A few years back Donald Knuth wrote a very enlightening open letter to the editorial board of his Journal of Algorithms. It gives a fascinating insider view of the process of journal publishing. My interpretation of Knuth's arguments is that traditional journals are now next to useless. Scientists already do all the typesetting and peer-reviewing of articles, and the internet means anybody can get together and start a free online journal with a very modest investment.
http://zesty.ca/voting/sim/ illustrates clearly some of the really wacky things that can happen with IRV. For example, in certain situations, a shift in the public opinion towards a certain candidate can actually cause that candidate to lose.
When it came to the area with the close-the-toilet-door scene, it was ridiculous. The trick for me was realizing that the timing of the nunchuck shake is important and needs to occur just after you point at the door; if you are just constantly shaking the nunchuck while pointing at doors, that doesn't work.
What if people chose to help each other, including themselves? I agree that such a society would be heavenly, but in order to create it, we would have to force people to act accordingly, thus depriving them of their free will, thus destroying them, etc.
What constitutes force which deprives free will? For example, the need for oxygen forces people to breathe, depriving them of the free will to choose not to breathe. So free will in this sense is impossible. But free will could mean free from imposition by another will. In other words, if the "force" is applied impartially to everybody, i.e. by nature, then this could solve your dilemma.
Everything he says applies to server>client. Producers>consumers.
On the contrary, it sounds to me like he's describing an egalitarian network where anybody who is connected to the internet can inject data into it with very little hosting overhead (because the data will be cached inside the network).
What about bittorrent though?
The first question at the end of the presentation was, what does he think of Bittorrent? And his problems with it are:
Only people actively downloading or seeding content are available to redistribute it, so there's a built-in time-dependence which makes in unsuitable for small pieces of data.
It has no knowledge of topology, so it can't take advantage of topology to cache data closer to endpoints.
I wish someone had asked about freenet, since that seems much closer to what he describes.
Here is the technique I used to get a Wii with a minimum of effort:
Once every few days, check a site like Fat Wallet for rumors of Wiis in a weekly ad at a major retailer.
You may have to wait a couple weeks, but probably not more than that.
Call the stores in your area the day before the ad comes out and see which is getting the most Wiis. You might also ask if they will hand out vouchers early.
Showing up an hour before vouchers are handed out, or 2-3 hours before the store opens, will pretty much guarantee you a Wii. You can probably show up later. I got mine at Toys R Us 2 weeks ago and there were still 20 systems unclaimed when they started handing out vouchers.
Re:Rails is Doomed
on
Rails Cookbook
·
· Score: 2, Insightful
That's a highly optimized algorithm. If you think that looks bad, check out the equivalent C code.
typedef struct {
int first;
int second; } intpair;
unsigned int sqr(unsigned int x) { return x*x; } intpair fibonacci(unsigned int nn) {
intpair out;
switch (nn) {
case 0: out.first = 0; out.second = 0; return out;
case 1: out.first = 1; out.second = 0; return out;
case 2: out.first = 1; out.second = 1; return out;
case 3: out.first = 2; out.second = 1; return out;
}
unsigned int mm = nn >> 1;
unsigned short int rr = nn & 1;
intpair f = fibonacci(mm);
unsigned int f0 = f.first;
unsigned int f1 = f.second;
if (rr == 0) {
out.first = f0*(f1*2 + f0);
out.second = f0*f0 + f1*f1;
} else {
out.first = f0*f0 + sqr(f0 + f1);
out.second = f0*(f1*2 + f0);
return out;
} }
Obviously if you already know C you'll find the C version easier to understand, but if you don't I think it's hard to argue that this is any more obvious than the Lisp version, and it's a lot longer. Oh yeah, and it doesn't even get the right answers:
A lot of people are saying that the sexes are inherently different, and we shouldn't try artificially to force them to become the same. If we were talking about nursemaids or prison guards, that might be a valid point. It's conceivable to imagine that, despite the fact that humans come with immensely flexible brains, there is just enough evolutionarily-inbuilt predisposition to, say, nurturing in one case, or violence in another case, that we should expect these careers to be gender-inbalanced. Of course, there is no scientific evidence to support this, but it's at least plausible.
Engineering and computer science, on the other hand, are just white collar jobs like any other. If you look at people who are at the top of any such field, they're all pretty much the same: they are smart, they are good at solving problems, and they enjoy what they do. So maybe some people enjoy making money, and some enjoy making programs, and some enjoy making businesses, but there's no reason to imagine that evolution has predisposed certain genders to care more about certain fields. That's like saying girls "naturally" prefer pink over blue; the only thing that could possibly influence that choice is culture.
What may be true is that women prefer jobs that involve more job security, more social aspects, and more respect from their coworkers. But surely nobody thinks that being antisocial is a requirement for being a good programmer? The culture of computer science is a result of the field being dominated by men, and is in no way related to the actual practice of computer science.
how difficult it was for them to understand that, when their Word document disappeared from the screen when they minimized the window, it did not also disappear from "the computer" I'm skeptical. Did these people also think that all the little people in their TV die when they turn it off?
You cannot package javascript up in a module which ensures that there will be no namespace collisions Sure you can, if you are willing to accept the risk of collisions in module names (after all, what happens if two Java packages use the same name)?
// define a new module org_slashdot_moduleX = new function() { var module_private_variable =... function private_function() { ... } this.public_function = new function() { ... } }// end module org_slashdot_moduleX
// import the module as "mX": var mX = new org_slashdot_moduleX(); mx.public_function();
You can even parameterize modules to create ML-style functors.
Your scheme doesn't work because triangles are not atomic; they are made out of lines, which are described in terms of endpoints, which have a finite resolution. For example, an inkjet printer would make a triangle by printing dots at some fixed resolution. A drafting machine or laser printer might be able to draw the triangle without resorting to dots, but the elements of the triangle can still only be positioned with finite accuracy.
Let's say that we're drawing very tiny triangles as close to our resolution limit as possible (which we must do if we want to fit a lot of them). Such a triangle might be, say, 3 x 3 resolution units, so a hollow, up-triangle might look like this:
010 101 111
But look: there are 2^9 (or 512) possible shapes that can be made in this grid -- so by using only 64 different triangles, we are actually underutilising our medium. It doesn't matter what technology you use, any shape other than a "dot" is itself made out of smaller units like "dots", so restricting our vocabulary to certain shapes (rather than arbitrary sequences of dots) will waste space.
For graphics purposes, just think of a quaternion as a "rotation": the real component is the angle, and the imaginary components form a 3D vector which is the axis of rotation. Why this works and is indeed a quaternion is complicated, but the application of it isn't.
"morse code [...] uses huffman-like compression for english only"
This is called "making it up as you go along"
Are you saying this because you don't know what Huffman coding is? It's indisputable that more frequent letters in english get shorter codes, just like Huffman coding. So using morse code for, say, Czech, would probably do a lot worse.
Your wishes are relics of an MSDOS era when only one application could run at one time, so your data was inextricably tied to the application you used it in. Here's a brilliant idea: store your calendar, emails, etc. in FILES, and provide a uniform API to search, retrieve, and update this information. Then your email program can add stuff to your calendar and visa versa without them having to share a window on your desktop. Why, it's almost like a DATABASE.
I think you've misspelt the title of the book. Surely you meant: The Honor's Clas's: Hilbert's Problem's and Their Solver's, a classic in the field of egregiou's misu'se of apostrophe's.
Literate programming (meaning stuff like WEB) is a disgusting hack to make up for the inefficiencies of a few poorly-designed languages. Things like costly subroutines, no higher-order functions, no continuations, no local variables, no declarations mixed in to function bodies. But really, it's precisely equivalent to sprinkling your code liberally with gotos, and hopefully we know why that's bad. There were more structured ways to make the literal flow of the program match the logical 20 years ago, but Knuth was too much of a NIH masochist to realize it.
If you need bracketing code, put it in a higher-order function and call that with a thunk. Or better yet, use a macro. (You are using a language with real macros and not that non-referentially-transparent cpp-style crap, right?)
"You don't own your land, you rent it from the government." Do you believe that you are somehow more entitled to its exclusive use than any other of your 6+ billion fellow humans? A capitalist like yourself should see the virtue of a system where you earn what you get -- you keep the land by virtue of adding value to it beyond what it's worth to the rest of humanity by itself (the rent).
So you don't believe in property taxes, or what? The police are *still* protecting the money, every year, so it makes sense that they should get paid every year to do so.
"If you don't spend the money, you see no benefit from having it." -- That's ridiculous. The benefit in having money isn't spending it, it's making more money so you can spend that instead, and basically live off the interest. I don't know how much it's worth to never have to work the rest of your life, but it's certainly worth more than nothing.
FYI: the higher level a language is, the *more* possible it is to optimize in general (though the optimizations may be difficult). It is languages like C that are "impossible to optimize", since they way overspecify, for example, the order-of-execution, and force optimizers to bend over backwards preserving semantics that the programmer doesn't even care about.
The word you're defining is either "lie" or "err", depending on if the person knew they were wrong. Irony is not just stating a falsehood. Though perhaps it was ironic that you jumped all over this guy for his misuse of the word.
Slashdot Mirror
A few years back Donald Knuth wrote a very enlightening open letter to the editorial board of his Journal of Algorithms. It gives a fascinating insider view of the process of journal publishing. My interpretation of Knuth's arguments is that traditional journals are now next to useless. Scientists already do all the typesetting and peer-reviewing of articles, and the internet means anybody can get together and start a free online journal with a very modest investment.
Did you mean "closures" and objects? Because I've never heard anybody compare continuations to objects (in the object-oriented sense).
http://zesty.ca/voting/sim/ illustrates clearly some of the really wacky things that can happen with IRV. For example, in certain situations, a shift in the public opinion towards a certain candidate can actually cause that candidate to lose.
- Only people actively downloading or seeding content are available to redistribute it, so there's a built-in time-dependence which makes in unsuitable for small pieces of data.
- It has no knowledge of topology, so it can't take advantage of topology to cache data closer to endpoints.
I wish someone had asked about freenet, since that seems much closer to what he describes.Here is the technique I used to get a Wii with a minimum of effort:
A lot of people are saying that the sexes are inherently different, and we shouldn't try artificially to force them to become the same. If we were talking about nursemaids or prison guards, that might be a valid point. It's conceivable to imagine that, despite the fact that humans come with immensely flexible brains, there is just enough evolutionarily-inbuilt predisposition to, say, nurturing in one case, or violence in another case, that we should expect these careers to be gender-inbalanced. Of course, there is no scientific evidence to support this, but it's at least plausible.
Engineering and computer science, on the other hand, are just white collar jobs like any other. If you look at people who are at the top of any such field, they're all pretty much the same: they are smart, they are good at solving problems, and they enjoy what they do. So maybe some people enjoy making money, and some enjoy making programs, and some enjoy making businesses, but there's no reason to imagine that evolution has predisposed certain genders to care more about certain fields. That's like saying girls "naturally" prefer pink over blue; the only thing that could possibly influence that choice is culture.
What may be true is that women prefer jobs that involve more job security, more social aspects, and more respect from their coworkers. But surely nobody thinks that being antisocial is a requirement for being a good programmer? The culture of computer science is a result of the field being dominated by men, and is in no way related to the actual practice of computer science.
Let's say that we're drawing very tiny triangles as close to our resolution limit as possible (which we must do if we want to fit a lot of them). Such a triangle might be, say, 3 x 3 resolution units, so a hollow, up-triangle might look like this:But look: there are 2^9 (or 512) possible shapes that can be made in this grid -- so by using only 64 different triangles, we are actually underutilising our medium. It doesn't matter what technology you use, any shape other than a "dot" is itself made out of smaller units like "dots", so restricting our vocabulary to certain shapes (rather than arbitrary sequences of dots) will waste space.
For graphics purposes, just think of a quaternion as a "rotation": the real component is the angle, and the imaginary components form a 3D vector which is the axis of rotation. Why this works and is indeed a quaternion is complicated, but the application of it isn't.
This is called "making it up as you go along"
Are you saying this because you don't know what Huffman coding is? It's indisputable that more frequent letters in english get shorter codes, just like Huffman coding. So using morse code for, say, Czech, would probably do a lot worse.
I'll let you know when Apple's finished with it. ;)
Your wishes are relics of an MSDOS era when only one application could run at one time, so your data was inextricably tied to the application you used it in. Here's a brilliant idea: store your calendar, emails, etc. in FILES, and provide a uniform API to search, retrieve, and update this information. Then your email program can add stuff to your calendar and visa versa without them having to share a window on your desktop. Why, it's almost like a DATABASE.
I think you've misspelt the title of the book. Surely you meant: The Honor's Clas's: Hilbert's Problem's and Their Solver's, a classic in the field of egregiou's misu'se of apostrophe's.
Literate programming (meaning stuff like WEB) is a disgusting hack to make up for the inefficiencies of a few poorly-designed languages. Things like costly subroutines, no higher-order functions, no continuations, no local variables, no declarations mixed in to function bodies. But really, it's precisely equivalent to sprinkling your code liberally with gotos, and hopefully we know why that's bad. There were more structured ways to make the literal flow of the program match the logical 20 years ago, but Knuth was too much of a NIH masochist to realize it.
If you need bracketing code, put it in a higher-order function and call that with a thunk. Or better yet, use a macro. (You are using a language with real macros and not that non-referentially-transparent cpp-style crap, right?)
"You don't own your land, you rent it from the government." Do you believe that you are somehow more entitled to its exclusive use than any other of your 6+ billion fellow humans? A capitalist like yourself should see the virtue of a system where you earn what you get -- you keep the land by virtue of adding value to it beyond what it's worth to the rest of humanity by itself (the rent).
Maybe you were imagining something like this: TouchStream keyboard
So you don't believe in property taxes, or what? The police are *still* protecting the money, every year, so it makes sense that they should get paid every year to do so.
"If you don't spend the money, you see no benefit from having it." -- That's ridiculous. The benefit in having money isn't spending it, it's making more money so you can spend that instead, and basically live off the interest. I don't know how much it's worth to never have to work the rest of your life, but it's certainly worth more than nothing.
FYI: the higher level a language is, the *more* possible it is to optimize in general (though the optimizations may be difficult). It is languages like C that are "impossible to optimize", since they way overspecify, for example, the order-of-execution, and force optimizers to bend over backwards preserving semantics that the programmer doesn't even care about.
Static typing doesn't require declarations, it just requires type inference.
The word you're defining is either "lie" or "err", depending on if the person knew they were wrong. Irony is not just stating a falsehood. Though perhaps it was ironic that you jumped all over this guy for his misuse of the word.