And if they tried to communicate with each other I'm sure they would be found out quickly, as that matches the pattern of a DDOS system.
There are several ways of communication. What about a virus/worm that just tries to infect pograms/systems, but when it succesfuly infects a program/system, it searches for other copies of itself within that program/system, and exchanges data with them. Next, new copies of the virus/worm are sent out with the new data.
If we use a genetic algorithm, the data would be the "genome", and we would have viruses/worms not only changing each generation, but actually evolving into more effective ones (from the viewpoint of the virus/worm), spreading information by "mating". -><- Grand Reverence Zan Zu, AB, DD, KSC
Ok, we all know France has got dirty hands too. But why is reporting you're going to investigate on other nations' spying activities wrong? Lets look at it the other way around, imagine an eavesdropping network, run by France, Canada and some french speaking states in Africa. I bet you the US would investigate it, at the very least!
Another point is that discussions like this obscure the fact that ECHELON eavesdrops on everyone -yes, this means you too- it makes no difference wether you live in France or the US. Sure the first nations to complain about it are the ones that have some issues with the US, but this does not mean ECHELON all of a sudden becomes a good thing because it can be used to spy on them.
Then there is the paranoid "if they wheren't hiding anything, they wouldn't be whining" argument. It has been discussed over and over in endless crypto threads. This isn't any different.
French foreign policy might not be the best in the world, but the US keeps "pissing off" (to put it in your terms) various nations all over the world too, with its endless trade wars and embargos. The only difference is the US being capable of and willing to take firm, even violent, action to enforce its policies. This raises the question wether it's better to be an 'irritating whiner' like France or a 'big bully' like the US. -><- Grand Reverence Zan Zu, AB, DD, KSC
Igor Aleksander postulates consciousness in NNs is based on awareness of World and of Self. As I see it the mindpixel project would train a NN in world-awareness, but what about the self-awareness? (Or don't you agree with Aleksander at all?) -><- Grand Reverence Zan Zu, AB, DD, KSC
...but a simpler LZ78 algorithm could probably be made compatible (with some tricks).
That would be more difficult than it sounds, IMHO. Have a look at the data compression reference center's page on the differences between LZ78 and LZW.
You might end up coding another LZWx algorithm, and the whole family of algorithms (the method) is patented (free for non-commercial use tho). -><- Grand Reverence Zan Zu, AB, DD, KSC
So why do I have to use the 9 bit to specify its uncompressed, if I'm not using a compression algorithm? If the file would be not compressed, I would have to set *one* "not compressed" flag for the entire file, not for each octet in the file. Since this is *not* possible with GIF I can't avoid my data being piped trough the decompression algorithm. So I'm stuck with using the algorithm, even if I dont want to compress... -><- Grand Reverence Zan Zu, AB, DD, KSC
Uncompressed GIF's are possible. LZW uses a system wwhere a number greater than 256 means a compressed or control code, and a number less than 256 means a plain colour. Uncompressed GIF's just use values less than 256. Its actually anti compressed because it needs at least 9 bits to store each value.
And how exactly is this "uncompressed"? You are still (ab)using LZW compression; only now you've created a bigger file... -><- Grand Reverence Zan Zu, AB, DD, KSC
But why? Even ignoring the influx of money, and assuming that you were actually from Tuvalu and had internet access, why would you care whether your domain name ended in.tv ,.net,.uk,.us or anything else?
But why? Even ignoring the $50 million, assuming dotTv is not located in Tuvalu, knowing.tv is the Tuvalu TLD, why in the world would they want their domain name to end in.tv?
Doesn't seem like a valid argument, as this goes both ways. Why should big buck be allowed to break the TLD conventions?
What would you say if microsoft where to buy.gov? -><- Grand Reverence Zan Zu, AB, DD, KSC
Like most (all?) protests, this is all fluff and no substance. By taking some action, no matter how silly, these guys make themselves feel better about the situation. A better course of action would be to work out a solution to prevent DDoS attacks from succeeding.
There is a solution to DDoS, but it has a huge problem: you have to do it on the network(s) from which the malicious packets are being sent. You can't do much about it on the receiving side, the packets are going to flood your lines to/from the Internet before they reach your firewall/router. DDoS is going to be around until every admin on the Internet plugs his network holes and until every user on the Internet secures his machines.
I bet you the ircops have sent a zillion emails to the admins of the offending networks, but what more can they do? Sue the hell out of them and disable large parts of the Internet in the process? Hmm... doesn't sound to good, lets try to make the general public aware of the situation. They tried to do just that by putting info on the situation on their webservers, in their/motd's and you have what. Guess what? It didn't work. So now they are going to do the one thing they can do to get the attention of the public.
Sure this is a big disappointment to all the "well behaved" irc users, but if it is the only way of making them aware of the situation, let them do it. (And all of a sudden, everyone is indeed aware of the situation, and all up in arms...) -><- Grand Reverence Zan Zu, AB, DD, KSC
The way I see it this is hardly going to affect anything. Your typical "script kiddie" is gonna try to get onto ircnet, fail and just go over to efnet for the day without even noticing.
I think the ircops are aiming more for the 1rc-kl4nz with huge bot/flood-nets. Your typical script kiddie doesn't do much harm on irc (he's dossing clients, not servers), its mostly the clans that do the damage (they dos the servers to generate netsplits).
For these clans, it will be a real pain to move their 300+ bots to an other irc net, just for one day. Without their botnets, they are not likely to do massive takeovers, especially on other irc networks as they will challenge clans over there (of unknown sk1llz level) to come and have a fight on their home turf.
So you have testdriven an IP stack that isn't even released yet? Or, if you have tested the 2.3.99-pre3 and the latest BSD IP stacks, could you post (a link to) your test results?
It does matter. When you have something that a law, however unjust, says you can't have, then it's no longer the "land of the free", it's the "land of the underground."
What underground? US laws don't have any jurisdiction in other parts of the world. Hey, guns are prohibited by law around here, so can I go on and sue all of you US citizens carrying illegal underground guns?
Serious question: What if I put up some sort of proxy service to CPhack outside of US territory; would US citizens be allowed to use it, or is that illegal too?
Yes, he wants Darwin in the BSD family, not Mac on Intel. From his avogato diary:
Though Darwin's role in the world is as the guts of Mac OS X, it is shaping up to be a viable BSD platform on its own. It take advantage of some of the great work underway in the NetBSD, FreeBSD, and Linux communities, as well as several of the neat things we're working on at Apple.
If we can offer fully-functional ASP and IIS on Linux, we can start to clean up the last pockets of resistance.
And while whe're at it, why stop at full functionality? Why stop at reimplementing micros~1 applications and not use their tactics? Why not embrace & extend our GPLled versions and put in some nice new features?
If the extensions are any good, micros~1 will be forced to implement them...
Things aren't always black and white, which is why I think it's a good idea to be objective when posting news. That's all.
Me too:)
But seriously, do you expect unbiased news on a site that clearly choses sides in the OS holy wars? After all/. claims to be "news for nerds" and since most "nerds" seem to be fighting Microsoft, whats/. gotta do? Tell them how great MSFT is?
What bothers me most is not the holy wars (they've been going on for ages), but that everyone is taking themselves so damned serious. Hacker culture always was ha ha, only serious. We seem to have forgotten about the ha ha. Just look at the quality of the average troll on/. (there hardly are any good ones, they're mostly flamebait labled as "troll"; and if you by any chance find a little gem its mostly marked as "funny").
As the hacker community grows, so does the number of wannabees, who think its all about being a full-time genius and a zealot...
Hey, I read/. because its fun, not because its the fastest and most reliable source of news (no offence guys).
Why not just report the news and let people discuss them here instead of spoon feeding them anti Microsoft propaganda every chance you get?
Ok, discuss this. Its the everchanging list of Top 30 Windows 2000 Bugs. This one is my favorite:
According to Microsoft, Windows 2000 Professional may hang after you install Microsoft IntelliPoint 2.2. Microsoft says that pressing CTRL-ALT-DELETE will not help. To resolve this problem, Microsoft says you have to reinstall Windows 2000 Professional.
Looking at the success of the open source movement, one might say full-time telecommuting works. Ok, it isn't exactly working for a company, but the problems with it a are very similar (CVS, project forking, other type of communication/management needed, clash of egos).
Some companies will not hire telecommuters because of these problems, other companies like the cost savings aspect of it (they don't need large office buildings).
Well, there are some nice trolls here, you just gotta look close and THINK about what you read... There is no such thing as an obvious troll.
I agree with you on goatse.cx being a big asshole; but is it a troll?
Lets call these guys "spam kiddies" or something and moderate them down. At the same time moderate up real trolls, just to show them how its done (if they are smart enough to figure out its a troll in the first place).
Ok, lets say our algorithmical solution isn't int i >= 100000000; but BigNum i >= CONSTANT_WITH_250_DIGITS.
You can wait your 30 years for hardware good enough, then run your code for 5 years or so, wondering just how fast you could do it in another 30 years. Meanwhile my code executes in under 5 minutes on a 386 16Mhz...
As to the hardware, I seem to remember something in neural networks theory that states that no matter how big you make your NN, it doesn't work right until you randomize its inputs. IOW we can make our hardware bigger, we can make it faster, but until the stuff we run on it will be a whole lot more complex than it is today, we'll never reach artificial human-like intelligence.
As other posters have pointed out, human-like AI can't be done with brute force (alone). You are suggesting we can do it with todays brute force methods if we do it fast enough and have enough memory to store state information.
<rant>If we make a monkey 1000 times bigger and 1000 times faster it will still be a monkey. It might run us into the ground before we see it, but this won't be because its any smarter now.<rant>
If we put out monkey (lets call it Willy) behind a typewriter it might produce maybe one interresting piece of literature in its lifetime, but is Willy intelligent just because he punches keys at the speed of light and got lucky once?
int main() {
int i= 0; while (i < 2) { ++i; } while (i < 100000000) { i*= i; } return 0; }
I'm talking about a different approach to the problem, a different algorithm in the sense that it changes the time-complexity of the problem. (We all know the classical example of naive vs. smart sorting algorithms.)
In our chess example I would build in heuristics and pattern matching. Actually, the winner app in my previous post was nearly all heuristics, which made it very strong against computer players. Its main tactics was to complicate positions as quickly as possible and make unexpected (non-optimal) moves. Ironicly, this is exactly how most human players try to beat todays heavy metal...
Yeah? Let's see you create a (arbitrary type of game here) chess program that will run on a 386 that wouldn't get pounded by Deep Blue.
If I am to use the same algorithm you're right, but if I where to come up with a smarter one you could get hosed.
In the early days of chess playing apps on homecomputers, you'ld rather frequently see superior software on inferior hardware winning the compu-chess tournaments.
Here is the top 5 of the dutch CSVN championships 1986:
Slashdot Mirror
There are several ways of communication. What about a virus/worm that just tries to infect pograms/systems, but when it succesfuly infects a program/system, it searches for other copies of itself within that program/system, and exchanges data with them. Next, new copies of the virus/worm are sent out with the new data.
If we use a genetic algorithm, the data would be the "genome", and we would have viruses/worms not only changing each generation, but actually evolving into more effective ones (from the viewpoint of the virus/worm), spreading information by "mating".
-><-
Grand Reverence Zan Zu, AB, DD, KSC
Another point is that discussions like this obscure the fact that ECHELON eavesdrops on everyone -yes, this means you too- it makes no difference wether you live in France or the US. Sure the first nations to complain about it are the ones that have some issues with the US, but this does not mean ECHELON all of a sudden becomes a good thing because it can be used to spy on them.
Then there is the paranoid "if they wheren't hiding anything, they wouldn't be whining" argument. It has been discussed over and over in endless crypto threads. This isn't any different.
French foreign policy might not be the best in the world, but the US keeps "pissing off" (to put it in your terms) various nations all over the world too, with its endless trade wars and embargos. The only difference is the US being capable of and willing to take firm, even violent, action to enforce its policies. This raises the question wether it's better to be an 'irritating whiner' like France or a 'big bully' like the US.
-><-
Grand Reverence Zan Zu, AB, DD, KSC
Ah, oh, uhhhh? Could you please show me how to compile a training set for self-awareness?
-><-
Grand Reverence Zan Zu, AB, DD, KSC
Igor Aleksander postulates consciousness in NNs is based on awareness of World and of Self. As I see it the mindpixel project would train a NN in world-awareness, but what about the self-awareness? (Or don't you agree with Aleksander at all?)
-><-
Grand Reverence Zan Zu, AB, DD, KSC
That would be more difficult than it sounds, IMHO. Have a look at the data compression reference center's page on the differences between LZ78 and LZW.
You might end up coding another LZWx algorithm, and the whole family of algorithms (the method) is patented (free for non-commercial use tho).
-><-
Grand Reverence Zan Zu, AB, DD, KSC
So why do I have to use the 9 bit to specify its uncompressed, if I'm not using a compression algorithm? If the file would be not compressed, I would have to set *one* "not compressed" flag for the entire file, not for each octet in the file. Since this is *not* possible with GIF I can't avoid my data being piped trough the decompression algorithm. So I'm stuck with using the algorithm, even if I dont want to compress...
-><-
Grand Reverence Zan Zu, AB, DD, KSC
And how exactly is this "uncompressed"? You are still (ab)using LZW compression; only now you've created a bigger file...
-><-
Grand Reverence Zan Zu, AB, DD, KSC
But why? Even ignoring the $50 million, assuming dotTv is not located in Tuvalu, knowing .tv is the Tuvalu TLD, why in the world would they want their domain name to end in .tv?
Doesn't seem like a valid argument, as this goes both ways. Why should big buck be allowed to break the TLD conventions?
What would you say if microsoft where to buy .gov?
-><-
Grand Reverence Zan Zu, AB, DD, KSC
There is a solution to DDoS, but it has a huge problem: you have to do it on the network(s) from which the malicious packets are being sent. You can't do much about it on the receiving side, the packets are going to flood your lines to/from the Internet before they reach your firewall/router. DDoS is going to be around until every admin on the Internet plugs his network holes and until every user on the Internet secures his machines.
I bet you the ircops have sent a zillion emails to the admins of the offending networks, but what more can they do? Sue the hell out of them and disable large parts of the Internet in the process? Hmm... doesn't sound to good, lets try to make the general public aware of the situation. They tried to do just that by putting info on the situation on their webservers, in their /motd's and you have what. Guess what? It didn't work. So now they are going to do the one thing they can do to get the attention of the public.
Sure this is a big disappointment to all the "well behaved" irc users, but if it is the only way of making them aware of the situation, let them do it. (And all of a sudden, everyone is indeed aware of the situation, and all up in arms...)
-><-
Grand Reverence Zan Zu, AB, DD, KSC
I think the ircops are aiming more for the 1rc-kl4nz with huge bot/flood-nets. Your typical script kiddie doesn't do much harm on irc (he's dossing clients, not servers), its mostly the clans that do the damage (they dos the servers to generate netsplits).
For these clans, it will be a real pain to move their 300+ bots to an other irc net, just for one day. Without their botnets, they are not likely to do massive takeovers, especially on other irc networks as they will challenge clans over there (of unknown sk1llz level) to come and have a fight on their home turf.
-><-
Grand Reverence Zan Zu, AB, DD, KSC
So you have testdriven an IP stack that isn't even released yet? Or, if you have tested the 2.3.99-pre3 and the latest BSD IP stacks, could you post (a link to) your test results?
-><-
Grand Reverence Zan Zu, AB, DD, KSC
It does matter. When you have something that a law, however unjust, says you can't have, then it's no longer the "land of the free", it's the "land of the underground."
What underground? US laws don't have any jurisdiction in other parts of the world. Hey, guns are prohibited by law around here, so can I go on and sue all of you US citizens carrying illegal underground guns?
Serious question: What if I put up some sort of proxy service to CPhack outside of US territory; would US citizens be allowed to use it, or is that illegal too?
-><-
Grand Reverence Zan Zu, AB, DD, KSC
Darwin is the Apple's OS X "kernel" or better "core OS".
Here is a link to Apple (including some marketing crap).
-><-
Grand Reverence Zan Zu, AB, DD, KSC
Yes, he wants Darwin in the BSD family, not Mac on Intel. From his avogato diary:
Though Darwin's role in the world is as the guts of Mac OS X, it is shaping up to be a viable BSD platform on its own. It take advantage of some of the great work underway in the NetBSD, FreeBSD, and Linux communities, as well as several of the neat things we're working on at Apple.
-><-
Grand Reverence Zan Zu, AB, DD, KSC
If we can offer fully-functional ASP and IIS on Linux, we can start to clean up the last pockets of resistance.
And while whe're at it, why stop at full functionality? Why stop at reimplementing micros~1 applications and not use their tactics? Why not embrace & extend our GPLled versions and put in some nice new features?
If the extensions are any good, micros~1 will be forced to implement them...
-><-
Grand Reverence Zan Zu, AB, DD, KSC
Things aren't always black and white, which is why I think it's a good idea to be objective when posting news. That's all.
Me too :)
But seriously, do you expect unbiased news on a site that clearly choses sides in the OS holy wars? After all /. claims to be "news for nerds" and since most "nerds" seem to be fighting Microsoft, whats /. gotta do? Tell them how great MSFT is?
What bothers me most is not the holy wars (they've been going on for ages), but that everyone is taking themselves so damned serious. Hacker culture always was ha ha, only serious. We seem to have forgotten about the ha ha. Just look at the quality of the average troll on /. (there hardly are any good ones, they're mostly flamebait labled as "troll"; and if you by any chance find a little gem its mostly marked as "funny").
As the hacker community grows, so does the number of wannabees, who think its all about being a full-time genius and a zealot...
Hey, I read /. because its fun, not because its the fastest and most reliable source of news (no offence guys).
-><-
Grand Reverence Zan Zu, AB, DD, KSC
oops, overlooked the other post about this link. I guess I'm redundant now?
-><-
Grand Reverence Zan Zu, AB, DD, KSC
Why not just report the news and let people discuss them here instead of spoon feeding them anti Microsoft propaganda every chance you get?
Ok, discuss this. Its the everchanging list of Top 30 Windows 2000 Bugs. This one is my favorite:
-><-
Grand Reverence Zan Zu, AB, DD, KSC
Looking at the success of the open source movement, one might say full-time telecommuting works. Ok, it isn't exactly working for a company, but the problems with it a are very similar (CVS, project forking, other type of communication/management needed, clash of egos).
Some companies will not hire telecommuters because of these problems, other companies like the cost savings aspect of it (they don't need large office buildings).
-><-
Grand Reverence Zan Zu, AB, DD, KSC
Well, there are some nice trolls here, you just gotta look close and THINK about what you read... There is no such thing as an obvious troll.
I agree with you on goatse.cx being a big asshole; but is it a troll?
Lets call these guys "spam kiddies" or something and moderate them down. At the same time moderate up real trolls, just to show them how its done (if they are smart enough to figure out its a troll in the first place).
-><-
Grand Reverence Zan Zu, AB, DD, KSC
Ok, lets say our algorithmical solution isn't
int i >= 100000000; but
BigNum i >= CONSTANT_WITH_250_DIGITS.
You can wait your 30 years for hardware good enough, then run your code for 5 years or so, wondering just how fast you could do it in another 30 years. Meanwhile my code executes in under 5 minutes on a 386 16Mhz...
-><-
Grand Reverence Zan Zu, AB, DD, KSC
uhhhh that was half the answer :)
As to the hardware, I seem to remember something in neural networks theory that states that no matter how big you make your NN, it doesn't work right until you randomize its inputs. IOW we can make our hardware bigger, we can make it faster, but until the stuff we run on it will be a whole lot more complex than it is today, we'll never reach artificial human-like intelligence.
As other posters have pointed out, human-like AI can't be done with brute force (alone). You are suggesting we can do it with todays brute force methods if we do it fast enough and have enough memory to store state information.
<rant>If we make a monkey 1000 times bigger and 1000 times faster it will still be a monkey. It might run us into the ground before we see it, but this won't be because its any smarter now.<rant>
If we put out monkey (lets call it Willy) behind a typewriter it might produce maybe one interresting piece of literature in its lifetime, but is Willy intelligent just because he punches keys at the speed of light and got lucky once?
-><-
Grand Reverence Zan Zu, AB, DD, KSC
Ok, but what about:
I'm talking about a different approach to the problem, a different algorithm in the sense that it changes the time-complexity of the problem. (We all know the classical example of naive vs. smart sorting algorithms.)
In our chess example I would build in heuristics and pattern matching. Actually, the winner app in my previous post was nearly all heuristics, which made it very strong against computer players. Its main tactics was to complicate positions as quickly as possible and make unexpected (non-optimal) moves. Ironicly, this is exactly how most human players try to beat todays heavy metal...
-><-
Grand Reverence Zan Zu, AB, DD, KSC
Yeah? Let's see you create a (arbitrary type of game here) chess program that will run on a 386 that wouldn't get pounded by Deep Blue.
If I am to use the same algorithm you're right, but if I where to come up with a smarter one you could get hosed.
In the early days of chess playing apps on homecomputers, you'ld rather frequently see superior software on inferior hardware winning the compu-chess tournaments.
Here is the top 5 of the dutch CSVN championships 1986:
Name: Designers: CPU: Points:
1 Nona Morsch 6502 6,5 - 7
2 Rebel Schroeder, Louwman pseudo 6502 6 - 7
3 Chess 0.5X Elsenaar VAX 11/785 5,5 - 7
4 Shess van Bergen VAX 8600 4 - 7
5 Sylkar Hartmann, Kouwenhoven VAX 11/785 4 - 7
<rant>Note that the scores of different applications on the same hardware are not equal!</rant>
(I wish /. had a <TABLE>)
-><-
Grand Reverence Zan Zu, AB, DD, KSC