Note: I'm an atheist. But here's a bit of philosophy for you.
Science requires faith: particularly, faith in the scientific method, which is founded on a completely unjustifiable belief in inductive reasoning - that is, "if one thing causes another thing a lot, it'll probably cause that same thing in the future too". Inductive reasoning is why we require experiments to be repeatable. Once they're repeatable, then we can assume they apply to the real world.
But why do you believe in induction? It can't be derived from deductive logic (that's why it's not called deduction). Your reason for believing in induction is probably that it's always worked before. But then you're using induction to justify induction!
Really, the reason to believe in induction (and therefore the scientific method) is because it creates results that are useful to the progress of society, not because it is more justifiable.
A rational person, generally, is someone who chooses that induction is a basis for believing in things.
These are the same people who sold MIT their database of "legal" streamable music for the LAMP system. When the RIAA came down on MIT, MIT pointed to Loudeye and said "They said it was legal", and Loudeye essentially said "Uh, did we? Sorry, we meant, uh, legal to not do anything with."
Unless they back up the legality of their product, the music database that Loudeye sells is no different from Kazaa.
Ah yes, that page. "Google doesn't put my NameBase results on top! So Google is evil! Wah!"
There's some point where you have to realize that people aren't going to your site because it sucks.
And when it comes to blogs, it seems that now if you even mention Google on LiveJournal you get some random person you don't know evangelizing their favorite Google alternative, claiming that Google gave them pop-up ads, banners, herpes, or whatever.
Congratulations, you have rediscovered Markov chains.
As for related programs, what you describe has been done already by the "LiveJournal Poetry Generator", JWZ's "dadadodo", the fake AIs "MegaHAL" and "gNiall" and many many more (some of which unfortunately claim to be real AI), emacs' "dissociated-press", and so on.
So you believe English would have been perfectly phonetic without "grammar nazis" deciding that, say, the word pronounced "koff" should be spelled "cough"?
The irregularities in spelling come from ordinary people, and those same people would resist any attempt to make English phonetic.
Most people cannot make a consistent mapping from sounds to letters on their own. For example, you probably won't find anyone who realizes that the "ch" sound is pronounced as "t" followed by "sh" without being told so first. So the spelling of English mostly came from people writing down whatever sequence of letters reminded them of a word, until Webster came along and sorted things out, by listing the way each word was spelled by most people.
If wee speld wurdz laik wee prunounst dhem and roat sentinsiz laik wee spoak dhem, wee wudnt bee yoozing Inglish ennymoar.
Your ballot can be checked to ensure that it is a valid vote. The pixelating XOR stuff he did is to ensure that, while your vote can be checked for validity, it cannot be checked to see who you voted for, except by the board of trustees, who have the other half of the vote and have no information about who you are.
People who know CSS well should be people who know good design, right? But it seems that they all like to show off their CSS skillz by putting dotted lines everywhere on the page. Like here, in their "second skin" example.
What is the obsession with dotted lines? They're ugly and distracting. It's nothing but a way to scream to the world "Hey look, I know CSS and I'm using it to make my page look worse! Go me!"
Yeah, but that's okay. Given Boies' track record (losing the antitrust suit against Microsoft, losing the election for Al Gore) we don't even have to worry about him winning once.
That could make a good job title. "David Boies, Attorney-At-Law, Professional Loser"
Topcoder runs tournaments (mostly to attract good coders in the first place) and then there's a separate page for component development. If you do the component development, you get paid. That's the stuff they sell to companies. The "component competition" you linked to is where they're throwing an extra bonus on top of the pay you get for a component, to encourage more people to do it.
But most people do the tournaments, because they're much more fun. There was a time when TopCoder was only tournaments; basically, they gave out lots of money and hardly made any. That, of course, had to stop, so now they have the components section too.
It should be clear why they have to have extra incentives like the component competition: writing useful, non-specialized code that companies would want to buy is boring as all hell. So most TopCoders don't participate in that part of the site, even though the pay is good.
The problems that are asked for the coding tournaments, like Google's, have all been solved before (that's how they have a reference solution to compare your outputs to!) That's not the code they sell. These problems are purely for fun. Look at the medium-level problem from the championship: given a polynomial, find the largest root. This is not cutting edge code that a company will pay for. Your TI calculator can do that. However, writing the code to do it, from scratch, in less than an hour, is quite an interesting challenge.
And if you consider yourself a geek, but can't fathom the idea of people writing code for fun... be very embarrassed.
You boot from a Knoppix CD, and all you have to do is install a base system and apt to your hard disk, and you've got a Debian system that's already configured.
They should acknowledge this fact and officially support Knoppix as an install method for desktop users. Then they can still focus their installer on people who want to install Debian on an Alpha over their serial line.
I dislike IRV as much as you do, but at the moment you can't blame Australia - no country is doing any better than IRV for their national elections.
Good methods (like Condorcet) should start in small organizations and work their way up, so that people are already familiar with how good preferential voting works.
Debian, for example, has already worked out lots of kinks and unfairness in their voting system by switching to Condorcet. Some "rules of order" books now advocate using Condorcet when possible. Encouraging this is what will get good preferential voting accepted, not pointing an angry finger at the government that's using a slightly better method than everyone else but still isn't good enough.
The best part of the site is "Spontaneous knowledge" - when it gains just enough information to cross some threshold and draw its own conclusions about an object (which is usually not the object you had as your answer). At the end of the game, it tells you all of these conclusions.
For example, I played it with the answer "a search engine". Here's what it concluded: The mandelbrot set is probably not made of plastic The mandelbrot set probably doesn't live in the desert You might not wear the mandelbrot set on your finger The mandelbrot set probably doesn't wear a mask The mandelbrot set probably doesn't have a trunk The mandelbrot set probably doesn't crumble The mandelbrot set probably doesn't nest on top of chimneys The mandelbrot set probably doesn't burrow The mandelbrot set is probably not usually made from bricks The mandelbrot set is probably not made out of wood The mandelbrot set is probably not a pizza topping The mandelbrot set probably doesn't live in tundra The mandelbrot set probably doesn't have gears The mandelbrot set probably doesn't have flippers The mandelbrot set probably doesn't have a spring The mandelbrot set might not sting The mandelbrot set probably doesn't have tassels The mandelbrot set probably doesn't wear black The mandelbrot set is probably not radioactive The mandelbrot set is probably not sometimes kept as a house pet The mandelbrot set probably doesn't vibrate at a predetermined rate The mandelbrot set is probably not a type of grain The mandelbrot set is probably not a cartoon character The mandelbrot set probably doesn't help you clean your teeth The mandelbrot set is probably not an endangered species Something probably doesn't live in the mandelbrot set The mandelbrot set probably doesn't grow in a bunch The mandelbrot set is probably not a lazy animal The mandelbrot set probably doesn't hop The mandelbrot set is probably not long and thin The mandelbrot set is probably not a type of grass
and not only that, the "learning" isn't anything more adding a discriminator a search tree.
Have you used 20Q.net? The thing is certainly not the binary-tree-in-BASIC you're thinking of. It gets the right answer before you think it could possibly know. They've tuned this algorithm over several years, and it's smart.
How is the halting problem brute force? It's proven using a very short, elegant proof.
How is the Church-Turing thesis brute force? It's nothing but an assumption that allows us to make statements in computer science. Church didn't go around to a bunch of computers saying "Yep, this one only computes things that are Turing-computable... so does this one... so does this one..."
And of course, if you use brute force on your girlfriend, then you really suck.
Your post contains a sequence of more than one question mark; therefore, you are in violation of the Sounding Like A Fucking Dolt Act. Federal agents have already traced the source of the offending punctuation, and should be arriving shortly at your door.
It's my impression that Maxwell's equations can account for the existence of magnetic monopoles; in fact, there are some interesting physics problems you can do (using Maxwell's equations) if you assume the existence of magnetic monopoles.
The equation can be stated "the divergence of the magnetic field equals the density of magnetic monopoles", but of course, in most situations, it's easier to say "zero" than "the density of magnetic monopoles", since there aren't any magnetic monopoles around.
LyX defeats the point. It has menus. It has toolbars and buttons all over the place. It has far too many choices, which is what would cause a writer to play with them (in order to procrastinate) instead of writing.
Slashdot Mirror
Note: I'm an atheist. But here's a bit of philosophy for you.
Science requires faith: particularly, faith in the scientific method, which is founded on a completely unjustifiable belief in inductive reasoning - that is, "if one thing causes another thing a lot, it'll probably cause that same thing in the future too". Inductive reasoning is why we require experiments to be repeatable. Once they're repeatable, then we can assume they apply to the real world.
But why do you believe in induction? It can't be derived from deductive logic (that's why it's not called deduction). Your reason for believing in induction is probably that it's always worked before. But then you're using induction to justify induction!
Really, the reason to believe in induction (and therefore the scientific method) is because it creates results that are useful to the progress of society, not because it is more justifiable.
A rational person, generally, is someone who chooses that induction is a basis for believing in things.
These are the same people who sold MIT their database of "legal" streamable music for the LAMP system. When the RIAA came down on MIT, MIT pointed to Loudeye and said "They said it was legal", and Loudeye essentially said "Uh, did we? Sorry, we meant, uh, legal to not do anything with."
Unless they back up the legality of their product, the music database that Loudeye sells is no different from Kazaa.
Wow. That's the first slashborging ("All Slashdotters should have the same opinions! Be consistent, dammit!") post I've seen in a long time.
Even though they're stupid as hell, I was beginning to miss them.
Ah yes, that page. "Google doesn't put my NameBase results on top! So Google is evil! Wah!"
There's some point where you have to realize that people aren't going to your site because it sucks.
And when it comes to blogs, it seems that now if you even mention Google on LiveJournal you get some random person you don't know evangelizing their favorite Google alternative, claiming that Google gave them pop-up ads, banners, herpes, or whatever.
Congratulations, you have rediscovered Markov chains.
As for related programs, what you describe has been done already by the "LiveJournal Poetry Generator", JWZ's "dadadodo", the fake AIs "MegaHAL" and "gNiall" and many many more (some of which unfortunately claim to be real AI), emacs' "dissociated-press", and so on.
So you believe English would have been perfectly phonetic without "grammar nazis" deciding that, say, the word pronounced "koff" should be spelled "cough"?
The irregularities in spelling come from ordinary people, and those same people would resist any attempt to make English phonetic.
Most people cannot make a consistent mapping from sounds to letters on their own. For example, you probably won't find anyone who realizes that the "ch" sound is pronounced as "t" followed by "sh" without being told so first. So the spelling of English mostly came from people writing down whatever sequence of letters reminded them of a word, until Webster came along and sorted things out, by listing the way each word was spelled by most people.
If wee speld wurdz laik wee prunounst dhem and roat sentinsiz laik wee spoak dhem, wee wudnt bee yoozing Inglish ennymoar.
You misunderstand what he meant by "checking".
Your ballot can be checked to ensure that it is a valid vote. The pixelating XOR stuff he did is to ensure that, while your vote can be checked for validity, it cannot be checked to see who you voted for, except by the board of trustees, who have the other half of the vote and have no information about who you are.
People who know CSS well should be people who know good design, right? But it seems that they all like to show off their CSS skillz by putting dotted lines everywhere on the page. Like here, in their "second skin" example.
What is the obsession with dotted lines? They're ugly and distracting. It's nothing but a way to scream to the world "Hey look, I know CSS and I'm using it to make my page look worse! Go me!"
If the card cannot be read, it is destroyed, and the machine shuts down until someone can service it.
There's a good way to bring an election to a screeching halt.
Just stick unreadable cards into voting machines all over a precinct containing lots of voters whom you disagree with.
Yeah, but that's okay. Given Boies' track record (losing the antitrust suit against Microsoft, losing the election for Al Gore) we don't even have to worry about him winning once.
That could make a good job title. "David Boies, Attorney-At-Law, Professional Loser"
Holy crap are you confused.
Topcoder runs tournaments (mostly to attract good coders in the first place) and then there's a separate page for component development. If you do the component development, you get paid. That's the stuff they sell to companies. The "component competition" you linked to is where they're throwing an extra bonus on top of the pay you get for a component, to encourage more people to do it.
But most people do the tournaments, because they're much more fun. There was a time when TopCoder was only tournaments; basically, they gave out lots of money and hardly made any. That, of course, had to stop, so now they have the components section too.
It should be clear why they have to have extra incentives like the component competition: writing useful, non-specialized code that companies would want to buy is boring as all hell. So most TopCoders don't participate in that part of the site, even though the pay is good.
The problems that are asked for the coding tournaments, like Google's, have all been solved before (that's how they have a reference solution to compare your outputs to!) That's not the code they sell. These problems are purely for fun. Look at the medium-level problem from the championship: given a polynomial, find the largest root. This is not cutting edge code that a company will pay for. Your TI calculator can do that. However, writing the code to do it, from scratch, in less than an hour, is quite an interesting challenge.
And if you consider yourself a geek, but can't fathom the idea of people writing code for fun... be very embarrassed.
The easiest Debian installer is Knoppix.
You boot from a Knoppix CD, and all you have to do is install a base system and apt to your hard disk, and you've got a Debian system that's already configured.
They should acknowledge this fact and officially support Knoppix as an install method for desktop users. Then they can still focus their installer on people who want to install Debian on an Alpha over their serial line.
But you only have two of the important ones.
I dislike IRV as much as you do, but at the moment you can't blame Australia - no country is doing any better than IRV for their national elections.
Good methods (like Condorcet) should start in small organizations and work their way up, so that people are already familiar with how good preferential voting works.
Debian, for example, has already worked out lots of kinks and unfairness in their voting system by switching to Condorcet. Some "rules of order" books now advocate using Condorcet when possible. Encouraging this is what will get good preferential voting accepted, not pointing an angry finger at the government that's using a slightly better method than everyone else but still isn't good enough.
The best part of the site is "Spontaneous knowledge" - when it gains just enough information to cross some threshold and draw its own conclusions about an object (which is usually not the object you had as your answer). At the end of the game, it tells you all of these conclusions.
For example, I played it with the answer "a search engine". Here's what it concluded:
The mandelbrot set is probably not made of plastic
The mandelbrot set probably doesn't live in the desert
You might not wear the mandelbrot set on your finger
The mandelbrot set probably doesn't wear a mask
The mandelbrot set probably doesn't have a trunk
The mandelbrot set probably doesn't crumble
The mandelbrot set probably doesn't nest on top of chimneys
The mandelbrot set probably doesn't burrow
The mandelbrot set is probably not usually made from bricks
The mandelbrot set is probably not made out of wood
The mandelbrot set is probably not a pizza topping
The mandelbrot set probably doesn't live in tundra
The mandelbrot set probably doesn't have gears
The mandelbrot set probably doesn't have flippers
The mandelbrot set probably doesn't have a spring
The mandelbrot set might not sting
The mandelbrot set probably doesn't have tassels
The mandelbrot set probably doesn't wear black
The mandelbrot set is probably not radioactive
The mandelbrot set is probably not sometimes kept as a house pet
The mandelbrot set probably doesn't vibrate at a predetermined rate
The mandelbrot set is probably not a type of grain
The mandelbrot set is probably not a cartoon character
The mandelbrot set probably doesn't help you clean your teeth
The mandelbrot set is probably not an endangered species
Something probably doesn't live in the mandelbrot set
The mandelbrot set probably doesn't grow in a bunch
The mandelbrot set is probably not a lazy animal
The mandelbrot set probably doesn't hop
The mandelbrot set is probably not long and thin
The mandelbrot set is probably not a type of grass
and not only that, the "learning" isn't anything more adding a discriminator a search tree.
Have you used 20Q.net? The thing is certainly not the binary-tree-in-BASIC you're thinking of. It gets the right answer before you think it could possibly know. They've tuned this algorithm over several years, and it's smart.
You don't get it. It's a part of the game, and the only N-way question they ask is that one.
Plus, 20Q is tuned to answer yes-no questions, and it doesn't get much information from the first one.
I forget who it is who said this:
What in the world are you talking about?
How is the halting problem brute force? It's proven using a very short, elegant proof.
How is the Church-Turing thesis brute force? It's nothing but an assumption that allows us to make statements in computer science. Church didn't go around to a bunch of computers saying "Yep, this one only computes things that are Turing-computable... so does this one... so does this one..."
And of course, if you use brute force on your girlfriend, then you really suck.
Proportional representation includes more than party list voting. Learn about the Single Transferrable Vote system.
Your post contains a sequence of more than one question mark; therefore, you are in violation of the Sounding Like A Fucking Dolt Act. Federal agents have already traced the source of the offending punctuation, and should be arriving shortly at your door.
They could still put your domain on a message going through a spoofable system, right?
I believe this only prevents forging someone else's domain on your system.
It's my impression that Maxwell's equations can account for the existence of magnetic monopoles; in fact, there are some interesting physics problems you can do (using Maxwell's equations) if you assume the existence of magnetic monopoles.
The equation can be stated "the divergence of the magnetic field equals the density of magnetic monopoles", but of course, in most situations, it's easier to say "zero" than "the density of magnetic monopoles", since there aren't any magnetic monopoles around.
You mean the print dialog is missing the "Print to file" option?
LyX defeats the point. It has menus. It has toolbars and buttons all over the place. It has far too many choices, which is what would cause a writer to play with them (in order to procrastinate) instead of writing.