IRC is simply stated, dying. It is too insecure to begin with, plus people satisfy their needs with aol, yahoo and msn messenger; despite the fact that they install GATOR on people's computers.
Agreed on the security thing.
However, I think IRC is also dying because things like identd make it difficult for a lot of users, including those behind firewalls, to find a good irc server.
You're making it harder for Windows users to connect to your irc server when you know it can't be trusted to give you any sort of identifying information about people who connect from a Windows box.
Make it optional, so UNIX admins can run identd for the reason you mentioned.
I would be willing to believe that if we were still back in the early days of irc, but nowadays most everyone has their Windows box or Mac connected straight up to the network and relatively few of the script kiddies I have met on irc know how to use unix.
Now, which do you think is the more likely scenario: All the l-users here that have never run an IRC server and are taking out of their ass know best, or that hundreds of experienced server and network ops know what they're doing and require identd for a reason?
I'll choose the third scenario: The hundreds of experienced server and network ops are living in the last decade and have completely forgotten that the case for a giant portion of internet users is that they are on single-user machines but sharing IP addresses with other users because their network or ISP has more users than IP addresses and is using network address translation.
Thinking that identd is useful for security in this situation seems about as smart as assuming that anonymous ftp logs really do keep an accurate list of the e-mail addresses of people who have connected to the server.
I think we can now conclude with complete certanity that video games aren't just marketed at pimply faced hypercephalic adolescent and pre-adolescent boys. Nope, nosiree. They're also apparently marketed at any male creature thing that has never quite gotten used to the appearance of a female midriff. . .
Whether this is right or wrong depends on your fundamental belief of what a corporation's primary goal is: maximizing profits, or benefitting the world.
I think there are two conflicting goals coming from two separate parties - the first is the corporation itself, and it's primary goal is clearly to maximize profits.
On the other hand, the decision-makers in the corporation are people, and one would hope that they as individuals are primarily interested in benefitting the world.
Since the corporation is an abstract concept that is meant to serve humans' purposes, one would think that the human goal would win out. Of course, this gets into a really hairy subject of whether we should be concerned with things which don't affect us and whether profits are more important than morals.
I wish that after several thousand years of recorded human history, philosophical thinking, and civilization advancement that this question wouldn't be so hairy for so many people. . . the correct answer seems obvious to me.
I have a lot of friends who are international students, & those questions are pretty indicative of how closely INS is watching/controlling them.
If you're an international student, you practically have to fill out a form and send it to INS just to take a shit, and God Save You if it takes longer than 5 minutes to finish your 'business'.
My school has had a 15% drop in the number of international students over the past year from acts like SEVIS.
Id say asphyxiation and freezing to death wouldn't be a very big problem.
To freeze to death, you'd need to have some matter around to absorb the heat off your body. That ain't going to happen all that quickly in space.
For asphyxiation, you would have the air sucked out of your lungs, but you would probably still have enough oxygen in your bloodstream to keep you going for a minute or two.
I think someting related to part of your body bursting would be the worst problem.
So you propose that if you can do something evil and profit, it is ethical to do so if you know full well that if you don't do it, someone else will.
Let's extrapolate that to its logical extreme.
Let's say I offered to pay you $500,000 to shoot someone in the head, and you knew that if you didn't do it, I'd go to someone who you know doesn't really care about having blood on his/her hands and would do it if you wouldn't. Is it then okay to shoot our poor captive in the head?
How about if I were going to go shoot someone in the head, and asked you to sell me the gun. You know full well what my motive is. Would you really sell me the gun and ammo and defend yourself by saying that you had done nothing wrong, and were just engaging in commerce?
Blame capitalism all you want. Capitalism is just an ideal that has some shitty implications. If someone helps to bring them about, we blame the person, not the ideal. If a neo-nazi went around burning synagogues, we wouldn't say, "Oh, don't blame him, blame neo-nazism!" If I joined a cilque whose culture was such that the best way to gain respect is to spraypaint storefronts, I would still be to blame for spraypainting the storefronts if I did it.
Sorry, I don't think your ethics reasoning holds water, either.
On top of that, I don't think that stippling really applies to the gaming situation as a technique - they are trying to generate 3-D images given complex data sets based on x-ray transmission within the body. I have a feeling that generates points of information, and the old technique would be to either use voxels or translate the data into polygon information.
Beside the fact that modelling such information for a game would be ludicrously time-consuming, I fail to see why this technique could offer an advantage to the display of 3-D graphics in a gamin sense - and I doubt it's actually faster in terms of the amount of time it takes to get mathematical data translated onto a cathode ray tube. All the article says is that it's faster than previous techniques in medical imaging. The article doesn't say what those techniques are, but since I can't for the life of me see how a CT scanner would get polygon information out of x-rays, I think we can all be sure that they aren't at all similar to what the Quake 3 engine is doing.
Something like this can easily be avoided by all office programs being able to easily comply with AT LEAST one open format.
But they all already comply to at least one open format that I know of - namely, rich text format.
To avoid the problem, you'd have to either have all office software agree on a common native format, or have all office software able to read/write the data for all other office software.
I'm not saying that we should sue slashdot. I'm not saying that Slashdot shouldn't post articles. I'm saying that getting Slashdot to invest in some method to curb the effects of the slashdot effect on smaller servers by complaining in slashdot discussions isn't going to get a blamed thing done, and saying it using a common conversational tool called exaggeration. For those who don't know, exaggeration is a way of emphasising a point by inflating it to ridiculous proportions.
Methinks the only way to get Slashdot's editors to stop pulling this shit is to get a class action lawsuit into action so OSDN gets on/.'s butt over the issue.
We're still working on teaching Taco to communicate in human language. Getting him to understand more difficult concepts like ethics and not being a jerkbutt is going to take longer. Convincing the Slashdot team to figure out something on their own is a pipe dream.
Alabama requires stickers containing text very similar to the disclaimer I put in that parody to be put in every science textbook. Something about evolution being one of several equally valid theories.
In other news, Alabama has passed new legislation requiring all history textbooks to include the following disclaimer before discussing the Apollo program:
Humans landing on the moon is just one of many equally valid theories concerning the video footage and rock samples resulting from the Apollo program. The moon landing is a controversial theory. Instructional material associated with controversy should be approached with an open mind, studied carefully, and critically considered."
Alabama history teachers are encouraged to devote class time to discussing other theories about the Apollo mission.
Seeing how adults treat sex and violence, it makes complete sense to me.
I'm going to totally ignore what adults have told me through my teenage years. Actions speak louder than words, and I've seen a lot of adults acting in my life.
The distinction is simple: Violence is a perfectly reasonable and effective way to solve your problems with other people. Sex is at best a guilty pleasure.
I do not believe that the goal of having computers in classrooms is to have brighter children. Not the goal, but a very oft-cited goal among schools I have been to and worked with. Heck, I even went to one of the schools that was involved with the Microsoft/Toshiba Anytime, Anywhere Learning project. The whole idea of it is that computer technology has some magical ability to help teach kids. We had laptops assigned to every kid, we had the entire school wired for networking, and were jacked into the 'net all through class.
Of course, the whole thing was a complete bomb. Teachers didn't know how to use the Internet. Kids spent more time in Yahoo chatrooms than following whatever website the teacher wanted them to be following along in. E-textbooks need to die. And, of course, kids learned very quickly about all the webistes out there that catalog thousands of papers for them to plagarise.
Not to say that kids don't learn from the 'net. The 'net got me involved in the demoscene and open source, which taught me to program. Various political sites gave me access to information about the world that I never would have touched any other time. I can't begin to say how liberating some BBSes were when I was 11, and when Prodigy gave its users Internet access, it gave me oppotrunites to think that I never would have otherwise had growing up in a town of 7,000.
That, and I found a lot of porn. And jerked off. I have a hard time saying that was unhealthy, though, especially from the studies I've read giving evidence that boys who are exposed to a lot of porn as adolescents have a tendency to be much more well-adjusted sexually, more likely to use condoms, &c.
Robotics is my hobby, & I've read him, and think he's being a bit kooky in this article.
1) Computer Vision != human vision. Trying to compare two too closely is fallacious. We know that many computer techniques, such as laplacian-of-a-gaussian edge detection, are really quite similar to the human way of doing things. Other computer vision techniques, such as active vision, are clearly wrong. Some seem on the right track, but obviously don't do things in the same way humans do. Passive stereo, for example, is frequently used to get specific depth calculations. Research in human perception shows that the human visual system appears to do no such thing, and is only capable of extracting relative depth information. Because the methods being used in human vision are frequently different from those used in computer vision, we can't use the number of MIPS required to get a given performance on a given computer vision system to determine the MIPS of a human visual cortex.
2) Vision != whole brain activity. Ok, so we get past the first and figure out the equivalent computer performance to get all the same information in the same ways as the visual cortex, so we can claim to make a MIPS calculation. How can we extrapolate that to the whole brain? The neural networks in the visual cortex aren't even structurd the same way as many other parts of the brain.
3) Plausibility != proof. Near the beginning of the article, the author says that the MIPS of Deep Blue is 1/30 of his MIPS estimation for the brain. He then says that it's plausible that Kasparov uses 1/30th his cognitive power in playing chess, and calls that proof for his argument. I see two things wrong with tha. First, a more scientific mind wouldn't call it proof so much as a lack of a certain disproof. Second, the very language of the argument shows that he really doesn't understand brains at all. The idea of using foo fraction of your brain is one that most the people I know who study neuroscience have been trying to convince people is inherently fallacious since time began. A computer scientist can spend some time studying the theory and technique of ANN's and pretty quickly understand why this is the case.
4) Total computer design != brain design. Beyond the megahertz myth, there is the MIPS myth. Things like BUS bandwidth and memory speeds do play a role in computer performance. Furthermore, the brain is so integrated that it's hard to see how the design of computers, especially Von Neumann computers, wouldn't break the entire comparison down. For one, every part of the brain takes part in processing, and the computational parts are part of the BUS. Everything operates at the same speed. Even the parts having to do with memory operate at the same speed as everything else. Speaking of memory, it's important to remember that neural memory is associative memory, not sequential memory. How do you apply a MIPS rating to processing when it is built into the data retrieval process? MIPS ratings ignore hard drive reads - are we going to just ignore long-term memory retrieval on brains, too? If we do include long-term memory retrieval in the calculation for humans, it doesn't seem correct to not include that on a computer, in which case MIPS is obviously an unworkable metric.
While we're at it, a quick look at the references on his site shows just how little he's done his research on stuff other than computers that I can find - he read an IEEE article on nanoelectronics. I'm not even going to count him referencing himself.
That's not the entire brain, that's individual neurons.
Don't think of the brain as one really wide silicon wafer running at 40hz. If you absolutely have to try and make the comparison, though, think of the brain as a couple billion little tiny CPU's, each running at 40hz.
In that sense, your first statement is misleading. It's like implying that a beowulf cluster of 100 P100's is slower than a PIII-733 on the basis that the individual processors in the cluster run at a slower clock rate. And don't even get me in to the megahertz misconception;-)
Slashdot Mirror
IRC is simply stated, dying. It is too insecure to begin with, plus people satisfy their needs with aol, yahoo and msn messenger; despite the fact that they install GATOR on people's computers.
Agreed on the security thing.
However, I think IRC is also dying because things like identd make it difficult for a lot of users, including those behind firewalls, to find a good irc server.
You're making it harder for Windows users to connect to your irc server when you know it can't be trusted to give you any sort of identifying information about people who connect from a Windows box.
Make it optional, so UNIX admins can run identd for the reason you mentioned.
I would be willing to believe that if we were still back in the early days of irc, but nowadays most everyone has their Windows box or Mac connected straight up to the network and relatively few of the script kiddies I have met on irc know how to use unix.
Now, which do you think is the more likely scenario: All the l-users here that have never run an IRC server and are taking out of their ass know best, or that hundreds of experienced server and network ops know what they're doing and require identd for a reason?
I'll choose the third scenario: The hundreds of experienced server and network ops are living in the last decade and have completely forgotten that the case for a giant portion of internet users is that they are on single-user machines but sharing IP addresses with other users because their network or ISP has more users than IP addresses and is using network address translation.
Thinking that identd is useful for security in this situation seems about as smart as assuming that anonymous ftp logs really do keep an accurate list of the e-mail addresses of people who have connected to the server.
But Intel and Microsoft told me if I get a snazzier PC the Internet will run faster!
I think we can now conclude with complete certanity that video games aren't just marketed at pimply faced hypercephalic adolescent and pre-adolescent boys. Nope, nosiree. They're also apparently marketed at any male creature thing that has never quite gotten used to the appearance of a female midriff. . .
Whether this is right or wrong depends on your fundamental belief of what a corporation's primary goal is: maximizing profits, or benefitting the world.
I think there are two conflicting goals coming from two separate parties - the first is the corporation itself, and it's primary goal is clearly to maximize profits.
On the other hand, the decision-makers in the corporation are people, and one would hope that they as individuals are primarily interested in benefitting the world.
Since the corporation is an abstract concept that is meant to serve humans' purposes, one would think that the human goal would win out. Of course, this gets into a really hairy subject of whether we should be concerned with things which don't affect us and whether profits are more important than morals.
I wish that after several thousand years of recorded human history, philosophical thinking, and civilization advancement that this question wouldn't be so hairy for so many people. . . the correct answer seems obvious to me.
I never made any mention of legal/illegal.
When I say 'is it okay' and 'defend yourself', I mean in the moral sense.
Besides, in this particular case I think the arguments are equally valid (for different reasons) in both law and ethics.
a lallie is another word for a 'dental dam'. It's a little latex square you use to avoid exchange of body fluids during cunnilinguis.
I have a lot of friends who are international students, & those questions are pretty indicative of how closely INS is watching/controlling them.
If you're an international student, you practically have to fill out a form and send it to INS just to take a shit, and God Save You if it takes longer than 5 minutes to finish your 'business'.
My school has had a 15% drop in the number of international students over the past year from acts like SEVIS.
Id say asphyxiation and freezing to death wouldn't be a very big problem.
To freeze to death, you'd need to have some matter around to absorb the heat off your body. That ain't going to happen all that quickly in space.
For asphyxiation, you would have the air sucked out of your lungs, but you would probably still have enough oxygen in your bloodstream to keep you going for a minute or two.
I think someting related to part of your body bursting would be the worst problem.
So you propose that if you can do something evil and profit, it is ethical to do so if you know full well that if you don't do it, someone else will.
Let's extrapolate that to its logical extreme.
Let's say I offered to pay you $500,000 to shoot someone in the head, and you knew that if you didn't do it, I'd go to someone who you know doesn't really care about having blood on his/her hands and would do it if you wouldn't. Is it then okay to shoot our poor captive in the head?
How about if I were going to go shoot someone in the head, and asked you to sell me the gun. You know full well what my motive is. Would you really sell me the gun and ammo and defend yourself by saying that you had done nothing wrong, and were just engaging in commerce?
Blame capitalism all you want. Capitalism is just an ideal that has some shitty implications. If someone helps to bring them about, we blame the person, not the ideal. If a neo-nazi went around burning synagogues, we wouldn't say, "Oh, don't blame him, blame neo-nazism!" If I joined a cilque whose culture was such that the best way to gain respect is to spraypaint storefronts, I would still be to blame for spraypainting the storefronts if I did it.
Sorry, I don't think your ethics reasoning holds water, either.
Everyone needs condoms.
Even if you think you don't need condoms, you do.
You can fit a gallon of water in one - great if you're planning on going on a desert trip.
They don't break too easily if you don't put too much water in 'em, so they make nice re-usable water balloons.
You can cut 'em down the side and use them as lallies.
If smuggling's your thing, youc an always pack them with coke and swallow them!
blow a french tickler up a little bit and it looks like a mouse. Give it to your cat as a toy.
I tell ya, condoms are nature's duct tape!
On top of that, I don't think that stippling really applies to the gaming situation as a technique - they are trying to generate 3-D images given complex data sets based on x-ray transmission within the body. I have a feeling that generates points of information, and the old technique would be to either use voxels or translate the data into polygon information.
Beside the fact that modelling such information for a game would be ludicrously time-consuming, I fail to see why this technique could offer an advantage to the display of 3-D graphics in a gamin sense - and I doubt it's actually faster in terms of the amount of time it takes to get mathematical data translated onto a cathode ray tube. All the article says is that it's faster than previous techniques in medical imaging. The article doesn't say what those techniques are, but since I can't for the life of me see how a CT scanner would get polygon information out of x-rays, I think we can all be sure that they aren't at all similar to what the Quake 3 engine is doing.
Something like this can easily be avoided by all office programs being able to easily comply with AT LEAST one open format.
But they all already comply to at least one open format that I know of - namely, rich text format.
To avoid the problem, you'd have to either have all office software agree on a common native format, or have all office software able to read/write the data for all other office software.
I'm not saying that we should sue slashdot. I'm not saying that Slashdot shouldn't post articles. I'm saying that getting Slashdot to invest in some method to curb the effects of the slashdot effect on smaller servers by complaining in slashdot discussions isn't going to get a blamed thing done, and saying it using a common conversational tool called exaggeration. For those who don't know, exaggeration is a way of emphasising a point by inflating it to ridiculous proportions.
Methinks the only way to get Slashdot's editors to stop pulling this shit is to get a class action lawsuit into action so OSDN gets on /.'s butt over the issue.
We're still working on teaching Taco to communicate in human language. Getting him to understand more difficult concepts like ethics and not being a jerkbutt is going to take longer. Convincing the Slashdot team to figure out something on their own is a pipe dream.
Alabama requires stickers containing text very similar to the disclaimer I put in that parody to be put in every science textbook. Something about evolution being one of several equally valid theories.
In other news, Alabama has passed new legislation requiring all history textbooks to include the following disclaimer before discussing the Apollo program:
Humans landing on the moon is just one of many equally valid theories concerning the video footage and rock samples resulting from the Apollo program. The moon landing is a controversial theory. Instructional material associated with controversy should be approached with an open mind, studied carefully, and critically considered."
Alabama history teachers are encouraged to devote class time to discussing other theories about the Apollo mission.
On an unrelated note, I got a spam today featuring what appears to be a 16-year-old and an octagenarian having sex.
Seeing how adults treat sex and violence, it makes complete sense to me.
I'm going to totally ignore what adults have told me through my teenage years. Actions speak louder than words, and I've seen a lot of adults acting in my life.
The distinction is simple: Violence is a perfectly reasonable and effective way to solve your problems with other people. Sex is at best a guilty pleasure.
I do not believe that the goal of having computers in classrooms is to have brighter children.
Not the goal, but a very oft-cited goal among schools I have been to and worked with. Heck, I even went to one of the schools that was involved with the Microsoft/Toshiba Anytime, Anywhere Learning project. The whole idea of it is that computer technology has some magical ability to help teach kids. We had laptops assigned to every kid, we had the entire school wired for networking, and were jacked into the 'net all through class.
Of course, the whole thing was a complete bomb. Teachers didn't know how to use the Internet. Kids spent more time in Yahoo chatrooms than following whatever website the teacher wanted them to be following along in. E-textbooks need to die. And, of course, kids learned very quickly about all the webistes out there that catalog thousands of papers for them to plagarise.
Not to say that kids don't learn from the 'net. The 'net got me involved in the demoscene and open source, which taught me to program. Various political sites gave me access to information about the world that I never would have touched any other time. I can't begin to say how liberating some BBSes were when I was 11, and when Prodigy gave its users Internet access, it gave me oppotrunites to think that I never would have otherwise had growing up in a town of 7,000.
That, and I found a lot of porn. And jerked off. I have a hard time saying that was unhealthy, though, especially from the studies I've read giving evidence that boys who are exposed to a lot of porn as adolescents have a tendency to be much more well-adjusted sexually, more likely to use condoms, &c.
Robotics is my hobby, & I've read him, and think he's being a bit kooky in this article.
1) Computer Vision != human vision. Trying to compare two too closely is fallacious. We know that many computer techniques, such as laplacian-of-a-gaussian edge detection, are really quite similar to the human way of doing things. Other computer vision techniques, such as active vision, are clearly wrong. Some seem on the right track, but obviously don't do things in the same way humans do. Passive stereo, for example, is frequently used to get specific depth calculations. Research in human perception shows that the human visual system appears to do no such thing, and is only capable of extracting relative depth information. Because the methods being used in human vision are frequently different from those used in computer vision, we can't use the number of MIPS required to get a given performance on a given computer vision system to determine the MIPS of a human visual cortex.
2) Vision != whole brain activity. Ok, so we get past the first and figure out the equivalent computer performance to get all the same information in the same ways as the visual cortex, so we can claim to make a MIPS calculation. How can we extrapolate that to the whole brain? The neural networks in the visual cortex aren't even structurd the same way as many other parts of the brain.
3) Plausibility != proof. Near the beginning of the article, the author says that the MIPS of Deep Blue is 1/30 of his MIPS estimation for the brain. He then says that it's plausible that Kasparov uses 1/30th his cognitive power in playing chess, and calls that proof for his argument. I see two things wrong with tha. First, a more scientific mind wouldn't call it proof so much as a lack of a certain disproof. Second, the very language of the argument shows that he really doesn't understand brains at all. The idea of using foo fraction of your brain is one that most the people I know who study neuroscience have been trying to convince people is inherently fallacious since time began. A computer scientist can spend some time studying the theory and technique of ANN's and pretty quickly understand why this is the case.
4) Total computer design != brain design. Beyond the megahertz myth, there is the MIPS myth. Things like BUS bandwidth and memory speeds do play a role in computer performance. Furthermore, the brain is so integrated that it's hard to see how the design of computers, especially Von Neumann computers, wouldn't break the entire comparison down. For one, every part of the brain takes part in processing, and the computational parts are part of the BUS. Everything operates at the same speed. Even the parts having to do with memory operate at the same speed as everything else. Speaking of memory, it's important to remember that neural memory is associative memory, not sequential memory. How do you apply a MIPS rating to processing when it is built into the data retrieval process? MIPS ratings ignore hard drive reads - are we going to just ignore long-term memory retrieval on brains, too? If we do include long-term memory retrieval in the calculation for humans, it doesn't seem correct to not include that on a computer, in which case MIPS is obviously an unworkable metric.
While we're at it, a quick look at the references on his site shows just how little he's done his research on stuff other than computers that I can find - he read an IEEE article on nanoelectronics. I'm not even going to count him referencing himself.
Kind of like how you can't give a hyperlink to a 2,000lb bandwidth-eating beast and not scare the shit out of some sysadmin?
That's not the entire brain, that's individual neurons.
;-)
Don't think of the brain as one really wide silicon wafer running at 40hz. If you absolutely have to try and make the comparison, though, think of the brain as a couple billion little tiny CPU's, each running at 40hz.
In that sense, your first statement is misleading. It's like implying that a beowulf cluster of 100 P100's is slower than a PIII-733 on the basis that the individual processors in the cluster run at a slower clock rate. And don't even get me in to the megahertz misconception