> Some would say only God would allow a transcendental number raise to another transcendental number, raise to an imaginary number to come out so nice as just -1!
Would that be the same God who designed the relative sizes of penises and anuses?
Your method of "reasoning" could lead to some very alarming conclusions if you applied it the same to every observation.
> Something so incredibly nefarious like a gross invasion of civil liberties and free speech somehow gets segued into yet another blathering rant about the RIAA/MIAA?
Probably the real motive of centralizing the net is to make it a corporate playground rather than a citizens' playground. The bit about "security" is just an excuse for doing it.
One of my first posts on Slashdot, way back when, was "I fear for the internet". It lets us step on the toes of too many vested interests. Unfettered communication between citizens hasn't been the norm in 'free' countries, and some parties in those countries are deciding that they don't like it.
> We have been working on institutions in India and they are heavily leaning towards adopting Linux. Microsoft sales teams have been bending over backwards to prevent them from taking the plunge. The recent donations of funds as well as the offer to share the code all amount to last ditch attempts to keep Linux out.
Yes, The Register had an intersting take on billg's recent coin-scattering trip to India.
> The content of this book is spot on. Just as important is the layout of the book and Sobell has nailed the best way to present this sort of text. It is quick and easy to find exactly what you need.
FWIW, Sobel has a new book A Practical Guide to Red Hat Linux 8 in press, supposedly out this month. Shame about the distro-/version- specific presentation, but the TOC makes it look like it will be a good book for anyone who matches the relevant distro/version.
> I'm not sure you'd get a lot of improvement from the technique though - bots would tend to play well with bots this way, unless you somehow managed to get a fitness function that reflected playing with humans. You could also invite a lot of human beta testers and evolve them based on survival rates.
You could bootstrap the system by having it play itself. That doesn't always result in the desired "arms race" of continual improvements, but people are looking at ways of inducing arms races.
If you could ever get it up to some basic level of performance (say, about what the typical game AI does now), you could enter a phase of distributed learning where thousands of people who had bought the game downloaded the latest AI, played it a time or two, and uploaded the results. That way no individual would have to play through a lot of games with a stupid AI as it explored options to improve its behavior against humans. Presumably if you downloaded the latest AI every week you would get a more competitive opponent every time.
> Some algorithm that noticed patterns in actual gameplay would probably induce cooler behaviour - something in the lines of... These are both things real players do, and it'd help them avoid falling in the same traps every time.
Yeah, those are hard problems for AI if you try to generalize the behavior. I wish more AI researchers were working on this kind of stuff instead of the toy problems that so many of them are still on. Several AI researchers have suggested games as the best research domain for acheiving "human level" AI.
> Naturally, the AI has the shortest time frame in the software engineering, but there is no reason it should remain stagnent across the future patches.
Another problem is that lots of games are just engines that support an 'official' dataset plus whatever modpacks the players care to come up with, but even the cheatAI that ships with the game won't work worth a damn on the modpacks.
I hope in the future machine learning methods can help with both of these problems. I.e., a couple of months before release when the code is fairly stable and the graphics are in production, turn on the old Beowulf cluster and let reinforcement learning or an evolutionary algorithm train a good AI for the game. As for modpacks, the vendors could support something like sourceforge, where gamers could upload their modpacks and have the Beowulf cluster automagically re-tune the AI to work right with them.
And of course, the machine learning could continue in the background for as long as people were interested in the game, allowing them to download "new improved" AIs every few months.
> Please pass this information on to as many people as possible. Let's make a child's dream come true before its too late.
Be sure to mention how sad you were when you heard that he was dying from inrectocephalosis.
Re: The two skills of writing
on
Prey
·
· Score: 1
> I've always thought that there are two very distinct skill involved in writing. The first is storytelling, the ability to weave a yarn that is enthralling, touching, satisfying, etc. The second is skill with the language, the ability to create a rich imaginary world, enticing to all the senses, with only the written word.
> There are some writers who clearly excel at both. The first that comes to mind is Pat Conroy.
> Indeed, Spielberg did a fantastic job with Jurassic Park.
Feh. Superb dinosaurs, tolerable plot, third-rate actors, execrable script. Never have so few characters spoken so many bad lines in so few movies (the sequel was actually worse!). I've heard better dialog in porno flicks.
>...but 'convincing and specific' proof with 'photo graphic evidence' and maybe 'a diary or something' is needed to refute the existence of a hypothetical fairy tale creature.
I'm curious as to what sort of photograph would show that fairy tale creatures don't exist. Maybe a picture of the whole universe, that you could go through and show that there weren't any FTCs in?
Re: Something from nothing?
on
Shapes of Time
·
· Score: 2
> With every new dissertation on evolution, I find it interesting that no researcher seems to address the fundamental, underlying problem which dogs evolution:
Read a lot of dissertations on evolution, do you?
> Where does new genetic information come from?
Depends on how you define "information". If you use Shannon information, the less predictable an observation is then the more information you get from the observation. From that perspective a mudslide generates more information than a birth with a mutation does.
Evolution deniers have tried their hand at defining genomic information and making a claim that information so defined can't increase spontaneously, but those claims never stand up to the test. The closest thing to rigor was the attempt by Lee Spetner, but a close reading shows that he pulls a bait-n-switch argument when the chips are down. To all appearances, there is no law of nature that says "information can't increase".
> Why is it so important? Because they're saying we're getting something from nothing.
No, we're getting that "something" from evolution.
Scientists are very like creationists when they see something amazing and say "That could never have come about by chance!". The two camps part ways after that observation, with the creationist invoking the "goddidit" mantra and the scientist trying to figure out what actually caused the counterintuitive event. In this case, the explanation is neo-darwinian evolution. But there's nothing special about that; we don't get "something for nothing" when planets form, when hurricanes form, when snowflakes form, or when NaOH + HCl => NaCl + H2O. Science is all about understanding why those things happen rather than some other outcome.
> A system which isn't directed towards any goal teleologically goes nowhere.
Loaded semantics there. Is a mudslide "directed towards a goal"? A hurricane? A supernova?
> And if it is directed, it must have a net positive influx of information. What is that source?
Impossible to say without hearing your definition of "information", but the usual intuitive answer is "from the environment". By intuitive standards, genetic algorithms can "create information" simply by interacting with their environment (i.e., the fitness evaluation). Why can't the "biological algorithm" do the same?
The problem with your argument is that there's no empirical reason to be concerned with it. It's the armchair argument of people who want to show that evolution is wrong for reasons that have nothing to do with information.
Slashdot Mirror
> India is probably the only fully-functioning democratic country between Australia and Israel
Which way around?
> You wont need to learn your HINDI, because ALL Indian people who have a college education are fluent in British English
OK, we'll start working on our British English instead.
> when my trust Cd-n-Go software couldn't access freedb
Freedb was having lots of problems yesterday, presumably due to a Christmas-induced DDoS attack.
> Some would say only God would allow a transcendental number raise to another transcendental number, raise to an imaginary number to come out so nice as just -1!
Would that be the same God who designed the relative sizes of penises and anuses?
Your method of "reasoning" could lead to some very alarming conclusions if you applied it the same to every observation.
> "Those sirry Americans think they'll sell more products if they make them defective!"
And a generation from now we'll all be listening to Japanese music.
> 1) Apple produces a comprehensive set of UE guidelines for application developers to follow
What's the old saying about "the harder you squeeze"?
> Something so incredibly nefarious like a gross invasion of civil liberties and free speech somehow gets segued into yet another blathering rant about the RIAA/MIAA?
Probably the real motive of centralizing the net is to make it a corporate playground rather than a citizens' playground. The bit about "security" is just an excuse for doing it.
One of my first posts on Slashdot, way back when, was "I fear for the internet". It lets us step on the toes of too many vested interests. Unfettered communication between citizens hasn't been the norm in 'free' countries, and some parties in those countries are deciding that they don't like it.
> Yeah, keep postning this story...
Maybe he's submitting duplicate stories to Slashdot for his spamish revenge?
> We have been working on institutions in India and they are heavily leaning towards adopting Linux. Microsoft sales teams have been bending over backwards to prevent them from taking the plunge. The recent donations of funds as well as the offer to share the code all amount to last ditch attempts to keep Linux out.
Yes, The Register had an intersting take on billg's recent coin-scattering trip to India.
> The big problem with GA though, IIRC, is that the resulting solution is often incomprehensible to a human.
Same with perl, but people use it anyway.
> Tina Turner is SO 80's!
Not quite, but she must be getting close.
> No way. Road Warrior is the best of the three. MM3 was full of cliched post-apocalyptic child civilizations straight from some bad b-movie.
MM:BtTD is the apotheosis of the B Movie.
I love it; it's one of my five favorite movies of all time.
> Now, I know you think it sucked, but I bet you went and saw Episode 2 as well, eh?
Nope. TPM was bad. I don't have the least interest in seeing e2.
> The content of this book is spot on. Just as important is the layout of the book and Sobell has nailed the best way to present this sort of text. It is quick and easy to find exactly what you need.
FWIW, Sobel has a new book A Practical Guide to Red Hat Linux 8 in press, supposedly out this month. Shame about the distro-/version- specific presentation, but the TOC makes it look like it will be a good book for anyone who matches the relevant distro/version.
> I'm not sure you'd get a lot of improvement from the technique though - bots would tend to play well with bots this way, unless you somehow managed to get a fitness function that reflected playing with humans. You could also invite a lot of human beta testers and evolve them based on survival rates.
You could bootstrap the system by having it play itself. That doesn't always result in the desired "arms race" of continual improvements, but people are looking at ways of inducing arms races.
If you could ever get it up to some basic level of performance (say, about what the typical game AI does now), you could enter a phase of distributed learning where thousands of people who had bought the game downloaded the latest AI, played it a time or two, and uploaded the results. That way no individual would have to play through a lot of games with a stupid AI as it explored options to improve its behavior against humans. Presumably if you downloaded the latest AI every week you would get a more competitive opponent every time.
> Some algorithm that noticed patterns in actual gameplay would probably induce cooler behaviour - something in the lines of
Yeah, those are hard problems for AI if you try to generalize the behavior. I wish more AI researchers were working on this kind of stuff instead of the toy problems that so many of them are still on. Several AI researchers have suggested games as the best research domain for acheiving "human level" AI.
> Beowulf cluster? Reinforcement learning? Evolutionary algorithm? What are you, the buzzword king?
No, just mentioning some things I've been reading up on and experimenting with that I think may eventually offer a solution.
> First of all, how would this "solution" answer your problem with modpacks?
Read my post.
> Naturally, the AI has the shortest time frame in the software engineering, but there is no reason it should remain stagnent across the future patches.
Another problem is that lots of games are just engines that support an 'official' dataset plus whatever modpacks the players care to come up with, but even the cheatAI that ships with the game won't work worth a damn on the modpacks.
I hope in the future machine learning methods can help with both of these problems. I.e., a couple of months before release when the code is fairly stable and the graphics are in production, turn on the old Beowulf cluster and let reinforcement learning or an evolutionary algorithm train a good AI for the game. As for modpacks, the vendors could support something like sourceforge, where gamers could upload their modpacks and have the Beowulf cluster automagically re-tune the AI to work right with them.
And of course, the machine learning could continue in the background for as long as people were interested in the game, allowing them to download "new improved" AIs every few months.
> Please pass this information on to as many people as possible. Let's make a child's dream come true before its too late.
Be sure to mention how sad you were when you heard that he was dying from inrectocephalosis.
> I've always thought that there are two very distinct skill involved in writing. The first is storytelling, the ability to weave a yarn that is enthralling, touching, satisfying, etc. The second is skill with the language, the ability to create a rich imaginary world, enticing to all the senses, with only the written word.
> There are some writers who clearly excel at both. The first that comes to mind is Pat Conroy.
Jack Vance, Jack Vance!
Sorry, you pulled my string.
> Indeed, Spielberg did a fantastic job with Jurassic Park.
Feh. Superb dinosaurs, tolerable plot, third-rate actors, execrable script. Never have so few characters spoken so many bad lines in so few movies (the sequel was actually worse!). I've heard better dialog in porno flicks.
> This legislation has been abused like a village bicycle!
Heh. My village had a girl nicknamed 'bicycle', for reasons that shouldn't need explaining.
OMFG! Patent Office, here I come!> I have a friend at SETI, and he sent me the code for the best signal.
> As others have pointed out, we could pick up something that existed a few score or a few hundred years ago, and that would certainly be interesting.
Especially if they left any money on deposit.
>
I'm curious as to what sort of photograph would show that fairy tale creatures don't exist. Maybe a picture of the whole universe, that you could go through and show that there weren't any FTCs in?
> With every new dissertation on evolution, I find it interesting that no researcher seems to address the fundamental, underlying problem which dogs evolution:
Read a lot of dissertations on evolution, do you?
> Where does new genetic information come from?
Depends on how you define "information". If you use Shannon information, the less predictable an observation is then the more information you get from the observation. From that perspective a mudslide generates more information than a birth with a mutation does.
Evolution deniers have tried their hand at defining genomic information and making a claim that information so defined can't increase spontaneously, but those claims never stand up to the test. The closest thing to rigor was the attempt by Lee Spetner, but a close reading shows that he pulls a bait-n-switch argument when the chips are down. To all appearances, there is no law of nature that says "information can't increase".
> Why is it so important? Because they're saying we're getting something from nothing.
No, we're getting that "something" from evolution.
Scientists are very like creationists when they see something amazing and say "That could never have come about by chance!". The two camps part ways after that observation, with the creationist invoking the "goddidit" mantra and the scientist trying to figure out what actually caused the counterintuitive event. In this case, the explanation is neo-darwinian evolution. But there's nothing special about that; we don't get "something for nothing" when planets form, when hurricanes form, when snowflakes form, or when NaOH + HCl => NaCl + H2O. Science is all about understanding why those things happen rather than some other outcome.
> A system which isn't directed towards any goal teleologically goes nowhere.
Loaded semantics there. Is a mudslide "directed towards a goal"? A hurricane? A supernova?
> And if it is directed, it must have a net positive influx of information. What is that source?
Impossible to say without hearing your definition of "information", but the usual intuitive answer is "from the environment". By intuitive standards, genetic algorithms can "create information" simply by interacting with their environment (i.e., the fitness evaluation). Why can't the "biological algorithm" do the same?
The problem with your argument is that there's no empirical reason to be concerned with it. It's the armchair argument of people who want to show that evolution is wrong for reasons that have nothing to do with information.