"You could just as well call Perl a clean simple well designed language with an easy to learn syntax, if you're going to distort the facts so grossly. "
As a Perl hacker, I resent this. Perl is nothing if not easy to learn. You can code Perl that looks like C, if that's what you're used to (I was). You can code Perl that looks like Sed, or BASIC, or shell scripting - If you know one of these, Perl just requires adjusting a few switches in your head.
Effective Perl programs can be written only using a small subset of the language. Don't like (or haven't learned) its OO? Don't use it.
Perl is also simple. Instead of forcing you to tell it exactly what to do, it has sensible defaults, like $_. It does what you expect - and if you prefer something else, you can make it do something else.
WRT your article, it was dead on ten years ago. Now, X has a whole different set of problems:).
To plug NeWS as a solution is ridiculous. NeWS is dead. It could only come back if Sun freed the code, and if dedicated hackers ported it to Linux and made an X compatibility layer. And they won't do that, because they don't understand Freedom. We can't use a windowing system that's not Free (as in speech), because we can't improve it, or fix its bugs, or port it to our processor/OS/video card.
But if a kid can read a little, the net can help them learn more. Not through "remote education", or whatever, but because people want to read what's out there. I know there are plenty of people who learned basic english in school, and learned more by hanging out on the net.
"If a kid can't do better than rudimentary math, she can't program."
Not really... I learned sines and cosine through programming an Asteroids game in QBASIC. I've also learned about Voronoi diagrams and other stuff through programming. Think of all those 7-year-olds learning LOGO.
"They might screw up, but it's not going to be something fundamental as less powerful with shorter but multiple keys."
How do you know they know anything at all about crypto?
'what amuses me though is all these/.ers who think they outsmarted the profs. One smart guy said "and not even a white paper? not a good sign."'
We are smart people here. John fucking Carmack posts here from time to time. I'm no John Carmack, but I've got a good mind and I know how to use it. So, yeah, why shouldn't we be able to?
"LOL. It's not like it's some guy living under a basement."
And if it were? What does it matter? The point is the ideas, and we can easily see what's going on here.
I can totally see them screwing this up. Even college professors screw up from time to time. Besides, the main problems are the two that others pointed out: No one will buy it / client side security always fails.
Don't assume that professors are always right. Heck, for all you know, I could be Bruce Schneier in disguise (I'm not, but you get the idea).
"Having tons of keys to encrypt mutliple parts of a piece of music is essentially indistinguishable from just using one really long key to encrypt the same material. Perhaps this helps skirt encryption export laws, but beyond that it really doesn't buy you significantly better protection. "
Actually, it's worse than that. Consider a piece of music with the first half encrypted with a 40 bit key, and the second half with another 40 bit key. It's not the same as one 80 bit key, because each half can be decrypted separately. It's more like a 41 bit key.
Um. Nintendos had multiple processors - sound, sprite, and main (maybe more, I don't recall). They have been emulated for years. The reason you don't see a Saturn emulator is because it had, like, 3 games. -Dave Turner.
I don't think in the US a judge can do that - if there's an innocent interpretation, which is easy to arrange. Especially if there's no evidence that Alice and Bob are cooperating. If Alice claims her data is just a 1/nth XOR of part of the bible, and can prove it, she's safe, and her data is safe. Same with Bob. Mathematically, assuming there's no creation dates (important!), there's no way to know who started with random data, and who XORed against it.
WRT: "Outlaw the system that did this, as it is subversive (which it is!). " If we assume a police state, there's no hope for *any system*. If this is illegal, then encryption is illegal, and we're screwed anyhow. Let's focus on finding security holes in the existing legal system, and improving our system when they fix the holes. This hole is, IMHO, unfixable without destroying the presumption of innocence - and then we might as well move to Sweden:)
I read "The Terminal Experiment" by Robert Sawyer. It was a very good book, but had an premise that I found annoying.
SPOILER WARNING: If you don't want they "mystery" ruined, don't read on. But frankly, unless you have the same biases as the author, you'll never figure out whodunit. Skip to the last section of this review.
Here's the deal: The murderer is the construct without an idea of death. Why? Because without death, there's no hell, and without hell, there's no morality. EVEN THOUGH this person is an atheist. Hi, I'm a *real* atheist - I don't believe in hell, at all - I wasn't even raised Christian. I don't think murder is OK.
END SPOILER:
In short, as little as this guy understands fundamentalists, he understands atheists even less. Which makes me wonder why I should give a fuck about his opinions about religion/theology/etc.
Why not read some author who does know what s/he's talking about, like James Morrow or Mary Doria Russel.
This won't work. It won't work because there's no deniability.
Remeber the last anti-censorship system on/.? The XOR-based system? That had deniability. Consider:
Alice wants to share scientology secrets with the world. Alice looks around, finds Bob's chunk of random data, and XORs the scientology secrets with Bob's data. (In a real situation, she would also use Barry's, Bonnie's, and Billy's, but let's keep it simple). Then she gets her friend Christine to publish to usenet (or whatever) "Hey look: Alice xor Bob = Secrets!"
Helena Kobrin (scientology lawyer) drags Alice and Bob. Alice says "I put random data up, Bob XORed Scientology secrets with my data." Bob says "I put random data up, Alice XORed Scientology secrets with my data." Mathematically, there's no way to tell who did it. There's also an easy way for both Alice and Bob to show that their data is innocent - just have Charlene and Darlene xor bits of the bible with Alice's and Bob's data, respectively, and say "We were just (independently) trying to get the bible into Iran/China/Libya..."
Anyway, this "Eternity service" doesn't have anything like that. Here, Kobrin will say "Is there anything else that this data could have been?" and "Could these have been key shares to any other data?" Alice will have to say "No."
At best, Alice is an ISP - and under the DMCA, she'll have to take it down until it's proved non-infringing (never). In the XOR system, all she needs is Charlene to prove it's innocent.
It is not soon to be free as in speech. According to an e-mail from the author, they have decided not to free blender. One reason is because the author wants to force everyone to use his UI, and he is afraid that others will change it (it's considered pretty cryptic). Another reason (I've heard this one a lot) is that he doesn't want to support it. This seems like something that the Free Software community needs to work on.
1. Register selection is very easy in most cases. Also, I can't think of a machine that has 10 registers, and anyway the NP-Complete problems require N! time.
You're right, mostly. Register allocation is easy in the sense in that graph coloring is fairly fast and is able to roll up allocation, machine idioms and simple copy propagation all in one. It can also do things like web/live range splitting to pack even more variables into one register. Compiler register allocators are pretty damn smart. I doubt a programmer could do better. I didn't follow your "10 registers" comment - what are you getting at?
What I was saying was that the previous poster said "2^10", and I said 10 is a random number, and it wouldn't be 2^10 even if a machine did have 10 registers.
3. The programmer will always at least tie any optimizing compiler. Why? Because the programmer uses the compiler. When I last coded in assembly, I was writing a graphical application. 90% of it was in C. I wrote 100% of it in C, looked for the hot spots, saw what the compiler had done to them, and improved it.
How did you improve it? I don't think a programmer can always tie or beat a highly optimizing compiler. There are too many transformations hidden by the syntax of the high-level language. One might think that the hot spots can be coded in assembly to recover that, but it's very, very difficult to optimize for the pipeline, cache, etc. Not to mention non-portable.
A programmer will always at least tie an optimizing compiler, because the worst the programmer will do is to RUN the optimizing compiler, decide that its output can't be improved, and used it as-is.
It is difficult to optimize - no one ever said it was easy. But tools like Intel's VTune show you how your code will pipeline, etc (disclaimer: I've never used it, because I can't afford it). But yeah, it's possible to consider all of these things, and in some small sections of code, it's worth it. No one writes whole programs in assembly anymore (except 64k intros, but that's a different story).
Non-portable goes without saying - but if you need the speed, you pay the price.
4. How did I improve it? I used SIMD instructions (MMX). I was able to get a 4x increase in speed. *NO COMPILER COULD EVER DO THIS*. There are 2 reasons for this:
This has been addressed by another poster. Vectorizing compilers have existed for a long time.
There are compilers that can do some vector stuff, but not very well. Consider the following code:
int r, g, b, a, p;
a =....;
p = get_pixel (my_bitmap, x, y);
r = (p >> 16) & 0xff; g = (p >> 8) & 0xff; b = p & 0xff;
r = (r * a) >> 8; g= (g * a) >> 8; b = (b * a) >> 8;
p = (r << 16) | (g << 8) | b;
put_pixel (my_bitmap, x, y, p);
This multiplies p by a. I know that a is between 0 and 255. My compiler doesn't. (I could make it an unsigned char, but that would be slower for various reasons, and could cause problems in other sections of my code, where it does go over 255 and is then saturated to 255). Another thing the compiler is not aware of is that the MSbyte of p is irrelavent - it would probably waste time masking it out after the multiply. There's tons of little issues like this (which is related to the second point, but it's more fundamental).
b. A compiler doesn't know what the limits are on data that you'll pass to your function. You do. If you know all data will have an even number of pixels (say), you have an advantage over the compiler, which doesn't know this.
Now this is a valid point. The programmer should concentrate on the high-level algorithm, which should take stuff like this into account.
The code above is the best algorithm I can think of for darkening. I don't think it's possible to do better. But my assembly language version of that code was 2 or 3 times as fast. So yeah, it's not dead yet:)
The 1st amendment changes the text of the constitution, but yeah, it is part of the first amendment:
Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the government for a redress of grievances.
1. Register selection is very easy in most cases. Also, I can't think of a machine that has 10 registers, and anyway the NP-Complete problems require N! time.
2. But that's beside the point, because on a machine with a decent number of registers (anything but x86), you won't run out during a routine, or if you do, the scheduling will be fairly easy.
3. The programmer will always at least tie any optimizing compiler. Why? Because the programmer uses the compiler. When I last coded in assembly, I was writing a graphical application. 90% of it was in C. I wrote 100% of it in C, looked for the hot spots, saw what the compiler had done to them, and improved it.
4. How did I improve it? I used SIMD instructions (MMX). I was able to get a 4x increase in speed. *NO COMPILER COULD EVER DO THIS*. There are 2 reasons for this: a. C doesn't have the syntax to do pixel manipulation SIMD style. My C code was full of shifts and bitmasks (ands). My assembly (when I re-wrote it) had only 2 shifts (one per section of the unrolled loop), and no bitmasks. b. A compiler doesn't know what the limits are on data that you'll pass to your function. You do. If you know all data will have an even number of pixels (say), you have an advantage over the compiler, which doesn't know this.
I don't think you are using the system the way the author intended. I see it as more of a deniability system. It's not hiding the existance of data, it's hiding the legal responsibility for it.
The following situation ignores the author's explicit advice to use more than one pad. Oh well:)
Let's say Alice and Bob want to distribute a chunk of data - decss, for example. Alice makes (or finds) a chunk of random data. Bob XORs decss against Alice's chunk of data. (it's assumed that Alice doesn't leak data by downloading the file - that's one flaw in the system, but it might be fixable - an anonymizing proxy would do it).
Now Bob is distibuting one piece, and Alice is distributing another. Alice goes to the library, or uses an anonymous remailer, or whatever, and informs the world that DeCSS is available at http://www.alice.com/pad-f541e23a41c23b41f23.dat XOR http://www.bob.com/pad-1c23b441ce42bd93c5a.dat.
The cops haul Alice and Bob in, and they are put on trial for violating the DMCA. Alice says "I put up this random data, and Bob XORed DeCSS with it - he's the criminal." Bob says "I put up this random data, and Alice XORed DeCSS with it - she's the criminal." The chances of either being the actual guilty party (from the judge's point of view) are 49.9999999(lots of 9s)/49.99999(lots of 9s). The third possibility is that they're really both innocent and someone just happened to notice, but we'll ignore that. Under US law, I don't think the judge could convict either (in a criminal case).
If Alice wanted more deniability, she could ask Cindy to XOR the first 128k of the bible with her data, and post the result to http://www.cindy.com/pad-46a18b3e8f3b1934613.dat. Now Alice has an excues for her data - it's half of the bible, and it's for the poor Iranians (or Chinese, or whoever is being censored outside the US this week). Bob can do the same with Dave if he wants more deniability. And that's the point of the system. -Dave Turner.
HTML describes the structure of a document, not it's look. It was NEVER intended for graphic designers. Lynx (actually, Links, another text-mode browser, is better) can render HTML perfectly. Yes. Perfectly. But "perfect" just means that it interprets the tags that it chooses to, and displays stuff. But it doesn't necessarily do it the way hot-shot web designers want it to.
That's 'cause hot shot web designers aren't usually trained to design for the web - they design for print and hope it comes out OK on the web. Which is one of the reasons that a lot of sites are damn hard to use. But I digress. The point is that HTML doesn't describe look / feel - it describes structure. Using tables for layout is abusing the intent of tables (that said, I do it on my personal page).
If you don't like the intent of HTML, make some new standard. Don't use PDF - it's proprietary, and supporting proprietary standards gets us into shit like the gif and mp3 affairs.
"All output of this program is a derivative work of this program" would work in some cases. I don't really think it works well in the case of the GIMP, without other modifiers.
Maybe "All output of this program is a derivative work of this program. If you release any modifications to the program that you have made, you are granted copyright over the derivative works."
This is why RMS suggest you use the term Free Software. Free as in Freedom. With Free Software (and software that fits the "Open Source Definition", but they couldn't get a TM on "Open Source"), you have the Freedom to modify and distribute software. "Open Source" doesn't imply this - that's why RMS doesn't like it. I agree. -Dave Turner.
Ah yes, I'm so delighted that the government, a body that masterfully delivers our mail to the wrong address at ever increasing prices, is now placing it's surgeon-like hands into my industry. </i> <BR> Snail mail is delivered more reliably than e-mail. <BR><BR> <I>Where does IIS go? Where does IE go? Where does Bob go? </i> <BR><BR> RTFD (read the fucking decision) <BR><BR> <i> In the mean time, for god's sake, please take the web browser out of the hands of a software monopoly like Microsoft and put it in the hands of a media monopoly like AOL!! </i> <BR><BR> AOL is not a monopoly - and they (Netscape) will still have to compete with IE. No one likes AOL, but there's plenty of room for both browsers in the market. The reason M$ is being broken up is because it didn't let the market decide. <BR><BR> <i>Jim Clark is a real good guy. We should all love him. He bowed to AOL like the French to Hitler. Real dedication. </i> <BR><BR> That was an un-called-for attack. What would you have done, let your company go down in flames? Let your employees stock options, which they <a href =http://www.jwz.org>worked their asses of for</a> become no more than worthless pieces of toilet paper? If AOL had wanted them, it would have taken them over whether they liked it or not - by going peacefully, Netscape did the right thing. <BR><BR> <i> We've been operating under the assumption that we know better than the consumer. IE is on the desktop, so people will use IE. They are incapable of typing in www.netscape.com because they are drooling animals. </i> <BR><BR> Maybe people are too lazy. Maybe they don't know what to do - Would you know enough about cars to buck a muffler monopoly? Even if your muffler were inferior? (maybe you are a car buff and would - I wouldn't). <BR><BR> <I> On AOL, try changing email clients. Or web clients. Or FTP. Or anything. </i> <BR><BR> If people are smart enough to type in www.netscape.com from MSIE, they're smart enough to type in www.microsoft.com from NS. <BR><BR> <I>This anti-trust trial is irrelevant. IBM was evil. It fell. MS was evil. It is falling. </i> <BR><BR> Yeah, it's falling - with the government's help. <i>AOL is today's evil. And it's a far more formidable enemy. For millions, they control Internet access. For virtually everybody, they control content and information. </i> <BR><BR> Content is easy to find - there's plenty of it that isn't controlled by T/W. AOL isn't a monopoly. M$ is. To say that "AOL controls internet access" is a joke - switching ISPs is about as easy as switching browsers. <BR><BR> <i> Nobody here wants the government getting involved in the Internet. Yet everybody wants the government involved in the software industry. (That makes sense. Since obviously the Internet doesn't use software.) </i> <BR><BR> If you can't tell the difference between a browser and a web site, that's your problem. What we want is what we think is right. I'm not going to speak for anyone else, but I think that the anti-trust laws are good and were well applied here, and that the First Ammendment is good and was well applied in the CDA case. So, yeah, there is a difference. <BR><BR> <i> Everybody here rants how Linux is going to kick Windows' ass. Linux rules!! Yeah! Yet you have to enlist the government to slay Windows because Linux can't do it. </i> <BR><BR> Linux was not involved in this. Most Free Software hackers agree that Linux (in general) is more difficult to use for a normal user for desktop tasks (e.g. word processing). This was a case about desktop OSs. If this case were happening 5 years from now, yeah, Linux would be involved - but it's not. No one claims Linux is perfect or complete - that's why Alan Cox still has a job:) <BR><BR> <I> OPEN SOURCE! OPEN SOURCE! I won't run Windows because it's not open source! But let me haul ass to CompUSA to buy my closed-source copy of Diablo II and WordPerfect 2000. </i> <BR><BR> Some Slashdotters care about Free Software enough not to use proprietary software at all. Those people aren't buying Diablo. Some only care about Open Source software, and will use whatever works - they're buying Diablo. I don't see what your point is - did you really thing you could have a million people agree on everything? <BR><BR> <I> There's no fragmentation in the Linux world!! Nooo!!! But don't run that program, because it requires Xfree 32.2353, but it only comes standard with Red Hat 8.3. But if you have SuSE 5.2389 running the 2.44 kernel you need to download the 9.87 xstuff. Well, don't worry, because once you have everything, you need to copy it into folder/misc, unless you run Debian, where you put it in/misc/folder. But on Red Hat, you don't use that file. Linux supports you graphics card, but only Mandrake 663.3 has the right driver. If you have Corel, you won't have much luck. But Corel's based off of Debian, but the same software and desktop isn't bundled, so don't bother typing in that command.... </i> <BR><BR> If you're going to make this claim, try to use real examples. I haven't had this sort of problem. Then again, I know what I'm doing. If you don't like the lack of standards, get involved with the LSB project. <BR><BR> <i>Hey assholes, how about fixing the damn bug that crashes Netscape every ten minutes? </i> <BR><BR> 1. Why should we - because you want us to? <BR> 2. We can't. Netscape is not Free Software. <BR> 3. We're working on it. The fix is called Mozilla. <BR> <BR> <i>And why the fuck do you want Linux on a toaster anyway? </i> <BR><BR> For the same reason people climb Everest. Because it's there. (Or, because it's fun - y'know, fun, that thing people do for entertainment). <BR><BR> <i>BUSINESS IS EVIL, eh Katz? I bet you sure jumped at the chance to get some of that hot VA Linux IPO. </i> <BR><BR> Katz didn't write this article. Also, Katz doesn't say that business is evil. You're thinking of Marx. Katz says they corporatism is evil. <BR><BR><BR> And now that I'm done responding to that pile of crap, I'll say this: <STRONG>DO YOUR FUCKING RESEARCH.</STRONG> <BR><BR> -Dave Turner.
Nethack isn't in the same genre as FF7. You should probably be comparing it to Diablo. And, given that it was started 15 years ago, it is Diablo, but more fun. Why is it more fun? Because it has more secrets, more items, more interaction with the world, more little jokes, etc. In other words, it's a vastly more complex world. Without graphics. But there are plenty of graphic geeks out there - again, so one day, we'll see graphical Free games. One reason we don't see free graphical games is that most free game groups don't groups don't see art as important. That's a problem - as programmers, they can't see the art side. At Worldforge, we're doing things differently. We see art as one of the key pieces of a world, and as a result have recruited artists from Day One.
The problem with your capitalist stance is that it doesn't describe the real world. It describes a world in which people don't make games to have fun. But in the real world, having fun is the only reason to make games - if you just want to make cash, you're not going to be a good game designer. Game design is all about vision. Of course, technical skill is required - witness Daikatana. So if people are making games for fun anyway, why shouldn't people do it for free? Normal people, with normal jobs, who want a fun thing to do on a saturday afternoon.
Heh, free as in speech, not as in beer. Repeat it until you understand it. If you don't have time to learn Linux, then computers are probably not your job. Good. There are plenty of cool people who aren't geeks. Linux is still free to them. Why? They can give copies to their geeky friends, pay geeks to make Linux easy, or use Linux in their organizations, all without worrying about stuff like monopolies, reverse engineering, and shrink-wrap licensing.
Slashdot Mirror
"You could just as well call Perl a clean simple well designed language with an easy to learn syntax, if you're going to distort the facts so grossly. "
:).
As a Perl hacker, I resent this. Perl is nothing if not easy to learn. You can code Perl that looks like C, if that's what you're used to (I was). You can code Perl that looks like Sed, or BASIC, or shell scripting - If you know one of these, Perl just requires adjusting a few switches in your head.
Effective Perl programs can be written only using a small subset of the language. Don't like (or haven't learned) its OO? Don't use it.
Perl is also simple. Instead of forcing you to tell it exactly what to do, it has sensible defaults, like $_. It does what you expect - and if you prefer something else, you can make it do something else.
WRT your article, it was dead on ten years ago. Now, X has a whole different set of problems
To plug NeWS as a solution is ridiculous. NeWS is dead. It could only come back if Sun freed the code, and if dedicated hackers ported it to Linux and made an X compatibility layer. And they won't do that, because they don't understand Freedom. We can't use a windowing system that's not Free (as in speech), because we can't improve it, or fix its bugs, or port it to our processor/OS/video card.
-Dave Turner.
"If a kid can't read, he can't use the Internet."
But if a kid can read a little, the net can help them learn more. Not through "remote education", or whatever, but because people want to read what's out there. I know there are plenty of people who learned basic english in school, and learned more by hanging out on the net.
"If a kid can't do better than rudimentary math, she can't program."
Not really... I learned sines and cosine through programming an Asteroids game in QBASIC. I've also learned about Voronoi diagrams and other stuff through programming. Think of all those 7-year-olds learning LOGO.
So, the net can help these countries.
-Dave Turner.
The verb to sex means to give a sex to something. Consider the novel "Sexing the Cherry."
But if you insist on sex meaning fucking, I think there are other verbs that need helper verbs (have) to make sense.
-Dave Turner.
"They might screw up, but it's not going to be something fundamental as less powerful with shorter but multiple keys."
/.ers who think they outsmarted the profs. One smart guy said "and not even a white paper? not a good sign."'
How do you know they know anything at all about crypto?
'what amuses me though is all these
We are smart people here. John fucking Carmack posts here from time to time. I'm no John Carmack, but I've got a good mind and I know how to use it. So, yeah, why shouldn't we be able to?
"LOL. It's not like it's some guy living under a basement."
And if it were? What does it matter? The point is the ideas, and we can easily see what's going on here.
-Dave Turner.
I can totally see them screwing this up. Even college professors screw up from time to time. Besides, the main problems are the two that others pointed out: No one will buy it / client side security always fails.
Don't assume that professors are always right. Heck, for all you know, I could be Bruce Schneier in disguise (I'm not, but you get the idea).
-Dave Turner.
"Having tons of keys to encrypt mutliple parts of a piece of music is essentially indistinguishable from just using one really long key to encrypt the same material. Perhaps this helps skirt encryption export laws, but beyond that it really doesn't buy you significantly better protection. "
Actually, it's worse than that. Consider a piece of music with the first half encrypted with a 40 bit key, and the second half with another 40 bit key. It's not the same as one 80 bit key, because each half can be decrypted separately. It's more like a 41 bit key.
-Dave Turner.
You don't really need to read the ROM - you can download them from various sites (sorta like warez).
NESticle is one emulator, and there are a bunch of others.
-Dave Turner.
Um. Nintendos had multiple processors - sound, sprite, and main (maybe more, I don't recall). They have been emulated for years. The reason you don't see a Saturn emulator is because it had, like, 3 games.
-Dave Turner.
I don't think in the US a judge can do that - if there's an innocent interpretation, which is easy to arrange. Especially if there's no evidence that Alice and Bob are cooperating. If Alice claims her data is just a 1/nth XOR of part of the bible, and can prove it, she's safe, and her data is safe. Same with Bob. Mathematically, assuming there's no creation dates (important!), there's no way to know who started with random data, and who XORed against it.
:)
WRT: "Outlaw the system that did this, as it is subversive (which it is!). "
If we assume a police state, there's no hope for *any system*. If this is illegal, then encryption is illegal, and we're screwed anyhow. Let's focus on finding security holes in the existing legal system, and improving our system when they fix the holes. This hole is, IMHO, unfixable without destroying the presumption of innocence - and then we might as well move to Sweden
-Dave Turner.
I read "The Terminal Experiment" by Robert Sawyer. It was a very good book, but had an premise that I found annoying.
SPOILER WARNING:
If you don't want they "mystery" ruined, don't read on. But frankly, unless you have the same biases as the author, you'll never figure out whodunit. Skip to the last section of this review.
Here's the deal: The murderer is the construct without an idea of death. Why? Because without death, there's no hell, and without hell, there's no morality. EVEN THOUGH this person is an atheist. Hi, I'm a *real* atheist - I don't believe in hell, at all - I wasn't even raised Christian. I don't think murder is OK.
END SPOILER:
In short, as little as this guy understands fundamentalists, he understands atheists even less. Which makes me wonder why I should give a fuck about his opinions about religion/theology/etc.
Why not read some author who does know what s/he's talking about, like James Morrow or Mary Doria Russel.
-Dave Turner.
It is more or less linking, and with a proper server, it could be linking. Consider a mail bot that accepts emails with the subject:
URL;URL;...;URL
The mail bot downloads the files at these URLs, XORs them together, and mails them out to the requestor.
So, a mailto URL (a link) is enough to get the files.
-Dave Turner.
That's why there's Betty and Billy and Bilbo and ... With 5-8 files XORed together, it's much safer.
-Dave Turner.
This won't work. It won't work because there's no deniability.
/.? The XOR-based system? That had deniability. Consider:
Remeber the last anti-censorship system on
Alice wants to share scientology secrets with the world. Alice looks around, finds Bob's chunk of random data, and XORs the scientology secrets with Bob's data. (In a real situation, she would also use Barry's, Bonnie's, and Billy's, but let's keep it simple). Then she gets her friend Christine to publish to usenet (or whatever) "Hey look: Alice xor Bob = Secrets!"
Helena Kobrin (scientology lawyer) drags Alice and Bob. Alice says "I put random data up, Bob XORed Scientology secrets with my data." Bob says "I put random data up, Alice XORed Scientology secrets with my data." Mathematically, there's no way to tell who did it. There's also an easy way for both Alice and Bob to show that their data is innocent - just have Charlene and Darlene xor bits of the bible with Alice's and Bob's data, respectively, and say "We were just (independently) trying to get the bible into Iran/China/Libya..."
Anyway, this "Eternity service" doesn't have anything like that. Here, Kobrin will say "Is there anything else that this data could have been?" and "Could these have been key shares to any other data?" Alice will have to say "No."
At best, Alice is an ISP - and under the DMCA, she'll have to take it down until it's proved non-infringing (never). In the XOR system, all she needs is Charlene to prove it's innocent.
That's my 2 cents, anyhow.
-Dave Turner.
It is not soon to be free as in speech. According to an e-mail from the author, they have decided not to free blender. One reason is because the author wants to force everyone to use his UI, and he is afraid that others will change it (it's considered pretty cryptic). Another reason (I've heard this one a lot) is that he doesn't want to support it. This seems like something that the Free Software community needs to work on.
-Dave Turner.
You mean other than .gov, .mil, and .int?
-Dave Turner.
1. Register selection is very easy in most cases. Also, I can't think of a machine that has 10 registers, and anyway the NP-Complete problems require N! time.
....;
:)
You're right, mostly. Register allocation is easy in the sense in that graph coloring is fairly fast and is able to roll up allocation, machine idioms and simple copy propagation all in one. It can also do things like web/live range splitting to pack even more variables into one register. Compiler register allocators are pretty damn smart. I doubt a programmer could do better. I didn't follow your "10 registers" comment - what are you getting at?
What I was saying was that the previous poster said "2^10", and I said 10 is a random number, and it wouldn't be 2^10 even if a machine did have 10 registers.
3. The programmer will always at least tie any optimizing compiler. Why? Because the programmer uses the compiler. When I last coded in assembly, I was writing a graphical application. 90% of it was in C. I wrote 100% of it in C, looked for the hot spots, saw what the compiler had done to them, and improved it.
How did you improve it? I don't think a programmer can always tie or beat a highly optimizing compiler. There are too many transformations hidden by the syntax of the high-level language. One might think that the hot spots can be coded in assembly to recover that, but it's very, very difficult to optimize for the pipeline, cache, etc. Not to mention non-portable.
A programmer will always at least tie an optimizing compiler, because the worst the programmer will do is to RUN the optimizing compiler, decide that its output can't be improved, and used it as-is.
It is difficult to optimize - no one ever said it was easy. But tools like Intel's VTune show you how your code will pipeline, etc (disclaimer: I've never used it, because I can't afford it). But yeah, it's possible to consider all of these things, and in some small sections of code, it's worth it. No one writes whole programs in assembly anymore (except 64k intros, but that's a different story).
Non-portable goes without saying - but if you need the speed, you pay the price.
4. How did I improve it? I used SIMD instructions (MMX). I was able to get a 4x increase in speed. *NO COMPILER COULD EVER DO THIS*. There are 2 reasons for this:
This has been addressed by another poster. Vectorizing compilers have existed for a long time.
There are compilers that can do some vector stuff, but not very well. Consider the following code:
int r, g, b, a, p;
a =
p = get_pixel (my_bitmap, x, y);
r = (p >> 16) & 0xff;
g = (p >> 8) & 0xff;
b = p & 0xff;
r = (r * a) >> 8;
g= (g * a) >> 8;
b = (b * a) >> 8;
p = (r << 16) | (g << 8) | b;
put_pixel (my_bitmap, x, y, p);
This multiplies p by a. I know that a is between 0 and 255. My compiler doesn't. (I could make it an unsigned char, but that would be slower for various reasons, and could cause problems in other sections of my code, where it does go over 255 and is then saturated to 255). Another thing the compiler is not aware of is that the MSbyte of p is irrelavent - it would probably waste time masking it out after the multiply. There's tons of little issues like this (which is related to the second point, but it's more fundamental).
b. A compiler doesn't know what the limits are on data that you'll pass to your function. You do. If you know all data will have an even number of pixels (say), you have an advantage over the compiler, which doesn't know this.
Now this is a valid point. The programmer should concentrate on the high-level algorithm, which should take stuff like this into account.
The code above is the best algorithm I can think of for darkening. I don't think it's possible to do better. But my assembly language version of that code was 2 or 3 times as fast. So yeah, it's not dead yet
-Dave Turner.
The 1st amendment changes the text of the constitution, but yeah, it is part of the first amendment:
Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the government for a redress of grievances.
-Dave Turner.
You are wrong. Here's why:
:)
1. Register selection is very easy in most cases. Also, I can't think of a machine that has 10 registers, and anyway the NP-Complete problems require N! time.
2. But that's beside the point, because on a machine with a decent number of registers (anything but x86), you won't run out during a routine, or if you do, the scheduling will be fairly easy.
3. The programmer will always at least tie any optimizing compiler. Why? Because the programmer uses the compiler. When I last coded in assembly, I was writing a graphical application. 90% of it was in C. I wrote 100% of it in C, looked for the hot spots, saw what the compiler had done to them, and improved it.
4. How did I improve it? I used SIMD instructions (MMX). I was able to get a 4x increase in speed. *NO COMPILER COULD EVER DO THIS*. There are 2 reasons for this:
a. C doesn't have the syntax to do pixel manipulation SIMD style. My C code was full of shifts and bitmasks (ands). My assembly (when I re-wrote it) had only 2 shifts (one per section of the unrolled loop), and no bitmasks.
b. A compiler doesn't know what the limits are on data that you'll pass to your function. You do. If you know all data will have an even number of pixels (say), you have an advantage over the compiler, which doesn't know this.
Anyway, that's why you're wrong
-Dave Turner.
I don't think you are using the system the way the author intended. I see it as more of a deniability system. It's not hiding the existance of data, it's hiding the legal responsibility for it.
:)
The following situation ignores the author's explicit advice to use more than one pad. Oh well
Let's say Alice and Bob want to distribute a chunk of data - decss, for example. Alice makes (or finds) a chunk of random data. Bob XORs decss against Alice's chunk of data. (it's assumed that Alice doesn't leak data by downloading the file - that's one flaw in the system, but it might be fixable - an anonymizing proxy would do it).
Now Bob is distibuting one piece, and Alice is distributing another. Alice goes to the library, or uses an anonymous remailer, or whatever, and informs the world that DeCSS is available at http://www.alice.com/pad-f541e23a41c23b41f23.dat XOR http://www.bob.com/pad-1c23b441ce42bd93c5a.dat.
The cops haul Alice and Bob in, and they are put on trial for violating the DMCA. Alice says "I put up this random data, and Bob XORed DeCSS with it - he's the criminal." Bob says "I put up this random data, and Alice XORed DeCSS with it - she's the criminal." The chances of either being the actual guilty party (from the judge's point of view) are 49.9999999(lots of 9s)/49.99999(lots of 9s). The third possibility is that they're really both innocent and someone just happened to notice, but we'll ignore that. Under US law, I don't think the judge could convict either (in a criminal case).
If Alice wanted more deniability, she could ask Cindy to XOR the first 128k of the bible with her data, and post the result to http://www.cindy.com/pad-46a18b3e8f3b1934613.dat. Now Alice has an excues for her data - it's half of the bible, and it's for the poor Iranians (or Chinese, or whoever is being censored outside the US this week). Bob can do the same with Dave if he wants more deniability. And that's the point of the system.
-Dave Turner.
HTML describes the structure of a document, not it's look. It was NEVER intended for graphic designers. Lynx (actually, Links, another text-mode browser, is better) can render HTML perfectly. Yes. Perfectly. But "perfect" just means that it interprets the tags that it chooses to, and displays stuff. But it doesn't necessarily do it the way hot-shot web designers want it to.
That's 'cause hot shot web designers aren't usually trained to design for the web - they design for print and hope it comes out OK on the web. Which is one of the reasons that a lot of sites are damn hard to use. But I digress. The point is that HTML doesn't describe look / feel - it describes structure. Using tables for layout is abusing the intent of tables (that said, I do it on my personal page).
If you don't like the intent of HTML, make some new standard. Don't use PDF - it's proprietary, and supporting proprietary standards gets us into shit like the gif and mp3 affairs.
-Dave Turner.
"All output of this program is a derivative work of this program" would work in some cases. I don't really think it works well in the case of the GIMP, without other modifiers.
Maybe "All output of this program is a derivative work of this program. If you release any modifications to the program that you have made, you are granted copyright over the derivative works."
-Dave Turner.
This is why RMS suggest you use the term Free Software. Free as in Freedom. With Free Software (and software that fits the "Open Source Definition", but they couldn't get a TM on "Open Source"), you have the Freedom to modify and distribute software. "Open Source" doesn't imply this - that's why RMS doesn't like it. I agree.
-Dave Turner.
Ah yes, I'm so delighted that the government, a body that masterfully delivers our mail to the wrong address at ever increasing prices, is now placing it's surgeon-like hands into my industry. :) /misc/folder. But on Red Hat, you don't use that file. Linux supports you graphics card, but only Mandrake 663.3 has the right driver. If you have Corel, you won't have much luck. But Corel's based off of Debian, but the same software and desktop isn't bundled, so don't bother typing in that command.... </i>
</i>
<BR>
Snail mail is delivered more reliably than e-mail.
<BR><BR>
<I>Where does IIS go? Where does IE go? Where does Bob go? </i>
<BR><BR>
RTFD (read the fucking decision)
<BR><BR>
<i> In the mean time, for god's sake, please take the web browser out of the hands of a software monopoly like Microsoft and put it in the hands of a media monopoly like AOL!! </i>
<BR><BR>
AOL is not a monopoly - and they (Netscape) will still have to compete with IE. No one likes AOL, but there's plenty of room for both browsers in the market. The reason M$ is being broken up is because it didn't let the market decide.
<BR><BR>
<i>Jim Clark is a real good guy. We should all love him. He bowed to AOL like the French to Hitler. Real dedication. </i>
<BR><BR>
That was an un-called-for attack. What would you have done, let your company go down in flames? Let your employees stock options, which they <a href =http://www.jwz.org>worked their asses of for</a> become no more than worthless pieces of toilet paper? If AOL had wanted them, it would have taken them over whether they liked it or not - by going peacefully, Netscape did the right thing.
<BR><BR>
<i>
We've been operating under the assumption that we know better than the consumer. IE is on the desktop, so people will use IE. They are incapable of typing in www.netscape.com because they are drooling animals. </i>
<BR><BR>
Maybe people are too lazy. Maybe they don't know what to do - Would you know enough about cars to buck a muffler monopoly? Even if your muffler were inferior? (maybe you are a car buff and would - I wouldn't).
<BR><BR>
<I> On AOL, try changing email clients. Or web clients. Or FTP. Or anything. </i>
<BR><BR>
If people are smart enough to type in www.netscape.com from MSIE, they're smart enough to type in www.microsoft.com from NS.
<BR><BR>
<I>This anti-trust trial is irrelevant. IBM was evil. It fell. MS was evil. It is falling. </i>
<BR><BR>
Yeah, it's falling - with the government's help.
<i>AOL is today's evil. And it's a far more formidable enemy. For millions, they control Internet access. For virtually everybody, they control content and information. </i>
<BR><BR>
Content is easy to find - there's plenty of it that isn't controlled by T/W. AOL isn't a monopoly. M$ is. To say that "AOL controls internet access" is a joke - switching ISPs is about as easy as switching browsers.
<BR><BR>
<i> Nobody here wants the government getting involved in the Internet. Yet everybody wants the government involved in the software industry. (That makes sense. Since obviously the Internet doesn't use software.) </i>
<BR><BR>
If you can't tell the difference between a browser and a web site, that's your problem. What we want is what we think is right. I'm not going to speak for anyone else, but I think that the anti-trust laws are good and were well applied here, and that the First Ammendment is good and was well applied in the CDA case. So, yeah, there is a difference.
<BR><BR>
<i> Everybody here rants how Linux is going to kick Windows' ass. Linux rules!! Yeah! Yet you have to enlist the government to slay Windows because Linux can't do it. </i>
<BR><BR>
Linux was not involved in this. Most Free Software hackers agree that Linux (in general) is more difficult to use for a normal user for desktop tasks (e.g. word processing). This was a case about desktop OSs. If this case were happening 5 years from now, yeah, Linux would be involved - but it's not. No one claims Linux is perfect or complete - that's why Alan Cox still has a job
<BR><BR>
<I>
OPEN SOURCE! OPEN SOURCE! I won't run Windows because it's not open source! But let me haul ass to CompUSA to buy my closed-source copy of Diablo II and WordPerfect 2000. </i>
<BR><BR>
Some Slashdotters care about Free Software enough not to use proprietary software at all. Those people aren't buying Diablo. Some only care about Open Source software, and will use whatever works - they're buying Diablo. I don't see what your point is - did you really thing you could have a million people agree on everything?
<BR><BR>
<I> There's no fragmentation in the Linux world!! Nooo!!! But don't run that program, because it requires Xfree 32.2353, but it only comes standard with Red Hat 8.3. But if you have SuSE 5.2389 running the 2.44 kernel you need to download the 9.87 xstuff. Well, don't worry, because once you have everything, you need to copy it into folder/misc, unless you run Debian, where you put it in
<BR><BR>
If you're going to make this claim, try to use real examples. I haven't had this sort of problem. Then again, I know what I'm doing. If you don't like the lack of standards, get involved with the LSB project.
<BR><BR>
<i>Hey assholes, how about fixing the damn bug that crashes Netscape every ten minutes? </i>
<BR><BR>
1. Why should we - because you want us to? <BR>
2. We can't. Netscape is not Free Software. <BR>
3. We're working on it. The fix is called Mozilla. <BR>
<BR>
<i>And why the fuck do you want Linux on a toaster anyway? </i>
<BR><BR>
For the same reason people climb Everest. Because it's there. (Or, because it's fun - y'know, fun, that thing people do for entertainment).
<BR><BR>
<i>BUSINESS IS EVIL, eh Katz? I bet you sure jumped at the chance to get some of that hot VA Linux IPO. </i>
<BR><BR>
Katz didn't write this article. Also, Katz doesn't say that business is evil. You're thinking of Marx. Katz says they corporatism is evil.
<BR><BR><BR>
And now that I'm done responding to that pile of crap, I'll say this: <STRONG>DO YOUR FUCKING RESEARCH.</STRONG>
<BR><BR>
-Dave Turner.
Nethack isn't in the same genre as FF7. You should probably be comparing it to Diablo. And, given that it was started 15 years ago, it is Diablo, but more fun. Why is it more fun? Because it has more secrets, more items, more interaction with the world, more little jokes, etc. In other words, it's a vastly more complex world. Without graphics. But there are plenty of graphic geeks out there - again, so one day, we'll see graphical Free games. One reason we don't see free graphical games is that most free game groups don't groups don't see art as important. That's a problem - as programmers, they can't see the art side. At Worldforge, we're doing things differently. We see art as one of the key pieces of a world, and as a result have recruited artists from Day One.
The problem with your capitalist stance is that it doesn't describe the real world. It describes a world in which people don't make games to have fun. But in the real world, having fun is the only reason to make games - if you just want to make cash, you're not going to be a good game designer. Game design is all about vision. Of course, technical skill is required - witness Daikatana. So if people are making games for fun anyway, why shouldn't people do it for free? Normal people, with normal jobs, who want a fun thing to do on a saturday afternoon.
That's why we're in it - fun.
-Dave Turner.
Heh, free as in speech, not as in beer. Repeat it until you understand it. If you don't have time to learn Linux, then computers are probably not your job. Good. There are plenty of cool people who aren't geeks. Linux is still free to them. Why? They can give copies to their geeky friends, pay geeks to make Linux easy, or use Linux in their organizations, all without worrying about stuff like monopolies, reverse engineering, and shrink-wrap licensing.
-Dave Turner.