I can see someone buying one brand of endorsed sneaker because they are different from other sneakers. Sure, [famous player] endorses them, but they have unique properties that aren't available in other shoes (e.g., the cool patterns and pumps and stuff).
But, with a PC with publicized (and easily duplicated) specs, where's the benefit. If they add anything to the price for the "MTV" logo without adding some extra software or cool design features, there's nothing to stop Dell and every other computer manufacturer from creating their own "Compare to" models with equally cool colors, etc...
Alternatively, the yahoo database could've been screwed up and no story content was associated with the link. So, when users clicked the link they were fed the template (including banner ad) but no additional content. So, it could be a screw up... However, as far as I know there is no group that has oversight on such issues, and there would be nothing (except consumer reaction) that would stop Yahoo from doing something like that.
Contrast that to members of the American Society of Magazine Editors who must follow strict guidlines about the inclusion and identification of advertising content (in both paper and online publications). Their ethical standards (Here) include:
Neither links nor other references to special advertising sections, or "advertorials," shall appear in the table of contents, directory of contents, or in any listing of editorial content of an online publication.
and
The layout, design and type face of advertising pages should be distinctly different from the publication's normal layout, design and type faces.
Does anyone know if Yahoo or any other portals have been pressured to accept such codes of ethics?
Although you need ID, in theory it doesn't have to be state issued. If you were to carry around your passport you'd probably be fine (and could always jaunt off to Canada on a moments notice).
According to the article, The company says its written agreement with Internet subscribers offers no warranty against "uninterrupted use or operation of the equipment or services."
I think I'll go into the ISP business. I'll sell customers a 100gigabit access for $19.95 per month. When they complain that they ain't getting any service I'll point out the Rogers Clause in their contract, and try to upgrade them to my premiere package (1 zillion terabits for the low, low price of $39.95 pre month).
This was a great analysis, I've been thinking about the use of estimation models. And I think that comparing open and closed source models may be a bit trickier than one would first think.
Large open source projects are more likely than proprietary closed source projects to involve developers who are not physically co-located. This means that communication between developers is a bit more of a pain (e.g., you can't walk down the hall to discuss a problem). As a result of this and other factors, it's conceivable that physically co-located programmers may be more productive. As you may recall, there's evidence that a "war room" can result in 2x performance. I don't know much about the COCOMO model used in this paper, but I could imagine that it could be greatly affected by issues such as these.
Of course this doesn't take into account other benefits of distributed teams (e.g., more varied perspectives) or of open source programmers (intrinsic vs extrinsic motivation). But it's something to consider.
The Supreme Court decided that law enforcement agencies need a warrant to use technologies like thermal imaging to "look" into your home. In theory, the cops can't (without probable cause or permission, etc...) walk in and start searching your house without a judge's approval, the Supremes said that the police can't rely on technology to look inside your house unless they get the same type of approval. In that case the cops brought an infrared camera to his house and looked at the thermal activity.
This is interesting because now we have a systematically assembled and stored database.
The question is, if the spatial resolution on these scans were good enough to detect growing lamps in your house (who knows?), would the authorities need a search warrant to look at their own data?
This raises a more general question about the use of data mining for law enforcement. With increased collection of data about where we go and what we do (credit card records, electronic toll paying devices, face recognition software, satellite images, etc.) Can (and should) the government search datasets that it owns (or others) looking for suspicious patterns? Yes, they can use these data once they suspect you, but can they use it to find new suspects?
I knew I'd seen this before... IBM announced this back in November. They've been demoing prototypes of this thing since 1998.
Having seen the demo I can tell you this thing is amazing. Here's the scene whenever they demo the system at a trade show:
IBM Guy stands by display smiling as a slideshow of photographs, maps, and newsprint flashes across the screen.
Techies walking by are inexplicably drawn to the display, much drooling ensues.
IBM Guy explains that it takes tons of processing power to drive the damn thing.
Techies ask how much it costs.
IBM Guy says something like "This particular prototype cost $2 zillion."
Techies drool
IBM Guy prepares for the next question, which everyone, without fail, asks...
Techies ask, "Can I have it?"
IBM Guy smiles and says "No" instead of something clever like, "Gee, that's the first time I've heard that one. Have you considered a career in comedy?"
Techies go home with visions of megapixels dancing in their heads, and eventually, one special techie posts his observations. Finally overcoming the embarassment of realizing that his clever line, "Can I have it?" was not so clever after all.
Ah, but what if it turns out that platform does matter? Imagine if it turned out that people playing Mortal Kombat XVI on the Dreamcast were more likely to beat those on PS2...
There are already issues with people tweaking their computers a bit to gain unfair advantages in games like quake, this could lead to some interesting problems.
This type of device, while pretty cool for input, would be much cooler if it incorporated force feedback technology.
Until that's done, this doesn't seem much more impressive than Nintendo's Powerglove (which was cool, but didn't really do wonders for the game playing experience IMHO).
For more on various glove interfaces, see this page
A TV will choose programs the viewer enjoys. Better yet, commercials that annoy will not be repeated.
Given the current direction of television advertising, and technology like TiVo and the ability to use digital product placement, we're more likely to see TV shows tailored to the extent that you'll see a show where characters use products that are chosen to fit your demographic. So, if you're a rich guy you'll see a Rolex on the character's wrist, whereas I'll see some crappy armitron watch that fits my budget.
Either way, I'm guessing that me being annoyed by a commercial is not something that the broadcasters or advertisers will be too concerned with.
What was tested here isn't what latency people can really feel but what reported time causes them to look for another server.
You're absolutely right. A better experiment would involve manipulating ping times or other network factors (while managing to keep players blind to their experimental condition) and measuring things like satisfaction, score, etc.
Interestingly, even if you can't tell that you've got a bad connection you may show physiological signs of stress. Check out this cool paper which looks at the effects of frame rate (in video conferencing) on physiological symptoms such as heart rate and palm sweat. Even when people didn't know that the frame rate was reduced, they showed signs of stress.
That being said, I think this was a cool use of available data, with implications about judgments and preferences rather than about performance.
The salon article quotes the RIAA general counsel as saying that they would not sue Felton. Quoth the article:
RIAA general counsel Cary Sherman told The Associated Press late Thursday that the [threatening] letter [from Oppenheim] was a mistake.
"We simply weren't as sensitive as we should have been about how the letter would be perceived," Sherman said.
When asked if the RIAA would sue if the paper were published, as Felten fears, Sherman said "No."
The example the Ford lawyer used in the hearing (for more on that, see my other post) was that it was ok for someone to wear a t-shirt that says "F*ck the draft" but it isn't ok for you to put that shirt on an official with the draft board (undoubtedly an entertaining scene).
I just got back from watching the case unfold in the courtroom (and stopping off for a Slurpee on the way back). Here are some observations:
About 20 of the 25 or so people in the gallery were obviously on the 2600 side of the debate (white male twentysomethings with a penchant for wearing t-shirts (one in a 2600 T-shirt) -- some with O'Reilly books (XML and Apache)).
Ford had 4 lawyers, led by Thomas Lee. Corley had one lawyer, Eric Grimm.
The Ford case seemed to be:
- the web site (referred to as www.fgeneralmotors.com so as to save everyone the embarrassment) appears to the casual web surfer to be owned by Ford. Since it associates an obscene, caustic word referring to a close competitor of Ford with Ford's trademarked website, it was causing damage to Ford's intellectual property.
- this confusion couldn't be remedied by most users.
- If this is parody, then it isn't funny, and nobody gets it.
- Free speech rights aren't relevant in this case because this is about commerce and the sullying of trademarks.
- They made Corley re-sign his statement because it was originally signed with his pseudonym Emanual Goldstein - the judge thought that his re-signed signature was the same as the first, and Corley had to print his name next to the signature just to make it clear...
They presented printouts from a web site that allows you to query a database of domain names and provide links and whois info to them. They showed that if you queried "generalmotors" you'd get a bunch of stuff including "fgeneralmotors.com" which would link to whois info and the ford site. The whois info that it linked to either didn't include contact info or, as the defense claimed, had the contact info whited out.
The defense claimed that it was highly unlikely that someone would go to the web site without typing in www.fgeneralmotors.com, something that they would have heard via 2600 magazine or wired or something, and thus wouldn't be confused, and that the search engine the plaintiffs were referring to was designed specifically for searching for domain names, and would thus not be used by the consumers who would allegedly be confused. And that it wasn't Corley's responsibility to police other people's databases.
More importantly, they claimed that what they were doing wasn't commercial in nature, and that it was a free speech issue.
The judge tried to construct a few what-if scenarios involving situations where users could actually click on links rather than be forwarded to the site automatically.
My biggest concern stems from the apparent inept skills of Ford's programmers. The Ford lawyer repeatedly stated that Corley could remedy this situation with "a few computer keystrokes," while Ford's programmers would have to write complex software and test it on their hundreds of web sites in order to block traffic from fgeneralmotors.com. Ford cars have tons of computers in them, I hope those were programmed by people who are at least capable of doing some simple referrerd checking!
Corely's lawyer appeared a bit more nervous than Ford's lawyer, and he kept forgetting that you are supposed to stand up when addressing the court. He obviously knows his technology, but didn't appear as strong on the legal aspects as Ford's lawyers (who didn't bowl me over with their tech knowledge). The judge commented that he was a little flustered by receiving CD-ROMs from Corley's lawyer.
My prediction is that they'll get the injunction, and the ultimate outcome of the case will swing on Corley's lawyer dealing with Ford's reference to trademark law, and not relying strictly on the first amendment arguments.
You shouldn't have to use DeWHATEVER to excersize your fair use rights. First off, most people will be unaware or unable to use something like that. Second off, why should I be treated like a criminal for doing something perfectly legal?
When will the manufacturers ever learn that putting strong copy protection in to consumer devices simply does not work. It does not work because the first people who will adopt a new technology are the technology-savvy users, users who do not want to be told by the big media companies how they are allowed to use their technology.
[snip]
Don't think for one second that DVD would have caught on the way it has if mod chips to defeat region coding were not so readily available.
I think you'd be surprised at how few early adopters can actually circumvent DVD region coding.
While it would be great if people would stand up against schemes that infringe against fair use, the fact is that the lure of better image quality, coupled with wacky outtakes, interviews with the director, movie trailers, and never-before-seen behind the scenes footage was strong enough to build a critical mass of DVD users -- most of whom CANNOT excercise their fair use rights on materials the OWN.
By buying legislation like the DMCA, the MPAA and others have succeeded in overruling the Supreme Court's fair use decision.
The fact remains that if you can see it or if you can hear it, it can be duplicated.
While you are right (at least until analog recording devices are rendered obsolete), the mere fact that you can record anything is not the same as saying anyone with a non-infringing use can record the signal.
Yes, protected audio and video streams can be stored. Practically speaking though, >99% of users are unable to do so.
Besides, if I have a fair use right to time-shift content, I shouldn't have to point my videocamera at the TV.
This gets at an interesting point that I think was glossed over when the Librarian of Congress came up with exceptions to the DMCA. In deciding that the CSS content protection technology used on DVDs not be exempt from the law, the following argument was made:
...it is clear that, at present, most works available in DVD format are also available in analog format (VHS tape) as well.... When distributed in analog formats--formats in which distribution is likely to continue for the foreseeable future--
these works are not protected by any technological measures controlling access.... Therefore, any harm caused by the existence of access control measures used in DVDs can be avoided by obtaining a copy of the work in analog format.
They essentially said "don't worry about lack of fair use for DVDs, you can always rely on VHS." Imagine if the law had been passed 10 years earlier and CDs had similar content protection measures. The Librarian of Congress would be saying, "Don't worry, you still have LPs and cassettes."
New LPs are extremely rare, and it looks like cassettes are going the same way. THIS IS THE FUTURE FOR VIDEO. I have no doubt that the studios will start phasing out VHS tapes in the near future. As they go so shall the VHS player.
We already see differental treatment of DVDs and VHS. Let's say that you want to buy a copy of that hit movie Antitrust (starring Ryan Phillippe as a hacker -- realistic...). The DVD version can be yours for the low, low price of $19.99 at Amazon.com. But wait, you want to be able to make fair use of the video, so you take heart in the fact that the VHS version is available as well. Boy are you in for a surprise. The VHS version can be had for a mere $106.99 from our pals at Amazon.com.
Beyond the technical details (which are, of course, important), there are a number of important issues in building a collaboratory. Knowing whether video or a shared editor or presentation tools are important for a particular task is important. Also, understanding the interpersonal dynamics between users makes a difference (e.g., a system for teaching is different than one for research). Providing the right types of forums for communication and archiving is also useful.
One of the most active groups involved in the development of scientific collaboratories is CREW (the Collaboratory for Research on Electronic Work) at the University of Michigan. It's an interdisciplinary team that has worked on a number of successful Projects including Collaboratories for Space Physics, AIDS Research, and Breast Oncology.
Re:I can see the medical claims rolling in...
on
Surround Lights
·
· Score: 1
Epoleptics throughout the world beware poorly written 'light tracks' will send thousands into grand mal seizures.
I had a chance to talk with a CK employee about this very issue. They've read the literature on stroboscopic light induced seizures, and apparently there's a somewhat well defined zone that can induce seizures (mostly in kids), and they simply avoid those rates. Implementations of this technology will hopefully block out potentially dangerous flicker rates so that game developers don't knowingly or unknowingly start causing seizures...
I have a CK Sauce light stick. Nobody seems impressed when I show it to them. The typical response is "What use is that?" My answer: "They've got patents to use."
One last thing...could someone snag the source code behind the Speech-To-Text translation in Echelon for inclusion into this? I'd really dig being able to index all that radio/movie/tv stuff with a localized search engine, so I can easily find that scene from "The Ref" where Dennis Leary is commenting on why Gus should be so upset:)
Why waste your time (and cycles) with speech reco, use closed caption data.
I'd love to have a Tivo-like box that monitored a few channels for certain key terms in the closed captions, then recorded relevant shows (of course it would have to be buffering the previous x minutes like Tivo). For example, say you're interested in censorware, you could scan CNN, MSNBC, Fox News, TechTV, etc. for words like "filtering software" "Net Nanny" etc.
Tivo's wishlist feature kind of approaches this by allowing you to search TV show descriptions for actors, keywords, and directors. But I want more!
Slashdot Mirror
But, with a PC with publicized (and easily duplicated) specs, where's the benefit. If they add anything to the price for the "MTV" logo without adding some extra software or cool design features, there's nothing to stop Dell and every other computer manufacturer from creating their own "Compare to" models with equally cool colors, etc...
Where's the benefit?
Contrast that to members of the American Society of Magazine Editors who must follow strict guidlines about the inclusion and identification of advertising content (in both paper and online publications). Their ethical standards (Here) include:
Neither links nor other references to special advertising sections, or "advertorials," shall appear in the table of contents, directory of contents, or in any listing of editorial content of an online publication.
and
The layout, design and type face of advertising pages should be distinctly different from the publication's normal layout, design and type faces.
Does anyone know if Yahoo or any other portals have been pressured to accept such codes of ethics?
Although you need ID, in theory it doesn't have to be state issued. If you were to carry around your passport you'd probably be fine (and could always jaunt off to Canada on a moments notice).
I think I'll go into the ISP business. I'll sell customers a 100gigabit access for $19.95 per month. When they complain that they ain't getting any service I'll point out the Rogers Clause in their contract, and try to upgrade them to my premiere package (1 zillion terabits for the low, low price of $39.95 pre month).
Large open source projects are more likely than proprietary closed source projects to involve developers who are not physically co-located. This means that communication between developers is a bit more of a pain (e.g., you can't walk down the hall to discuss a problem). As a result of this and other factors, it's conceivable that physically co-located programmers may be more productive. As you may recall, there's evidence that a "war room" can result in 2x performance. I don't know much about the COCOMO model used in this paper, but I could imagine that it could be greatly affected by issues such as these.
Of course this doesn't take into account other benefits of distributed teams (e.g., more varied perspectives) or of open source programmers (intrinsic vs extrinsic motivation). But it's something to consider.
This is interesting because now we have a systematically assembled and stored database.
The question is, if the spatial resolution on these scans were good enough to detect growing lamps in your house (who knows?), would the authorities need a search warrant to look at their own data?
This raises a more general question about the use of data mining for law enforcement. With increased collection of data about where we go and what we do (credit card records, electronic toll paying devices, face recognition software, satellite images, etc.) Can (and should) the government search datasets that it owns (or others) looking for suspicious patterns? Yes, they can use these data once they suspect you, but can they use it to find new suspects?
Having seen the demo I can tell you this thing is amazing. Here's the scene whenever they demo the system at a trade show:
IBM Guy stands by display smiling as a slideshow of photographs, maps, and newsprint flashes across the screen.
Techies walking by are inexplicably drawn to the display, much drooling ensues.
IBM Guy explains that it takes tons of processing power to drive the damn thing.
Techies ask how much it costs.
IBM Guy says something like "This particular prototype cost $2 zillion."
Techies drool
IBM Guy prepares for the next question, which everyone, without fail, asks...
Techies ask, "Can I have it?"
IBM Guy smiles and says "No" instead of something clever like, "Gee, that's the first time I've heard that one. Have you considered a career in comedy?"
Techies go home with visions of megapixels dancing in their heads, and eventually, one special techie posts his observations. Finally overcoming the embarassment of realizing that his clever line, "Can I have it?" was not so clever after all.
There are already issues with people tweaking their computers a bit to gain unfair advantages in games like quake, this could lead to some interesting problems.
Ok, this is odd. If this is satire, how come this isn't?
[/sarcasm]
Until that's done, this doesn't seem much more impressive than Nintendo's Powerglove (which was cool, but didn't really do wonders for the game playing experience IMHO).
For more on various glove interfaces, see this page
Given the current direction of television advertising, and technology like TiVo and the ability to use digital product placement, we're more likely to see TV shows tailored to the extent that you'll see a show where characters use products that are chosen to fit your demographic. So, if you're a rich guy you'll see a Rolex on the character's wrist, whereas I'll see some crappy armitron watch that fits my budget.
Either way, I'm guessing that me being annoyed by a commercial is not something that the broadcasters or advertisers will be too concerned with.
This is so cool. Linux users don't have to pay taxes!
The CIA has invested $1 Million in Safeweb, and uses it for its own agents (and I believe they use triangleboy when in the field.
So, if they trust it, then why shouldn't you... Of course, if they have a stake, whose to say they don't have a bit of insider access...
"The issue is not whether you are paranoid, the issue is whether you are paranoid enough."
- Max, "Strange Days"
You're absolutely right. A better experiment would involve manipulating ping times or other network factors (while managing to keep players blind to their experimental condition) and measuring things like satisfaction, score, etc.
Interestingly, even if you can't tell that you've got a bad connection you may show physiological signs of stress. Check out this cool paper which looks at the effects of frame rate (in video conferencing) on physiological symptoms such as heart rate and palm sweat. Even when people didn't know that the frame rate was reduced, they showed signs of stress.
That being said, I think this was a cool use of available data, with implications about judgments and preferences rather than about performance.
RIAA general counsel Cary Sherman told The Associated Press late Thursday that the [threatening] letter [from Oppenheim] was a mistake. "We simply weren't as sensitive as we should have been about how the letter would be perceived," Sherman said.
When asked if the RIAA would sue if the paper were published, as Felten fears, Sherman said "No."
I'm counting that as a win for the good guys...
The example the Ford lawyer used in the hearing (for more on that, see my other post) was that it was ok for someone to wear a t-shirt that says "F*ck the draft" but it isn't ok for you to put that shirt on an official with the draft board (undoubtedly an entertaining scene).
I just got back from watching the case unfold in the courtroom (and stopping off for a Slurpee on the way back). Here are some observations:
About 20 of the 25 or so people in the gallery were obviously on the 2600 side of the debate (white male twentysomethings with a penchant for wearing t-shirts (one in a 2600 T-shirt) -- some with O'Reilly books (XML and Apache)).
Ford had 4 lawyers, led by Thomas Lee. Corley had one lawyer, Eric Grimm.
The Ford case seemed to be:
- the web site (referred to as www.fgeneralmotors.com so as to save everyone the embarrassment) appears to the casual web surfer to be owned by Ford. Since it associates an obscene, caustic word referring to a close competitor of Ford with Ford's trademarked website, it was causing damage to Ford's intellectual property.
- this confusion couldn't be remedied by most users.
- If this is parody, then it isn't funny, and nobody gets it.
- Free speech rights aren't relevant in this case because this is about commerce and the sullying of trademarks.
- They made Corley re-sign his statement because it was originally signed with his pseudonym Emanual Goldstein - the judge thought that his re-signed signature was the same as the first, and Corley had to print his name next to the signature just to make it clear...
They presented printouts from a web site that allows you to query a database of domain names and provide links and whois info to them. They showed that if you queried "generalmotors" you'd get a bunch of stuff including "fgeneralmotors.com" which would link to whois info and the ford site. The whois info that it linked to either didn't include contact info or, as the defense claimed, had the contact info whited out.
The defense claimed that it was highly unlikely that someone would go to the web site without typing in www.fgeneralmotors.com, something that they would have heard via 2600 magazine or wired or something, and thus wouldn't be confused, and that the search engine the plaintiffs were referring to was designed specifically for searching for domain names, and would thus not be used by the consumers who would allegedly be confused. And that it wasn't Corley's responsibility to police other people's databases.
More importantly, they claimed that what they were doing wasn't commercial in nature, and that it was a free speech issue.
The judge tried to construct a few what-if scenarios involving situations where users could actually click on links rather than be forwarded to the site automatically.
My biggest concern stems from the apparent inept skills of Ford's programmers. The Ford lawyer repeatedly stated that Corley could remedy this situation with "a few computer keystrokes," while Ford's programmers would have to write complex software and test it on their hundreds of web sites in order to block traffic from fgeneralmotors.com. Ford cars have tons of computers in them, I hope those were programmed by people who are at least capable of doing some simple referrerd checking!
Corely's lawyer appeared a bit more nervous than Ford's lawyer, and he kept forgetting that you are supposed to stand up when addressing the court. He obviously knows his technology, but didn't appear as strong on the legal aspects as Ford's lawyers (who didn't bowl me over with their tech knowledge). The judge commented that he was a little flustered by receiving CD-ROMs from Corley's lawyer.
My prediction is that they'll get the injunction, and the ultimate outcome of the case will swing on Corley's lawyer dealing with Ford's reference to trademark law, and not relying strictly on the first amendment arguments.
You shouldn't have to use DeWHATEVER to excersize your fair use rights. First off, most people will be unaware or unable to use something like that. Second off, why should I be treated like a criminal for doing something perfectly legal?
[snip]
Don't think for one second that DVD would have caught on the way it has if mod chips to defeat region coding were not so readily available.
I think you'd be surprised at how few early adopters can actually circumvent DVD region coding.
While it would be great if people would stand up against schemes that infringe against fair use, the fact is that the lure of better image quality, coupled with wacky outtakes, interviews with the director, movie trailers, and never-before-seen behind the scenes footage was strong enough to build a critical mass of DVD users -- most of whom CANNOT excercise their fair use rights on materials the OWN.
By buying legislation like the DMCA, the MPAA and others have succeeded in overruling the Supreme Court's fair use decision.
While you are right (at least until analog recording devices are rendered obsolete), the mere fact that you can record anything is not the same as saying anyone with a non-infringing use can record the signal.
Yes, protected audio and video streams can be stored. Practically speaking though, >99% of users are unable to do so.
Besides, if I have a fair use right to time-shift content, I shouldn't have to point my videocamera at the TV.
They essentially said "don't worry about lack of fair use for DVDs, you can always rely on VHS." Imagine if the law had been passed 10 years earlier and CDs had similar content protection measures. The Librarian of Congress would be saying, "Don't worry, you still have LPs and cassettes."
New LPs are extremely rare, and it looks like cassettes are going the same way. THIS IS THE FUTURE FOR VIDEO. I have no doubt that the studios will start phasing out VHS tapes in the near future. As they go so shall the VHS player.
We already see differental treatment of DVDs and VHS. Let's say that you want to buy a copy of that hit movie Antitrust (starring Ryan Phillippe as a hacker -- realistic...). The DVD version can be yours for the low, low price of $19.99 at Amazon.com. But wait, you want to be able to make fair use of the video, so you take heart in the fact that the VHS version is available as well. Boy are you in for a surprise. The VHS version can be had for a mere $106.99 from our pals at Amazon.com.
Why do I feel like I've been mugged?
One of the most active groups involved in the development of scientific collaboratories is CREW (the Collaboratory for Research on Electronic Work) at the University of Michigan. It's an interdisciplinary team that has worked on a number of successful Projects including Collaboratories for Space Physics, AIDS Research, and Breast Oncology.
Forget the black lights. Get some 16 million color led arrays inside your PC, let it perform light shows and use your pc case as a winamp plug in. It's bound to be cheaper than mood lighting for the whole damn room.
I had a chance to talk with a CK employee about this very issue. They've read the literature on stroboscopic light induced seizures, and apparently there's a somewhat well defined zone that can induce seizures (mostly in kids), and they simply avoid those rates. Implementations of this technology will hopefully block out potentially dangerous flicker rates so that game developers don't knowingly or unknowingly start causing seizures...
I have a CK Sauce light stick. Nobody seems impressed when I show it to them. The typical response is "What use is that?" My answer: "They've got patents to use."
Why waste your time (and cycles) with speech reco, use closed caption data.
I'd love to have a Tivo-like box that monitored a few channels for certain key terms in the closed captions, then recorded relevant shows (of course it would have to be buffering the previous x minutes like Tivo). For example, say you're interested in censorware, you could scan CNN, MSNBC, Fox News, TechTV, etc. for words like "filtering software" "Net Nanny" etc.
Tivo's wishlist feature kind of approaches this by allowing you to search TV show descriptions for actors, keywords, and directors. But I want more!