You wouldn't want to do something as crude as looking at the filesize, instead there are some pretty crude techniques for measuring image similarity. One of the simplest, and most common methods for a computer to assess image quality is to calculate the Peak Signal-to-Noise Ratio (PSNR) between the current and previous frame. PSNR is based on a calculation of the Mean Squared Error (MSE) of the luminance differences between equivalent pixels on the old (f) and new images (f').
MSE= (Sigma[f(i,j)-f'(i,j)]^2)/N^2
Where N is the number of pixels. PSNR in decibels (dB) is calculated as
PSNR=20 log10 ((255^2)/MSE)
A higher PSNR between two images indicates a greater degree of similarity, the PSNR of identical images will be infinity. Although this calculation is a useful way of determining overall similarity between images, it does not necessarily correlate with human judgments particularly well. One crucial limitation of PSNR is that it is a global measure that treats all deviations the same. Therefore, a slight, barely perceptible, uniform degradation over the entire image may result in a PSNR that is identical to that of an image with an obvious, severe degradation in a small, prominent location of the picture.
Alternatively, you could look into DCTune, a proprietary algorithm developed at NASA. It is based on some principles of the human visual perceptual system, and gives a score that's more well correlated with human judgments
This past week there was an article on some other source (can't remember) that focused on the whole issue of whether it was a trojan (last paragraph of this story).
As I recall, the spyware also sent the urls that users visited to a machine with an odd domain name (something like 2001-007.com) EVEN IF PEOPLE WANTED TO AVOID INSTALLING THE SPYWARE. This is why it was called a trojan.
I'm not sure if it turns out the software wasn't sending the info (reporter error) or if they've glossed over that fact...
Either way. Blah. Spyware is why I don't play Snood anymore. They use gator which does all sorts of lame stuff to hide itself on install.
Look, if you have to trick users or hide your program, then it probably isn't a "valuable bonus program." Stupid marketing bastards...
Re:Once more, for all the slow JBT's.
on
The Eyes Have It
·
· Score: 3, Informative
This technique has been used for years in psychology studies (and by bad cops on naive suspects). It's called the "Bogus Pipeline"[1]. The basic idea is that, given a quick demonstration of its effectiveness (instruct the target to tell a couple of lies, and show that the machine detects them), along with an incentive to avoid being caught (if you're honest with us, we'll be easy on you), people are more likely to behave honestly.
[1] Jones, E. E., and Sigall, H., "The Bogus Pipeline: A New Paradigm for Measuring Affect and Attitude," Psychological Bulletin 76: 349-364, 1971.
Handspring and all the other PalmOS licensees use the graffiti method, so it could affect them too. My guess is that it will end up meaning extra $ for the OS -- passed along to consumers.
Man, Palm is really having a hard time these days. Could this, coupled with their recent downturn help microsoft innovate them out of business a la netscape?
Sure, Palm was the original, and the only one (along with OS licencees) that offers PDAs that aren't overloaded with pricey color screens, 64mb of memory, and desktop applications. (Well they offer those too, but they still have some good straightforward PDAs). But, with the market crowding, and lots of new Wince apps being written, are we seeing the beginning of the end?
I'd hate to have to buy an overloaded PDA because MS becomes the only game in town...
I love slashback. Except that the links to the original stories are only sometimes listed in the slashback blurbs. Wouldn't it make sense to always link to the original slashdot story?
Just my $1.34 Canadian
Re:Cryptonomicon.....
on
Looking At Turing
·
· Score: 4, Informative
Although Turing was in Cryptonomicon, you have to remember that it is fiction. The fact is, it does take liberties with the truth in the name of dramatic license.
For a good biography of Turing, you should check out Hodges' Alan Turing: The Enigma. A bit of a dense read, but informative and comprehensive.
I agree that the study of brain structure is important (and promising). What pisses me off is when people confuse structure and process. Here's an example:
There's a hormone secreted in the brain called CRH[1], and there's a protein in the brain called CRH Binding Protein. Just like it sounds, CRHBP binds to CRH that's released. One theory says that CRHBP binds to CRH and prevents it from reaching receptors that would respond to it by releasing stress hormones. So, the more CRHBP you have, the less CRH will make it through to the receptors. You'd think then, that an organism with an excess of CRHBP would release less stress hormone (less CRH makes it through), well it turns out that this doesn't happen. What does happen is that animals with excess CRHBP release proportionally more CRH to make it through, thus releasing the appropriate amount of stress hormone, and behaving no different than normal animals.
What does this say? Well, we have a brain difference which, although interesting (and important) has no real bearing on behavior. It is perfectly acceptable to explore the CRH/CRHBP relationship, and in fact it is a scientifically interesting question. Yet, it would be dishonest to look at CRHBP levels and say anything about how the organism will behave.
It's equally dishonest to say that other differences in brain structure are meaningful beyond the fact that they are differences in structure. Until you show some effect of these differences, don't jump to conclusions. My complaint about the article was that they made claims about intelligence based on structural similarities. Bad PR person. Bad.
---
[1] Corticotropin Releasing Hormone
This isn't news, and it isn't the whole story
on
Intelligence is Inherited
·
· Score: 4, Informative
IAACP (I am a cognitive psychologist), so I know the field pretty well.
The nature nurture debate has raged for centuries in respect to intelligence. Ever since Darwin's cousin Francis Galton proposed his theories of hereditary intelligence.
At first glance, it seems neat that these guys did a study on twins. You might think, "wow, what a great approach." You'd be right, then you'd hopefully realize that other people had these ideas in the past, and they did the right studies and came to the right conclusions.
When Thomas Bouchard was in charge of the Minnesota Twin Studies, he and his colleagues compared twins raised together and apart.
Bouchard and colleagues tested 56 pairs of monozygotic (identical) twins who had been raised apart and compared them to hundreds of identical twins who had been reared together. The twins were tested on dozens of capacities. This approach allowed them to examine spatial ability, verbal ability, mathematical ability, personality, etc. Rather than the antiquated "g" or general intelligence factor.
So, here's a study with many more subjects, much better comparisons, and more detailed data. Here's the cool part: if you correlate the scores of twins on the Wechsler Adult Intelligence Scale (WAIS) (the standard IQ test in the US) and Ravens Progressive Matrices (An allegedly culturally unbiased test of reasoning skills), you see a 65% (WAIS) or 50% (Raven's) agreement among scores of twins raised together, and among twins raised apart, the WAIS scores are about 35% in agreement with nearly the same 50% agreement on Raven's scores.
What does this mean? Well, it means that even in the best cases, IQ scores are about 50% hereditary. In addition, the big drop off between WAIS scores indicates that environment has an important role in intelligence. Given that most of the twins were raised in similar middle class families (who adopt kids), the estimates of the role of environment are probably inflated.
That being said, the study mentioned in the story focuses primarily on brain structures. This pisses me off. One of the lamest things about current cognitive neuroscience is the common misunderstanding of the difference between structure and process. Given the little we know about how the brain works (yeah, we're getting good info about the molecular level, and we know in general what larger regions like Broca's area are responsible for, but we have woefully little info about how these structures actually work), idiot PR people take brain findings and blow them out of proportion.
The fact is, if you want to talk about intelligence, you have to talk about behavioral measurements. Looking at structures can, at best, tell you if something is wrong. So, yeah big deal, major structures are heritable, given the fact that environment has been shown in better studies to play a key role in intelligence, we shouldn't rely too much on these findings.
Bottom line: Yes of course nature plays an important role in intelligence. You'd have to be an idiot not to realize that. This study is not groundbreaking. Moreover, the headline of the article is sensationalistic and only a half-truth. Environment does play a role in intelligence. My god, meet a person who ate lead paint as a kid and you'll realize that.
Actually, there is little reason to doubt the NCTA data. Although it's called a "survey," their findings are better thought of as sales reports. The NCTA is an industry association, and will certainly try to put a spin on their findings (e.g., "Consumers' strong response to digital cable services, in spite of difficult times, confirms the excellent value of these new services"), but they'll still report accurate sales figures.
The RIAA (boo) would LOVE it if album sales plummeted at the same time that Napster was taking off. Yet, that wasn't the case (although CD singles did suffer a drop last year, the increased sales of full albums was significantly greater), and the RIAA reported the numbers correctly. Although they put their spin on it, ("Look at the drop in Singles sales"), they reported valid sales figures.
It looks like the first article was a guy trying to create a news story when there really wasn't one. Sure, people will switch from high bandwidth to low bandwidth, but if the general trend overshadows it, we end up with a very different story.
The newer story, which is just a rehash of an NCTA press release says nothing about the people who are installing (or uninstalling) cable modems. It talks about sales trends. This doesn't negate the other story, it just indicates that it isn't all that big a trend.
It's the equivalent of C|Net reporting that I just bought an AMD processor, so Intel had better watch out. Who cares about me? If there are 50,000 people like me, then you notice.
So, yes there are people switching. But no, it doesn't seem to be affecting the industry. Two separate stories, no conflict.
One interesting issue that these guys confronted, and that others must confront as well is whether or not to attempt to imitate the way humans (or other animals) perform or to simply engineer a solution that works.
As an example, imagine that you are creating a computer vision system. If you want to be able to have it measure aesthetic quality you may wish to rely on our knowledge of how humans judge visual pleasantness. Alternatively, if you are interested in the detection of faint patterns, you may find that humans don't do it as well as you'd like, and that an alternative approach is better.
The discussion of the Honda Robot compared to this one is sort of the same thing. Just because someone else made a robot that doesn't do things like people do them, doesn't mean the robot is not as good. If the goals are to simulate people, then yes this robot is better. If however, the goals are different, then this may not be better (e.g., it can't carry as much weight).
When I tried to access the story I got an error message:
Proxy Error
The proxy server received an invalid response from an upstream server.
The proxy server could not handle the request GET/site-stories/2001/1025.errorhandling.html.
Reason: Could not connect to remote machine: Connection refused
Additionally, a 404 Not Found error was encountered while trying to use an ErrorDocument to handle the request.
I used proxomitron to set my user-agent to some nonsense (it was "!b" without the quotes) and went to msn.com and got the standard "upgrade to IE" message. Then I went to msn.com.br and got this message:
Microsoft VBScript runtime error '800a000d'
Type mismatch: '[string: ""]'
/include/browser.asp, line 45
I figured I'd be able to find out a bit about their filter (written in vbscript) by altering the input a bit. So then I changed my user agent to "!b(msie)" and got
Microsoft VBScript runtime error '800a000d'
Type mismatch: '[string: "e"]'
/include/browser.asp, line 45
My hypothesis was that it looked at something right before the ")". So then I tried "!b(msie;msie;windows)" and got
Nota ao usuário
Infelizmente seu navegador não é compatível com essa versão, portanto não será possível visualizar todo o conteúdo que disponibilizamos especialmente para você. Para ver corretamente o novo portal MSN, baixe agora a versão mais recente do Internet Explorer ou o novo MSN Explorer. É grátis!
Se você acha que recebeu essa mensagem por engano, por favor confirme que seu navegador possui capacidade de visualização instalada. (No Internet Explorer, acesso pelo menu Visualizar | Opções).
which looks like it says something about having to download IE and that it's free. So I changed back to !b and then to !b(msie) and kept getting that final message. Looks like they must have changed the script or something. Just as it was getting good.
Overall this looks like a nice gift to the big movie and record companies at the cost of consumers fair use and other rights. However, section 103(b) states:
PERSONAL TIME-SHIFTING COPIES CANNOT BE BLOCKED. -- No person may apply a security measure that uses a certified security technology to prevent a lawful recipient from making a personal copy for time-shifting purposes of programming at the time it is lawfully performed, on an over-the-air broadcast, non-premium cable channel, or non-premium satellite channel, by a television broadcast station (as defined in section 122(j)(5)(A) of title 17, United States Code), a cable system (as defined in section 111(f) of such title), or a satellite carrier (as defined in section 119(d)(6) of such title.)
Which is a good first step. But this arguably would let HBO or any other premium service (do other packaged channels count as premium services) prevent me from recording The Sopranos or any of their America Undercover series...
This seems like an odd thing to do. So, for example, when the Sopranos goes into syndication I'll be able to record it off of UPN but not HBO.
I'm a teacher. I use recorded materials to teach my students. I'm allowed to do so under fair use. This law could make me into a criminal. That sucks.
The sickening part of this is that the RIAA is essentially using the deaths of thousands of people as cover to expand their power.
No matter how bad copyright infringment may be, it is much less morally repulsive than using a tragedy like this to advance a political agenda. The RIAA has stooped to a new low.
A. Little perceived need by the masses
B. Hassle to use
and more recently
C. Government rumblings
A. could be dealt with by some good old FUD. I've always been amazed that NAI and others have resisted the evil urge to play on naive users' fears of "hackers." Come on, companies with lame IDS and Firewall products have been playing the fear card for a while. Imagine how effective a campaign would be if the product were actually good... (Not that I'm a fan of these tactics).
B. is a more difficult problem. Although the product has come a long way since the old DOS version with it's confusing options, it has a way to go to acheive true ease of use. People don't necessarily "get it." I'm not a huge fan of dumbing down interfaces, but a real simple set of wizards that handled all the stages of key creation and software integration would be helpful. Plug-ins for email are good, but a deal with MS or Eudora to bundle it would be better. Plug-in with ICQ is good but a bit clumsy at times. Maybe playing up the Envelope metaphor in email programs would be better... Also, encouraging users to get their email contacts to install the freeware version would be great. Maybe, a window that popped up when people tried to send an encrypted email to a person whose key isn't know. The window could mention the problem, and offer to send the recipient an email with a link to the freeware (or perhaps a free "reader" that allowed for key creation and email integration).
With C. the issue is just a big hassle. At some point you'd hope the Gov't would realize that restricting strong encryption will have no effect on criminals, only business and home users.
GPS is cool, and future emergency GPS location will be cool too.
What concerns me is that the phone relies on the network to get restaurant and driving directions. My Visor with appropriate GPS springboard and something like Vindigo allows me to find nearby stuff using a local database (i.e., on the client). My SprintPCS phone has a feature that lets me set an alarm at a specific time. However, this feature only works when my phone is in the phone network. That is, even though my phone should be perfectly capable of keeping time, it needs a digital PCS signal to actually do so.
If I use this phone to do anything with the GPS aside from finding my own location, I'm forced to send the PCS network my location. While I really don't care that much about the privacy of my whereabouts, I certainly don't need anyone sending me ads targeted to my present location.
The long audio delays can come from two different sources. First off, there is the amount of time it takes for the signal to travel from the phone to the tv studio. Second, it typically takes a bit of time for the video decompression to occur. Rather than lose AV syncro, the audio is actually delayed to be played with the appropriate video.
Here's where it gets interesting: When someone asks you a question in real life, they rely on the amount of hesitation on your part to assess how honest you're being. With these things, you always appear to be hesitating before speaking. Imagine a debate or discussion with someone in studio and someone else in the field. You'd have an asymmetry in how honest these people are perceived.
When he demonstrated the orginal computer mouse, Doug Engelbart used a one handed chorded keyboard for his left hand at the same time. The system that they developed allowed users to type all characters using just the chords on the left hand, but they still had a normal keyboard... Why? Because the standard keyboard was pretty efficient. So, what did they use the chorded keyboard for? Functions. Copy, paste, print, etc were all associated with chord combinations. This way, you move your hands away from the keyboard to use the mouse and chord when you want to do functional stuff. But, when you're typing you put both hands on the keyboard and pound away.
The learning curve for a two-handed chorded keyboard is sort of long, but not rediculous. Learning a new layout (e.g., Dvorak) on a standard keyboard takes about 20 hours (e.g., 1 hour a day every weekday for a month). Chording is a learned skill, which can be acquired relatively quickly. I'm guessing it's faster if you have piano skills...
They don't even compare to the utter crappiness that was SimEarth...
Re:It's easy to see why nobody is buying these
on
No One Wants The Not-Coms
·
· Score: 3, Insightful
I think that, more importantly, these new domain names don't add anything..biz provides no useful semantic information. It's not like someone would say, "hmm, I'm not looking for a (.)company, I'm looking for a (.)business." Moreover, since the target market is companies, they already are likely to have.com names, thus making this redundant.
This is the equivalent of adding.com2 and.com3 -- nothing useful.
.kids or.sex or.personal or.protest would be useful. These are categories that aren't simply mirrors of existing TLDs.
One challenge in all of this is that our elected representatives don't want to be seen as being soft on terrorism.
What happened was terrible, steps must be taken to protect us, but for some authorities to take advantage of the pain and suffering of others to pass through their pet legislation is sickening.
Don't try to cripple crypto, it won't do anything except erode the rights of people who aren't terrorists. Real terrorists will use non-crippled systems, while those who use off the shelf groupware or banking software will be stuck with the crappy stuff.
Don't let the authorities conduct wiretaps (phone or computer) without warrants. If someone is really a suspect, then you should be able to get a warrant.
But sure, increase security checks at airports, designate more money to terrorism prevention, take appropriate action against those who did this. Just don't try to erode our rights while using this as a smokescreen.
I guess I'm lucky, because my representative, Lynn Rivers seems to get it...
Slashdot Mirror
You wouldn't want to do something as crude as looking at the filesize, instead there are some pretty crude techniques for measuring image similarity. One of the simplest, and most common methods for a computer to assess image quality is to calculate the Peak Signal-to-Noise Ratio (PSNR) between the current and previous frame. PSNR is based on a calculation of the Mean Squared Error (MSE) of the luminance differences between equivalent pixels on the old (f) and new images (f').
MSE= (Sigma[f(i,j)-f'(i,j)]^2)/N^2
Where N is the number of pixels. PSNR in decibels (dB) is calculated as
PSNR=20 log10 ((255^2)/MSE)
A higher PSNR between two images indicates a greater degree of similarity, the PSNR of identical images will be infinity. Although this calculation is a useful way of determining overall similarity between images, it does not necessarily correlate with human judgments particularly well. One crucial limitation of PSNR is that it is a global measure that treats all deviations the same. Therefore, a slight, barely perceptible, uniform degradation over the entire image may result in a PSNR that is identical to that of an image with an obvious, severe degradation in a small, prominent location of the picture.
Alternatively, you could look into DCTune, a proprietary algorithm developed at NASA. It is based on some principles of the human visual perceptual system, and gives a score that's more well correlated with human judgments
This past week there was an article on some other source (can't remember) that focused on the whole issue of whether it was a trojan (last paragraph of this story).
As I recall, the spyware also sent the urls that users visited to a machine with an odd domain name (something like 2001-007.com) EVEN IF PEOPLE WANTED TO AVOID INSTALLING THE SPYWARE. This is why it was called a trojan.
I'm not sure if it turns out the software wasn't sending the info (reporter error) or if they've glossed over that fact...
Either way. Blah. Spyware is why I don't play Snood anymore. They use gator which does all sorts of lame stuff to hide itself on install.
Look, if you have to trick users or hide your program, then it probably isn't a "valuable bonus program." Stupid marketing bastards...
This technique has been used for years in psychology studies (and by bad cops on naive suspects). It's called the "Bogus Pipeline"[1]. The basic idea is that, given a quick demonstration of its effectiveness (instruct the target to tell a couple of lies, and show that the machine detects them), along with an incentive to avoid being caught (if you're honest with us, we'll be easy on you), people are more likely to behave honestly.
[1] Jones, E. E., and Sigall, H., "The Bogus Pipeline: A New Paradigm for Measuring Affect and Attitude," Psychological Bulletin 76: 349-364, 1971.
Handspring and all the other PalmOS licensees use the graffiti method, so it could affect them too. My guess is that it will end up meaning extra $ for the OS -- passed along to consumers.
Man, Palm is really having a hard time these days. Could this, coupled with their recent downturn help microsoft innovate them out of business a la netscape?
Sure, Palm was the original, and the only one (along with OS licencees) that offers PDAs that aren't overloaded with pricey color screens, 64mb of memory, and desktop applications. (Well they offer those too, but they still have some good straightforward PDAs). But, with the market crowding, and lots of new Wince apps being written, are we seeing the beginning of the end?
I'd hate to have to buy an overloaded PDA because MS becomes the only game in town...
I love slashback. Except that the links to the original stories are only sometimes listed in the slashback blurbs. Wouldn't it make sense to always link to the original slashdot story?
Just my $1.34 Canadian
Although Turing was in Cryptonomicon, you have to remember that it is fiction. The fact is, it does take liberties with the truth in the name of dramatic license.
For a good biography of Turing, you should check out Hodges' Alan Turing: The Enigma. A bit of a dense read, but informative and comprehensive.
I agree that the study of brain structure is important (and promising). What pisses me off is when people confuse structure and process. Here's an example:
There's a hormone secreted in the brain called CRH[1], and there's a protein in the brain called CRH Binding Protein. Just like it sounds, CRHBP binds to CRH that's released. One theory says that CRHBP binds to CRH and prevents it from reaching receptors that would respond to it by releasing stress hormones. So, the more CRHBP you have, the less CRH will make it through to the receptors. You'd think then, that an organism with an excess of CRHBP would release less stress hormone (less CRH makes it through), well it turns out that this doesn't happen. What does happen is that animals with excess CRHBP release proportionally more CRH to make it through, thus releasing the appropriate amount of stress hormone, and behaving no different than normal animals.
What does this say? Well, we have a brain difference which, although interesting (and important) has no real bearing on behavior. It is perfectly acceptable to explore the CRH/CRHBP relationship, and in fact it is a scientifically interesting question. Yet, it would be dishonest to look at CRHBP levels and say anything about how the organism will behave.
It's equally dishonest to say that other differences in brain structure are meaningful beyond the fact that they are differences in structure. Until you show some effect of these differences, don't jump to conclusions. My complaint about the article was that they made claims about intelligence based on structural similarities. Bad PR person. Bad.
---
[1] Corticotropin Releasing Hormone
The nature nurture debate has raged for centuries in respect to intelligence. Ever since Darwin's cousin Francis Galton proposed his theories of hereditary intelligence.
At first glance, it seems neat that these guys did a study on twins. You might think, "wow, what a great approach." You'd be right, then you'd hopefully realize that other people had these ideas in the past, and they did the right studies and came to the right conclusions.
When Thomas Bouchard was in charge of the Minnesota Twin Studies, he and his colleagues compared twins raised together and apart.
Bouchard and colleagues tested 56 pairs of monozygotic (identical) twins who had been raised apart and compared them to hundreds of identical twins who had been reared together. The twins were tested on dozens of capacities. This approach allowed them to examine spatial ability, verbal ability, mathematical ability, personality, etc. Rather than the antiquated "g" or general intelligence factor.
So, here's a study with many more subjects, much better comparisons, and more detailed data. Here's the cool part: if you correlate the scores of twins on the Wechsler Adult Intelligence Scale (WAIS) (the standard IQ test in the US) and Ravens Progressive Matrices (An allegedly culturally unbiased test of reasoning skills), you see a 65% (WAIS) or 50% (Raven's) agreement among scores of twins raised together, and among twins raised apart, the WAIS scores are about 35% in agreement with nearly the same 50% agreement on Raven's scores.
What does this mean? Well, it means that even in the best cases, IQ scores are about 50% hereditary. In addition, the big drop off between WAIS scores indicates that environment has an important role in intelligence. Given that most of the twins were raised in similar middle class families (who adopt kids), the estimates of the role of environment are probably inflated.
That being said, the study mentioned in the story focuses primarily on brain structures. This pisses me off. One of the lamest things about current cognitive neuroscience is the common misunderstanding of the difference between structure and process. Given the little we know about how the brain works (yeah, we're getting good info about the molecular level, and we know in general what larger regions like Broca's area are responsible for, but we have woefully little info about how these structures actually work), idiot PR people take brain findings and blow them out of proportion.
The fact is, if you want to talk about intelligence, you have to talk about behavioral measurements. Looking at structures can, at best, tell you if something is wrong. So, yeah big deal, major structures are heritable, given the fact that environment has been shown in better studies to play a key role in intelligence, we shouldn't rely too much on these findings.
Bottom line: Yes of course nature plays an important role in intelligence. You'd have to be an idiot not to realize that. This study is not groundbreaking. Moreover, the headline of the article is sensationalistic and only a half-truth. Environment does play a role in intelligence. My god, meet a person who ate lead paint as a kid and you'll realize that.
Actually, there is little reason to doubt the NCTA data. Although it's called a "survey," their findings are better thought of as sales reports. The NCTA is an industry association, and will certainly try to put a spin on their findings (e.g., "Consumers' strong response to digital cable services, in spite of difficult times, confirms the excellent value of these new services"), but they'll still report accurate sales figures.
The RIAA (boo) would LOVE it if album sales plummeted at the same time that Napster was taking off. Yet, that wasn't the case (although CD singles did suffer a drop last year, the increased sales of full albums was significantly greater), and the RIAA reported the numbers correctly. Although they put their spin on it, ("Look at the drop in Singles sales"), they reported valid sales figures.
It looks like the first article was a guy trying to create a news story when there really wasn't one. Sure, people will switch from high bandwidth to low bandwidth, but if the general trend overshadows it, we end up with a very different story.
The newer story, which is just a rehash of an NCTA press release says nothing about the people who are installing (or uninstalling) cable modems. It talks about sales trends. This doesn't negate the other story, it just indicates that it isn't all that big a trend.
It's the equivalent of C|Net reporting that I just bought an AMD processor, so Intel had better watch out. Who cares about me? If there are 50,000 people like me, then you notice.
So, yes there are people switching. But no, it doesn't seem to be affecting the industry. Two separate stories, no conflict.
unless you charge money and have a 2 drink minimum.
One interesting issue that these guys confronted, and that others must confront as well is whether or not to attempt to imitate the way humans (or other animals) perform or to simply engineer a solution that works.
As an example, imagine that you are creating a computer vision system. If you want to be able to have it measure aesthetic quality you may wish to rely on our knowledge of how humans judge visual pleasantness. Alternatively, if you are interested in the detection of faint patterns, you may find that humans don't do it as well as you'd like, and that an alternative approach is better.
The discussion of the Honda Robot compared to this one is sort of the same thing. Just because someone else made a robot that doesn't do things like people do them, doesn't mean the robot is not as good. If the goals are to simulate people, then yes this robot is better. If however, the goals are different, then this may not be better (e.g., it can't carry as much weight).
When I tried to access the story I got an error message:
/site-stories/2001/1025.errorhandling.html.
Proxy Error
The proxy server received an invalid response from an upstream server.
The proxy server could not handle the request GET
Reason: Could not connect to remote machine: Connection refused
Additionally, a 404 Not Found error was encountered while trying to use an ErrorDocument to handle the request.
I used proxomitron to set my user-agent to some nonsense (it was "!b" without the quotes) and went to msn.com and got the standard "upgrade to IE" message. Then I went to msn.com.br and got this message:
Microsoft VBScript runtime error '800a000d'
Type mismatch: '[string: ""]'
/include/browser.asp, line 45
I figured I'd be able to find out a bit about their filter (written in vbscript) by altering the input a bit. So then I changed my user agent to "!b(msie)" and got
Microsoft VBScript runtime error '800a000d'
Type mismatch: '[string: "e"]'
/include/browser.asp, line 45
My hypothesis was that it looked at something right before the ")". So then I tried "!b(msie;msie;windows)" and got
Nota ao usuário
Infelizmente seu navegador não é compatível com essa versão, portanto não será possível visualizar todo o conteúdo que disponibilizamos especialmente para você. Para ver corretamente o novo portal MSN, baixe agora a versão mais recente do Internet Explorer ou o novo MSN Explorer. É grátis!
Se você acha que recebeu essa mensagem por engano, por favor confirme que seu navegador possui capacidade de visualização instalada. (No Internet Explorer, acesso pelo menu Visualizar | Opções).
© 2000 Microsoft Corporation. Todos os direitos reservados. Termos de Uso.
which looks like it says something about having to download IE and that it's free. So I changed back to !b and then to !b(msie) and kept getting that final message. Looks like they must have changed the script or something. Just as it was getting good.
Any thoughts?
Overall this looks like a nice gift to the big movie and record companies at the cost of consumers fair use and other rights. However, section 103(b) states:
PERSONAL TIME-SHIFTING COPIES CANNOT BE BLOCKED. -- No person may apply a security measure that uses a certified security technology to prevent a lawful recipient from making a personal copy for time-shifting purposes of programming at the time it is lawfully performed, on an over-the-air broadcast, non-premium cable channel, or non-premium satellite channel, by a television broadcast station (as defined in section 122(j)(5)(A) of title 17, United States Code), a cable system (as defined in section 111(f) of such title), or a satellite carrier (as defined in section 119(d)(6) of such title.)
Which is a good first step. But this arguably would let HBO or any other premium service (do other packaged channels count as premium services) prevent me from recording The Sopranos or any of their America Undercover series...
This seems like an odd thing to do. So, for example, when the Sopranos goes into syndication I'll be able to record it off of UPN but not HBO.
I'm a teacher. I use recorded materials to teach my students. I'm allowed to do so under fair use. This law could make me into a criminal. That sucks.
The sickening part of this is that the RIAA is essentially using the deaths of thousands of people as cover to expand their power.
No matter how bad copyright infringment may be, it is much less morally repulsive than using a tragedy like this to advance a political agenda. The RIAA has stooped to a new low.
PGP had a few of strikes against it:
A. Little perceived need by the masses
B. Hassle to use
and more recently
C. Government rumblings
A. could be dealt with by some good old FUD. I've always been amazed that NAI and others have resisted the evil urge to play on naive users' fears of "hackers." Come on, companies with lame IDS and Firewall products have been playing the fear card for a while. Imagine how effective a campaign would be if the product were actually good... (Not that I'm a fan of these tactics).
B. is a more difficult problem. Although the product has come a long way since the old DOS version with it's confusing options, it has a way to go to acheive true ease of use. People don't necessarily "get it." I'm not a huge fan of dumbing down interfaces, but a real simple set of wizards that handled all the stages of key creation and software integration would be helpful. Plug-ins for email are good, but a deal with MS or Eudora to bundle it would be better. Plug-in with ICQ is good but a bit clumsy at times. Maybe playing up the Envelope metaphor in email programs would be better... Also, encouraging users to get their email contacts to install the freeware version would be great. Maybe, a window that popped up when people tried to send an encrypted email to a person whose key isn't know. The window could mention the problem, and offer to send the recipient an email with a link to the freeware (or perhaps a free "reader" that allowed for key creation and email integration).
With C. the issue is just a big hassle. At some point you'd hope the Gov't would realize that restricting strong encryption will have no effect on criminals, only business and home users.
GPS is cool, and future emergency GPS location will be cool too.
What concerns me is that the phone relies on the network to get restaurant and driving directions. My Visor with appropriate GPS springboard and something like Vindigo allows me to find nearby stuff using a local database (i.e., on the client). My SprintPCS phone has a feature that lets me set an alarm at a specific time. However, this feature only works when my phone is in the phone network. That is, even though my phone should be perfectly capable of keeping time, it needs a digital PCS signal to actually do so.
If I use this phone to do anything with the GPS aside from finding my own location, I'm forced to send the PCS network my location. While I really don't care that much about the privacy of my whereabouts, I certainly don't need anyone sending me ads targeted to my present location.
I'm not sure, but I think that may be part of the Word XP EULA
The long audio delays can come from two different sources. First off, there is the amount of time it takes for the signal to travel from the phone to the tv studio. Second, it typically takes a bit of time for the video decompression to occur. Rather than lose AV syncro, the audio is actually delayed to be played with the appropriate video.
Here's where it gets interesting: When someone asks you a question in real life, they rely on the amount of hesitation on your part to assess how honest you're being. With these things, you always appear to be hesitating before speaking. Imagine a debate or discussion with someone in studio and someone else in the field. You'd have an asymmetry in how honest these people are perceived.
When he demonstrated the orginal computer mouse, Doug Engelbart used a one handed chorded keyboard for his left hand at the same time. The system that they developed allowed users to type all characters using just the chords on the left hand, but they still had a normal keyboard... Why? Because the standard keyboard was pretty efficient. So, what did they use the chorded keyboard for? Functions. Copy, paste, print, etc were all associated with chord combinations. This way, you move your hands away from the keyboard to use the mouse and chord when you want to do functional stuff. But, when you're typing you put both hands on the keyboard and pound away.
The learning curve for a two-handed chorded keyboard is sort of long, but not rediculous. Learning a new layout (e.g., Dvorak) on a standard keyboard takes about 20 hours (e.g., 1 hour a day every weekday for a month). Chording is a learned skill, which can be acquired relatively quickly. I'm guessing it's faster if you have piano skills...
They don't even compare to the utter crappiness that was SimEarth...
I think that, more importantly, these new domain names don't add anything. .biz provides no useful semantic information. It's not like someone would say, "hmm, I'm not looking for a (.)company, I'm looking for a (.)business." Moreover, since the target market is companies, they already are likely to have .com names, thus making this redundant.
.com2 and .com3 -- nothing useful.
.sex or .personal or .protest would be useful. These are categories that aren't simply mirrors of existing TLDs.
This is the equivalent of adding
.kids or
One challenge in all of this is that our elected representatives don't want to be seen as being soft on terrorism.
What happened was terrible, steps must be taken to protect us, but for some authorities to take advantage of the pain and suffering of others to pass through their pet legislation is sickening.
Don't try to cripple crypto, it won't do anything except erode the rights of people who aren't terrorists. Real terrorists will use non-crippled systems, while those who use off the shelf groupware or banking software will be stuck with the crappy stuff.
Don't let the authorities conduct wiretaps (phone or computer) without warrants. If someone is really a suspect, then you should be able to get a warrant.
But sure, increase security checks at airports, designate more money to terrorism prevention, take appropriate action against those who did this. Just don't try to erode our rights while using this as a smokescreen.
I guess I'm lucky, because my representative, Lynn Rivers seems to get it...
which is why Katz said
two of the biggest buildings on the planet
and not
the two biggest buildings on the planet