a few in a copyrighted cookbook, or it is a violation.
A buyers guide without comparisons isn't really all that useful. I'd like to know which machines have relative top quality and take separate cartridges for each color.
Well, if your requirements aren't too beefy (technical term) try labjack. They have kernel drivers for the cheaper model ($119) and the pricier one ($399) uses ethernet and works over TCP/IP. There are links to Perl/Java/.NET/MATLAB interfaces, among others, on their site. The cheaper U12 model support a dozen analog inputs, 20 digital I/O, and 2 analog out. The pricier model has PWM output and support for various timers and whatnot.
This switch back to KDE doesn't quite resolve everything. From the slashdot story Suse Linux Founder Exits Novell, it's clear that there has been a big internal fight over the changes happening at Novell. This comment under the above story sheds some light on just what.
It looks like Novell is being pushed to make some bad moves by a major investor or two, with lots of R&D layoffs and pressure to sell off some of the technology that strongly identifies Novell with quality. Rings bells of the sort of thing that happend at HP, with all the PHB talk of changing Novell's "vision," but with little articulated beyond "Linux" as a strategy. This is just a wee bit broad and smacks of a solution looking for a problem.
That's the card I used, it states a PIII-750 as the minimum requirement, and that's using the included software, which uses ATI Radeon hardware for accelerated decoding. It claims to work un-accelerated on a 1.6Ghz P4.
Also important is the memory bandwidth of your card - I've heard 10gb/s is required for full HD decoding, which I assume is 1080p. To figure out how much you have, plug and chug with this formula -
(buswidth/8 * memclk) (Double the result for DDR memory)
For a 9600 with a 128 bit wide bus, running DDR400 128/8 * 400 Mhz = 6.4GB/s
Well the machine I was talking about barely handles 1080i using the card spec'ed above, and stutters if you try and do anything. But it works.
Now I doubt this setup would work with MCE. I've read online that some people who run smoothly with Fusion's software get stuttering in MCE, so it may just be that Fusion has a more efficient decoder implementation. MCE uses third party DVD decoders like PowerDVD and Nvidia's decoder to do decoding. They may take advantage of your video card, but seem to be less efficient at utilizing video memory bandwidth. A 9600XT with 600Mhz of bandwidth would likely help that.
Case in point - I've watched OTA HDTV in flawless quality on a PIII-800, using an ATI 9600 series card to accelerate the decoding to a 1920x1200 LCD panel. Even baseline machines will be including powerful graphics coprocessors come the Vista generation.
Also, consider the architecture of a processor like the SONY/IBM cell. The DSP (digital signal processing) features that it provides will blaze through HDTV streams with far fewer Mhz than than a regular PC. Powerful DSPs can not only decode in realtime but can encode and transcode to other formats in real time as well, with far more efficiency than a regular PC. I believe SONY ran a demo where they were decoding something like 48 HDTV streams simultaneously using the PS3 version of the CELL. I expect this sort of tech to make it's way into regular PCs in coming years.
Speed really shouldn't be all that relevant in the long term, as processors are evolving to handle media intensive tasks with much greater efficiency.
This bill is meant to mandate measures with similar ends to those in that article. Right now the means exist but aren't required by law, and aren't implemented widely at all.
Out of all the counties that used e-voting last election, what percentage of those counties implemented systems with measures like those that article? Very, very few. Is that likely to change without legislation? No.
My point is to encourage others to request support for sane mandantory standard. It's facing considerable opposition and needs to build support now, not at the last minute, just before it hits the floor. If it fails to pass, half the people here will be outraged and wonder why. Not to mention we'll be lamenting all over again after the 2008 elections, wondering if there was any tampering in light of circumstantial evidence in key counties.
The success of an e-vote is hard to verify, and a poorly designed system - like many of those used in the US - makes it fairly trivial to alter even a presidential election through tampering. With no far reaching conspiracies required either, just a few key municipalities in Ohio would need to manipulated. This would be ridiculously easy for a few corrupt local election officials, who through diebold's interface can alter tallies without an audit log. This is a built in feature for making "corrections" and incorporating things like absentee votes of course. There is so much reward involved that the potential abuse here is astronomical.
Sound ridiculous?
Yeah, I know. So while your at it, please check out this bill and write your representatives about it. Some republicans are already poo-pooing these much needed reforms and they need more momentum.
...TCPA type security as well? There probably are some useful aspects to Trusted Computing. I'd imagine it would make it much harder for people to cheat in online video games, or to spoof identities.
It's a mess, craigslist wasn't really designed for this. You should be able to register yourself and your family, and others searching for you should be able to receive updates on matches as more people register.
This would be far simpler to use and more useful than craigslist. All one would need is a former address, employer, or family name. More criteria could be added later.
I bet those who are worried about their friends and family would be pretty happy to have this.
How about some slashdotters set up a database driven site where people can register to be found and find others? They could list their employer, address, any significant information. I don't have the resources to do this but would be glad to help in throwing together some php and sql if given some server space.
I was there yesterday and quite a few devices were sitting above the sliding doors and in a row along the ceiling as you came into the terminal, and they were oval shaped. They rotated on a platform and spun on a spindle, giving them 360 degrees of freedom. Each white oval was maybe 1.5 feet by 1 foot in diameter and they seemed to follow and track things, mobilizing suddenly at times, but remaining in default position most of the time.
The thing is the each egg shaped "camera" seemed to point with either a lens on one end the oval or a square shaped opening on the opposite side. The square shaped side I imagine has some other sort of detection ability. They looked big and expensive, and I was kind of curious what sort of tech goes into these.
Is anyone on slashdot working on these sorts of applications? Maybe someone could shed some light on what sort of sensory abilities these things have?
They could always pay to have it fixed. The author says much of the code is secure, so why not take undertake a massive effort to overhaul the suspect portions, and then offer a $75 cash incentive for each router a tech patches or a substantial discount for a replacement router? They do have serial #'s so patching could be tracked, perhaps they could even use some relatively inexpensive hardware or software verification module. It could generate a code to verify proper patch status, or even incorporate patching functions in this simple device.
This might hurt business less in the long run than a widespread, debilitating breakdown. It will be expensive, probably ~$120 a pop in the end, considering payout, as well as the cost of verification hardware/software devlopment and production, but they'll reduce the destruction for their customer's businesses and to their own image.
I don't know just how much this would cut into Cisco's revenues, which would of course reduce short term profits and thus investment interest. Someone up there should be weighing something like this though, however painful it sounds. It would also set Cisco apart in market where cheaper competitors are taking away Cisco's profits. How many of them would go to such lengths in the event of a vulnerability? Companies love insuring themselves against everything.
Agreed. For some tasks, better sensory input will be necessary, skin sensations being a particulary tricky one.
Still though, audio and video recognition should make robots far more human than they have been, and I suspect this will take much of the public by surprise. The environments they operate in will no longer have to be so consistent, and the consequence will be that they will appear far less deterministic and mechanical.
We should find hacks around not having skin sensations, at least for some applications. For example, a human might cook a steak by checking grill surface temp with the back of his hand, but a robot could use an infrared pyrometer. It could also estimate the doneness of a steak not by feel, as we might by pressing to check the elasticity of the meat, but by measuring it's initial temperature and thickness, as doneness can be calculated as a factor of these two plus cooking time.
Some problems will still elude, but I suspect we can make the tech adapt using atypical methods, not unlike how a blind person uses alternate senses to do tasks that we normally associate with vision.
This is the parent poster. I'll post this anonymous to avoid karma whoring, but I wanted to expand on what I said, because most readers probably won't undestand the significance of DSPs.
DSPs are good at tasks like pattern matching, filtering out noise, finding statistical correlations, inferring probabilities, and simulating neural networks - among other things. These sorts of tasks are can be done by traditional processors, but such processors aren't designed for this. Something a cheap DSP might handle can hog inordinate amounts of CPU time on a pricey general purpose processor, because architectures like x86 and ppc weren't designed for heavy vector processing. It takes lots of clock cycles on these processors to simulate with software what a DSP does in hardware.
This page lists many DSP capabilities to give you a better idea.
A bipedal robot with stereoscopic vision and so many degrees of freedom could potentially perform some complex tasks that traditional robots have been laughed at for trying.
Robots are often clumsy, and a big reason for this is inadequate sensory ability. Robots are already good at responding very precisely with their limbs, because they have a computer model of how their limbs react and can sense how they are presently positioned, at least with respect to the robot itself. But clumsiness becomes a factor when limbs must be positioned with respect to outside objects - this requires recognition.
The big deal is that processor technology is getting to the point where a robot can analyze high bandwidth sources like video in real-time, for use as feedback for its limbs, or whatever else it may control. With stereoscopic vision the robot could constantly refactor it's movement based on analyses of video, so it could say, go into the garage and roll out the garbage while avoiding obstacles. Or how about the watering plants? Or returning scattered items to their pre-defined places? Wiping down the kitchen counter? Doing laundry? Robots will play a big role in our future, for better or worse. The market will find them hard to resist once they actually become useful.
Fast digital signal processors will play a role in this, not unlike the 7 SPEs found in the Sony/IBM CELL processor. Fast, commoditized (read: cheap) DSPs for this sort of application haven't been available, and that will soon be changing. I suspect PS3 games will use those vector processing units for responding to and learning from high bandwidth inputs like audio, video, or even EEG data to sense states of mind. I mean they called the thing the CELL after all. Likely because of it's seemingly organic response to stimuli.
Firefox is getting to the point that even though it is still first generation browser, administrators of large networks like those at universities are taking note of it's speed, sturdiness, and ability to resist malware.
People are sick of malware, and their authors are increasingly using search engines to cast a wide net. It is not hard to get snagged while searching for popular subjects like mp3s or porn - many of the sites that manage to get their rankings up on these topics use activex installers for malware, or even worse, IE exploits, on their pages. These exploits sometimes install apps without any explicit user notification, though sometimes excess hard drive activity can be noticed while these things are sneaking in.
Even users with machines that are patched against exploits often let activex installers for malware in. This is because IE requests permission to run privileged code through a popup dialogue that demands a response. The problem is that sites often present pop up dialogues for more mundane things, like disclaimers and verification of site-specific information. Clicking no or cancel usually results in something not working, and users have become habituated to clicking yes. So why ask an unfamiliar security question using the same mechanism? The answer is poor identification with users, and an inconsiderate eagerness to "innovate". Both must go hand in hand though.
Firefox doesn't support ActiveX out of the box, but even once it is added, privileged code requests are handled differently. A simple notification that a control on the page doesn't have security permission to execute is overlayed at the top of the web site, and can be safely ignored. The simple message and consistency of purpose in the format lets users easily infer that when they are granting serious privileges. Popups for less privileged contexts, ignorable notifications that default to saftey for critical ones - how clever and innovative! Even better, it is rare that the advertised purpose of a site spreading malware is hindered by ignoring Firefox's notification message, though that may change as it picks up steam. Still though, users will be more aware of when they are granting wider authority.
The tough part with increasing adoption though, is that most users today don't make the browser-malware inference; identifying that the malware that keeps slowing and crashing their machines often stems from a simple page visit and/or dialogue box response. They look towards more overtly dangerous things, like explicit downloads via email attachment or website.
Firefox's notification model and its pending adoption by numerous sysadmins will change that. A critical mass of users will be forced to acknowledge the changes, and so should be more inclined to adopt them at home - through increased familiarity, and by authoritative association of that other browser with the familiar and frustrating ailment of malware.
IE 7.5 will then be forced to acknowledge the state of user experience before offering its own XPerience. Either that or MS will just mimick what it can of the Firefox recipe and rely on it's monopoly to propogate. I wonder which.
Why not allow unlimited TLDs? Official structure and organization through TLDs is overrated.
For example, it seems silly to rely on something like.mobi for mobile sites rather than embedding that in the application protocol. We should be relying on meta-data to define such distinctions, including the categorization companies, non-profit organizations and ISPs. Meta-data would be more flexible, as distinctions do overlap in ways that domains can't singularly cover.
If you need something authoritative, private authorities could use public/private keys as proof to do that. Indiviuals could then decide which private authorities have standards worth trusting. The U.N. could set up such an authority to authenticate government sites. When a user visits a government site, it could refer the application to the whichever authorities it chooses.
Limiting TLDs just creates conflict as different powerful interests vie for their own distinctions. Sure people can more quickly categorize this way, but the limitations seem to outweight the benefits.
With all the tech people here, it'd be neat to figure out the earliest sources of mp3 trading. If someone has an earlier reference, please share.
I forget the handle of the guy that was serving there. It may have started with Anim.. It was surprising when the first portable player was announced - this mp3 "market" took off much faster than I expected.
On a side note, it is weird when these things suddenly turn like that. Suddenly everyone was talking about mp3s so seriously and I just wanted throw some cold water in their faces and say chill! I guess I could have foreseen the consequences but didn't care so much.
A more modern example of this is the blog. These rambling personal sites/journals are now part of the "blogosphere" - and talked about endlessly as a revolutionizing force. Blog is an over-generalization though, even slashdot is considered a "community blog" by the mainstream now. The format doesn't matter so much as the content, and the availibility the internet brings.
The first blog I knew of were some friends that lived across the hall from me at UMD College Park, and they ran a site called thirteen.org, filled with existantial ramblings like you see today. It was neat because I knew them, and since it took some ingenuity at the time. It no longer takes ingenuity, and so the signal-to-noise ratio has gotten pretty bad. Also, as something like this gets popular, I think newcomers arrive with expectations of the format and then conform to them, so overall there is less quality, creatively speaking.
Thirteen.org belongs to PBS now, cause they sold it. They probably stopped blogging soon after people started calling it that. Probably got tired of saying, "No it's not a _blog_, its my site you dumb shit."
I first got wind of them on EFNet in late 1995. In #mp3 there were only a handful of users, and they were talking about WinPlay3 from the Fraunhofer site. The first songs I downloaded were cheeseball dance music tracks, and needed a Pentium 60 or 66 to play at full quality.
So our interdependence makes us slaves, because some men must work for other men. The root of all problems is economic, the state should build industry and divide the procedes equally so we can live without the hinderance of other men, so we can be left to pursue happiness. How else could we get everyone to cooperate? Natural cooperation leads to happiness, but by forcing cooperation through such absolute control of markets, can happiness be achieved?
In such a system who decides what professional roles we take? Why should I study to become a doctor and work tirelessly when I am getting paid just the same as the farmer that sits on his ass and doesn't even spend the effort to grow decent crops? Who goes to what school to study what, and who decides what range of material will be taught? Where are the incentives for the brilliant to pursue society's most urgent needs? Who will make sure resources get directed to the "best" public needs? Is there a set of "best" public needs? Can a singular government decide priorities better than people with cash in hand, ready to pay a premium for the most wanted products and services? Some innovations require more than a government salary to develop, and many require hard work that won't happend without additional reward.
On top of this, when you disallow markets they surface in the most insidious and corrupt ways, just look at Russia. When market structures are stricly enforced, not even executions can deter alternate ones from forming. Witness Stalin's massacres, a testament that humanely enforcing such a limited market is impossible, even with the strongest of deterrents. Worse yet, these ad-hoc markets are often corrupt, as competitive pressures go unregulated and get brutish, because of a lack of transparency and thus enforcement.
The problem with Marxism is that it complains of the alienation that competitive pressures and jealousy cause, but offers no better solution. It blames the game rather than the individual - ever hear the saying, "don't hate the playa! hate the game!"? It makes about as much sense as marixism or free market capitalism. If we are to be happy, some measure of moderation should exist in the game, so we can cooperate without trampling each other. Such moderation should involve regulations to ensure that people and the environment aren't being abused. Minimum wage comes to mind, but regulations should also allow for upward mobility, offering free or discount upper education to the poorest who show relative merit. Monopolies and regulations protecting business interests over public ones also need to be watched with care, so innovators can rise and claim their stake for better meeting public will.
Considering the deliberate nature of writing a virus, I'd think that a regulations protecting businesses should compensate them for damages caused, as much as the author can pay till bankrupcy. Intent should also be considered in punishment. If the virus results in a death, manslaughter or murder charges ought to be considered. That'll probably mean that losses on the virus writer's side will be minimal with respect to the loss incurred, but should be sufficiently devastating to act as a deterrent to others.
Finally, remember that the alternative to _choosing_ an employer in the game is having one's plays dictated by a less providing and unshakeable governmental oppressor.
Slashdot Mirror
Will this look any better than a movie broadcast in 1080i and played back on an HDTV with 3:2 pulldown detection? I was under the impression that the pull-up done to convert 24fps film into interlaced video is completely reversible, and so 1080 progressive quality is already available through 1080i broadcasts.
It will be nice to have discs of HD content eventually, but I don't see what is so impressive that makes this worthy of coverage.
No no, you've got it all wrong. How many of those 120,000 will have brown skin and a beard? That's the dirty little secret that'll make this system viable.
Not exactly. It's kind of like an algorithm, you can rewrite it under some circumstances using different constructs. You can't do this for more than a few in a copyrighted cookbook, or it is a violation.
r owse_thread/thread/89a947eca7b6b016/fb97fc55d1d9af f2
http://groups.google.com/group/rec.food.cooking/b
A buyers guide without comparisons isn't really all that useful. I'd like to know which machines have relative top quality and take separate cartridges for each color.
Well, if your requirements aren't too beefy (technical term) try labjack. They have kernel drivers for the cheaper model ($119) and the pricier one ($399) uses ethernet and works over TCP/IP. There are links to Perl/Java/.NET/MATLAB interfaces, among others, on their site. The cheaper U12 model support a dozen analog inputs, 20 digital I/O, and 2 analog out. The pricier model has PWM output and support for various timers and whatnot.
http://www.labjack.com/labjack_u12_downloads.html
This switch back to KDE doesn't quite resolve everything. From the slashdot story Suse Linux Founder Exits Novell, it's clear that there has been a big internal fight over the changes happening at Novell. This comment under the above story sheds some light on just what.
It looks like Novell is being pushed to make some bad moves by a major investor or two, with lots of R&D layoffs and pressure to sell off some of the technology that strongly identifies Novell with quality. Rings bells of the sort of thing that happend at HP, with all the PHB talk of changing Novell's "vision," but with little articulated beyond "Linux" as a strategy. This is just a wee bit broad and smacks of a solution looking for a problem.
Which HDTV tuner software are you using? Not all companies use software that takes advantage of ATI cards.
x
http://www.fusionhdtv.co.kr/eng/Products/ATSC.asp
That's the card I used, it states a PIII-750 as the minimum requirement, and that's using the included software, which uses ATI Radeon hardware for accelerated decoding. It claims to work un-accelerated on a 1.6Ghz P4.
Also important is the memory bandwidth of your card - I've heard 10gb/s is required for full HD decoding, which I assume is 1080p. To figure out how much you have, plug and chug with this formula -
(buswidth/8 * memclk) (Double the result for DDR memory)
For a 9600 with a 128 bit wide bus, running DDR400
128/8 * 400 Mhz = 6.4GB/s
Well the machine I was talking about barely handles 1080i using the card spec'ed above, and stutters if you try and do anything. But it works.
Now I doubt this setup would work with MCE. I've read online that some people who run smoothly with Fusion's software get stuttering in MCE, so it may just be that Fusion has a more efficient decoder implementation. MCE uses third party DVD decoders like PowerDVD and Nvidia's decoder to do decoding. They may take advantage of your video card, but seem to be less efficient at utilizing video memory bandwidth. A 9600XT with 600Mhz of bandwidth would likely help that.
Case in point - I've watched OTA HDTV in flawless quality on a PIII-800, using an ATI 9600 series card to accelerate the decoding to a 1920x1200 LCD panel. Even baseline machines will be including powerful graphics coprocessors come the Vista generation.
Also, consider the architecture of a processor like the SONY/IBM cell. The DSP (digital signal processing) features that it provides will blaze through HDTV streams with far fewer Mhz than than a regular PC. Powerful DSPs can not only decode in realtime but can encode and transcode to other formats in real time as well, with far more efficiency than a regular PC. I believe SONY ran a demo where they were decoding something like 48 HDTV streams simultaneously using the PS3 version of the CELL. I expect this sort of tech to make it's way into regular PCs in coming years.
Speed really shouldn't be all that relevant in the long term, as processors are evolving to handle media intensive tasks with much greater efficiency.
This bill is meant to mandate measures with similar ends to those in that article. Right now the means exist but aren't required by law, and aren't implemented widely at all.
Out of all the counties that used e-voting last election, what percentage of those counties implemented systems with measures like those that article? Very, very few. Is that likely to change without legislation? No.
My point is to encourage others to request support for sane mandantory standard. It's facing considerable opposition and needs to build support now, not at the last minute, just before it hits the floor. If it fails to pass, half the people here will be outraged and wonder why. Not to mention we'll be lamenting all over again after the 2008 elections, wondering if there was any tampering in light of circumstantial evidence in key counties.
Send off a quick note to your representative.
The success of an e-vote is hard to verify, and a poorly designed system - like many of those used in the US - makes it fairly trivial to alter even a presidential election through tampering. With no far reaching conspiracies required either, just a few key municipalities in Ohio would need to manipulated. This would be ridiculously easy for a few corrupt local election officials, who through diebold's interface can alter tallies without an audit log. This is a built in feature for making "corrections" and incorporating things like absentee votes of course. There is so much reward involved that the potential abuse here is astronomical.
Sound ridiculous?
Yeah, I know. So while your at it, please check out this bill and write your representatives about it. Some republicans are already poo-pooing these much needed reforms and they need more momentum.
...TCPA type security as well? There probably are some useful aspects to Trusted Computing. I'd imagine it would make it much harder for people to cheat in online video games, or to spoof identities.
http://www.findkatrina.net/
Has just what I had mind. Whew, at least I won't have to stay up building it. Good work. =)
It's a mess, craigslist wasn't really designed for this. You should be able to register yourself and your family, and others searching for you should be able to receive updates on matches as more people register.
This would be far simpler to use and more useful than craigslist. All one would need is a former address, employer, or family name. More criteria could be added later.
I bet those who are worried about their friends and family would be pretty happy to have this.
How about some slashdotters set up a database driven site where people can register to be found and find others? They could list their employer, address, any significant information. I don't have the resources to do this but would be glad to help in throwing together some php and sql if given some server space.
I was there yesterday and quite a few devices were sitting above the sliding doors and in a row along the ceiling as you came into the terminal, and they were oval shaped. They rotated on a platform and spun on a spindle, giving them 360 degrees of freedom. Each white oval was maybe 1.5 feet by 1 foot in diameter and they seemed to follow and track things, mobilizing suddenly at times, but remaining in default position most of the time.
The thing is the each egg shaped "camera" seemed to point with either a lens on one end the oval or a square shaped opening on the opposite side. The square shaped side I imagine has some other sort of detection ability. They looked big and expensive, and I was kind of curious what sort of tech goes into these.
Is anyone on slashdot working on these sorts of applications? Maybe someone could shed some light on what sort of sensory abilities these things have?
Seriously, how about we build one of those sound beaming devices from MIT and then follow Darl around with it?
We could laugh at him and make fun of his pathetic case while is strolling about in public.
They could always pay to have it fixed. The author says much of the code is secure, so why not take undertake a massive effort to overhaul the suspect portions, and then offer a $75 cash incentive for each router a tech patches or a substantial discount for a replacement router? They do have serial #'s so patching could be tracked, perhaps they could even use some relatively inexpensive hardware or software verification module. It could generate a code to verify proper patch status, or even incorporate patching functions in this simple device.
This might hurt business less in the long run than a widespread, debilitating breakdown. It will be expensive, probably ~$120 a pop in the end, considering payout, as well as the cost of verification hardware/software devlopment and production, but they'll reduce the destruction for their customer's businesses and to their own image.
I don't know just how much this would cut into Cisco's revenues, which would of course reduce short term profits and thus investment interest. Someone up there should be weighing something like this though, however painful it sounds. It would also set Cisco apart in market where cheaper competitors are taking away Cisco's profits. How many of them would go to such lengths in the event of a vulnerability? Companies love insuring themselves against everything.
Agreed. For some tasks, better sensory input will be necessary, skin sensations being a particulary tricky one.
Still though, audio and video recognition should make robots far more human than they have been, and I suspect this will take much of the public by surprise. The environments they operate in will no longer have to be so consistent, and the consequence will be that they will appear far less deterministic and mechanical.
We should find hacks around not having skin sensations, at least for some applications. For example, a human might cook a steak by checking grill surface temp with the back of his hand, but a robot could use an infrared pyrometer. It could also estimate the doneness of a steak not by feel, as we might by pressing to check the elasticity of the meat, but by measuring it's initial temperature and thickness, as doneness can be calculated as a factor of these two plus cooking time.
Some problems will still elude, but I suspect we can make the tech adapt using atypical methods, not unlike how a blind person uses alternate senses to do tasks that we normally associate with vision.
This is the parent poster. I'll post this anonymous to avoid karma whoring, but I wanted to expand on what I said, because most readers probably won't undestand the significance of DSPs.
DSPs are good at tasks like pattern matching, filtering out noise, finding statistical correlations, inferring probabilities, and simulating neural networks - among other things. These sorts of tasks are can be done by traditional processors, but such processors aren't designed for this. Something a cheap DSP might handle can hog inordinate amounts of CPU time on a pricey general purpose processor, because architectures like x86 and ppc weren't designed for heavy vector processing. It takes lots of clock cycles on these processors to simulate with software what a DSP does in hardware.
This page lists many DSP capabilities to give you a better idea.
A bipedal robot with stereoscopic vision and so many degrees of freedom could potentially perform some complex tasks that traditional robots have been laughed at for trying.
Robots are often clumsy, and a big reason for this is inadequate sensory ability. Robots are already good at responding very precisely with their limbs, because they have a computer model of how their limbs react and can sense how they are presently positioned, at least with respect to the robot itself. But clumsiness becomes a factor when limbs must be positioned with respect to outside objects - this requires recognition.
The big deal is that processor technology is getting to the point where a robot can analyze high bandwidth sources like video in real-time, for use as feedback for its limbs, or whatever else it may control. With stereoscopic vision the robot could constantly refactor it's movement based on analyses of video, so it could say, go into the garage and roll out the garbage while avoiding obstacles. Or how about the watering plants? Or returning scattered items to their pre-defined places? Wiping down the kitchen counter? Doing laundry? Robots will play a big role in our future, for better or worse. The market will find them hard to resist once they actually become useful.
Fast digital signal processors will play a role in this, not unlike the 7 SPEs found in the Sony/IBM CELL processor. Fast, commoditized (read: cheap) DSPs for this sort of application haven't been available, and that will soon be changing. I suspect PS3 games will use those vector processing units for responding to and learning from high bandwidth inputs like audio, video, or even EEG data to sense states of mind. I mean they called the thing the CELL after all. Likely because of it's seemingly organic response to stimuli.
Firefox is getting to the point that even though it is still first generation browser, administrators of large networks like those at universities are taking note of it's speed, sturdiness, and ability to resist malware.
People are sick of malware, and their authors are increasingly using search engines to cast a wide net. It is not hard to get snagged while searching for popular subjects like mp3s or porn - many of the sites that manage to get their rankings up on these topics use activex installers for malware, or even worse, IE exploits, on their pages. These exploits sometimes install apps without any explicit user notification, though sometimes excess hard drive activity can be noticed while these things are sneaking in.
Even users with machines that are patched against exploits often let activex installers for malware in. This is because IE requests permission to run privileged code through a popup dialogue that demands a response. The problem is that sites often present pop up dialogues for more mundane things, like disclaimers and verification of site-specific information. Clicking no or cancel usually results in something not working, and users have become habituated to clicking yes. So why ask an unfamiliar security question using the same mechanism? The answer is poor identification with users, and an inconsiderate eagerness to "innovate". Both must go hand in hand though.
Firefox doesn't support ActiveX out of the box, but even once it is added, privileged code requests are handled differently. A simple notification that a control on the page doesn't have security permission to execute is overlayed at the top of the web site, and can be safely ignored. The simple message and consistency of purpose in the format lets users easily infer that when they are granting serious privileges. Popups for less privileged contexts, ignorable notifications that default to saftey for critical ones - how clever and innovative! Even better, it is rare that the advertised purpose of a site spreading malware is hindered by ignoring Firefox's notification message, though that may change as it picks up steam. Still though, users will be more aware of when they are granting wider authority.
The tough part with increasing adoption though, is that most users today don't make the browser-malware inference; identifying that the malware that keeps slowing and crashing their machines often stems from a simple page visit and/or dialogue box response. They look towards more overtly dangerous things, like explicit downloads via email attachment or website.
Firefox's notification model and its pending adoption by numerous sysadmins will change that. A critical mass of users will be forced to acknowledge the changes, and so should be more inclined to adopt them at home - through increased familiarity, and by authoritative association of that other browser with the familiar and frustrating ailment of malware.
IE 7.5 will then be forced to acknowledge the state of user experience before offering its own XPerience. Either that or MS will just mimick what it can of the Firefox recipe and rely on it's monopoly to propogate. I wonder which.
Why not allow unlimited TLDs? Official structure and organization through TLDs is overrated.
.mobi for mobile sites rather than embedding that in the application protocol. We should be relying on meta-data to define such distinctions, including the categorization companies, non-profit organizations and ISPs. Meta-data would be more flexible, as distinctions do overlap in ways that domains can't singularly cover.
For example, it seems silly to rely on something like
If you need something authoritative, private authorities could use public/private keys as proof to do that. Indiviuals could then decide which private authorities have standards worth trusting. The U.N. could set up such an authority to authenticate government sites. When a user visits a government site, it could refer the application to the whichever authorities it chooses.
Limiting TLDs just creates conflict as different powerful interests vie for their own distinctions. Sure people can more quickly categorize this way, but the limitations seem to outweight the benefits.
With all the tech people here, it'd be neat to figure out the earliest sources of mp3 trading. If someone has an earlier reference, please share.
I forget the handle of the guy that was serving there. It may have started with Anim.. It was surprising when the first portable player was announced - this mp3 "market" took off much faster than I expected.
On a side note, it is weird when these things suddenly turn like that. Suddenly everyone was talking about mp3s so seriously and I just wanted throw some cold water in their faces and say chill! I guess I could have foreseen the consequences but didn't care so much.
A more modern example of this is the blog. These rambling personal sites/journals are now part of the "blogosphere" - and talked about endlessly as a revolutionizing force. Blog is an over-generalization though, even slashdot is considered a "community blog" by the mainstream now. The format doesn't matter so much as the content, and the availibility the internet brings.
The first blog I knew of were some friends that lived across the hall from me at UMD College Park, and they ran a site called thirteen.org, filled with existantial ramblings like you see today. It was neat because I knew them, and since it took some ingenuity at the time. It no longer takes ingenuity, and so the signal-to-noise ratio has gotten pretty bad. Also, as something like this gets popular, I think newcomers arrive with expectations of the format and then conform to them, so overall there is less quality, creatively speaking.
Thirteen.org belongs to PBS now, cause they sold it. They probably stopped blogging soon after people started calling it that. Probably got tired of saying, "No it's not a _blog_, its my site you dumb shit."
I first got wind of them on EFNet in late 1995. In #mp3 there were only a handful of users, and they were talking about WinPlay3 from the Fraunhofer site. The first songs I downloaded were cheeseball dance music tracks, and needed a Pentium 60 or 66 to play at full quality.
So our interdependence makes us slaves, because some men must work for other men. The root of all problems is economic, the state should build industry and divide the procedes equally so we can live without the hinderance of other men, so we can be left to pursue happiness. How else could we get everyone to cooperate? Natural cooperation leads to happiness, but by forcing cooperation through such absolute control of markets, can happiness be achieved?
In such a system who decides what professional roles we take? Why should I study to become a doctor and work tirelessly when I am getting paid just the same as the farmer that sits on his ass and doesn't even spend the effort to grow decent crops? Who goes to what school to study what, and who decides what range of material will be taught? Where are the incentives for the brilliant to pursue society's most urgent needs? Who will make sure resources get directed to the "best" public needs? Is there a set of "best" public needs? Can a singular government decide priorities better than people with cash in hand, ready to pay a premium for the most wanted products and services? Some innovations require more than a government salary to develop, and many require hard work that won't happend without additional reward.
On top of this, when you disallow markets they surface in the most insidious and corrupt ways, just look at Russia. When market structures are stricly enforced, not even executions can deter alternate ones from forming. Witness Stalin's massacres, a testament that humanely enforcing such a limited market is impossible, even with the strongest of deterrents. Worse yet, these ad-hoc markets are often corrupt, as competitive pressures go unregulated and get brutish, because of a lack of transparency and thus enforcement.
The problem with Marxism is that it complains of the alienation that competitive pressures and jealousy cause, but offers no better solution. It blames the game rather than the individual - ever hear the saying, "don't hate the playa! hate the game!"? It makes about as much sense as marixism or free market capitalism. If we are to be happy, some measure of moderation should exist in the game, so we can cooperate without trampling each other. Such moderation should involve regulations to ensure that people and the environment aren't being abused. Minimum wage comes to mind, but regulations should also allow for upward mobility, offering free or discount upper education to the poorest who show relative merit. Monopolies and regulations protecting business interests over public ones also need to be watched with care, so innovators can rise and claim their stake for better meeting public will.
Considering the deliberate nature of writing a virus, I'd think that a regulations protecting businesses should compensate them for damages caused, as much as the author can pay till bankrupcy. Intent should also be considered in punishment. If the virus results in a death, manslaughter or murder charges ought to be considered. That'll probably mean that losses on the virus writer's side will be minimal with respect to the loss incurred, but should be sufficiently devastating to act as a deterrent to others.
Finally, remember that the alternative to _choosing_ an employer in the game is having one's plays dictated by a less providing and unshakeable governmental oppressor.