Just wait. Next it will be VMWare's ESX server with every version of Windows ever released on the left, and a random sample of Linux distributions on the right.
If MS buys AOL, the slashdot readers who hate MS win, because I don't think that even MS could survive that acquisition (look what it did to Time Warner). What are they thinking? Wow, now we'll have MSN AND AOL... the two most popular ISPs on the planet!
If Google buys AOL, we can count on them FINALLY doing something evil, right? All of you "Google is evil because people want to work there" people will be able to watch Google tank from an acquisition.
Your usage isn't as incorrect as your primary school english teacher might have led you to believe though. It's perfectly correct (and the prior, incorrect in this case) to say, Amazon might sue/. or me.
I don't know much about quantum computing, but taking an exponential algorithm into poly time makes it theoretically tractible.
The difference isn't a factor of 10 or 20 speedup, it's going from taking a number of seconds equal to all of the protons in the universe, to something like, the number of nails in the bin at HQ.
Think in these terms. If modern crypto (not an expert on crypto) takes 2^n seconds to crack on a conventional computer, it'll take n seconds to crack on one of these.
What about when the code is downloaded to the local machine for execution?
So, I write a hacked version of NFS. I take allowed machine off the network (since I'm root). I clone its mac, I have its SSH keys (since I'm root), I set up with identical IPs and what not. My hacked version of NFS downloads the code and now I have it.
I don't see a way that they can do this aside from obfuscation, with modern technology. Even crypto requires the file to be decrypted to execute (something that I attended a talk on just last week).
Either way, I sincerely don't believe that there is anything to be learned from a PERL script that you can't even look at.
I was just picking a few common complexities, not making any theoretical statements. Still, the intial poster was asking if there's a reason to do this, like, a computational one. I'm still inclined to say "not really." Aren't you?
I have to think about this in terms of asymptotic complexity.
The difference between 14 and 28 days in these terms, is pure BS, nobody would care, right? If n translated to a process that took 14 days, the next we'd care about is nlog(n), then n^2!
nlog(n) 14... 21 n^2 14... 196
Research that I've seen in NP complete and PSPACE complete problems, of course, suggest solving schemes whose times vary wildly, but no assurance could be made of cracking within any reasonable time interval. It would be akin to "any time between 10 seconds and the end of time."
So, the only thing remaining is empirical evidence, not based on any theoretical result, that says "well, normally we can do this in x days."
Now, if that's the case, and it's some sort of distributed process (lets assume it is), well, scalability concerns aside, they could scale up the system, and protect personal libterties a bit more.
Pentium was just a way to highlight the difference between the new chip and the 486. Now that we're coming out with Pentium 5s, it makes sense to rename them before more educated folks catch the reference and go to the company that has Hexium 5's.
Eugenics is a program by which breeding is controlled. The Nazi eugenics program began like that, and then took on the killing people bit.
That's not to defend the Nazis or eugenics, it's just to clarify what it means.
Here's a bit of trivia. Eugenics is a classic socialist idea. Modern socialists usually leave that out of conversations, since the Nazis made it unpopular. Also, the Nazis were socialist. Nazi is short for German National Social Workers Party.
Again, not a dig on socialism, just a few observations regarding the roots of what you're discussing.
Interestingly, I met up with a chat buddy of mine from HS who saw me posting on/. Also, a bunch of Cornelians seem to post here. I've had a couple chit-chats with people who even posted other student's research papers here (perhaps a couple who posted their own, but said that someone else did;-))
Bayesian Inferrence refers to a rather large class of algorithms. It would be nice if something were more specific.
To give a heads up as to what this all is. Bayesian statistics are based on the idea that a probability can be updated based on additional information.
So, perhaps you have a prior of 0.5. There is a 50/50 chance that whoever you're looking at is better than another player.
Ok, so, 0.5 is the prior. Or, perhaps he's one 90% of the games played, so 0.9 is the prior. Now, 50% of games against the 2nd ranked person, he won, but only 20% against the third... and so on. That would be one form of bayesian inferrence.
Other forms? Naive Bayes is a type of, fairly simple machine learning algorithm. There are also graphical models, which are rather advanced bayesian machine learning models.
Slashdot Mirror
Just wait. Next it will be VMWare's ESX server with every version of Windows ever released on the left, and a random sample of Linux distributions on the right.
Agreed, and it totally rocks too.
I agree. Just look what it did for Time-Warner.
Wait...
I see it this way.
If MS buys AOL, the slashdot readers who hate MS win, because I don't think that even MS could survive that acquisition (look what it did to Time Warner). What are they thinking? Wow, now we'll have MSN AND AOL... the two most popular ISPs on the planet!
If Google buys AOL, we can count on them FINALLY doing something evil, right? All of you "Google is evil because people want to work there" people will be able to watch Google tank from an acquisition.
I guess that there's something for everyone here.
That's /. or I. Not me or /.
/. or me.
Your usage isn't as incorrect as your primary school english teacher might have led you to believe though. It's perfectly correct (and the prior, incorrect in this case) to say, Amazon might sue
I don't know much about quantum computing, but taking an exponential algorithm into poly time makes it theoretically tractible.
The difference isn't a factor of 10 or 20 speedup, it's going from taking a number of seconds equal to all of the protons in the universe, to something like, the number of nails in the bin at HQ.
Think in these terms. If modern crypto (not an expert on crypto) takes 2^n seconds to crack on a conventional computer, it'll take n seconds to crack on one of these.
What about when the code is downloaded to the local machine for execution?
So, I write a hacked version of NFS. I take allowed machine off the network (since I'm root). I clone its mac, I have its SSH keys (since I'm root), I set up with identical IPs and what not. My hacked version of NFS downloads the code and now I have it.
I don't see a way that they can do this aside from obfuscation, with modern technology. Even crypto requires the file to be decrypted to execute (something that I attended a talk on just last week).
Either way, I sincerely don't believe that there is anything to be learned from a PERL script that you can't even look at.
Not a problem.
Just a quick question... what are these students supposedly learning from this script anyway?
If it was a systems program, well it probably wouldn't have been written in perl, lets be honest.
AI? If the students can't learn about the algorithm, it's not much use.
Theory? What will you prove about a black box perl script?
I was just picking a few common complexities, not making any theoretical statements. Still, the intial poster was asking if there's a reason to do this, like, a computational one. I'm still inclined to say "not really." Aren't you?
Two thoughts.
The first, root can remount the filesystem.
The second, I've actually never seen ONLY execute. I've seen disallowing only execute though.
I have to think about this in terms of asymptotic complexity.
... 21 ... 196
The difference between 14 and 28 days in these terms, is pure BS, nobody would care, right? If n translated to a process that took 14 days, the next we'd care about is nlog(n), then n^2!
nlog(n) 14
n^2 14
Research that I've seen in NP complete and PSPACE complete problems, of course, suggest solving schemes whose times vary wildly, but no assurance could be made of cracking within any reasonable time interval. It would be akin to "any time between 10 seconds and the end of time."
So, the only thing remaining is empirical evidence, not based on any theoretical result, that says "well, normally we can do this in x days."
Now, if that's the case, and it's some sort of distributed process (lets assume it is), well, scalability concerns aside, they could scale up the system, and protect personal libterties a bit more.
There's this funny thing about root access...
Pentium was just a way to highlight the difference between the new chip and the 486. Now that we're coming out with Pentium 5s, it makes sense to rename them before more educated folks catch the reference and go to the company that has Hexium 5's.
Not at all.
Apparently I missed the fine point of who invented eugenics, but as another poster pointed out, conceptually I'm dead on.
For that matter, I find it in poor taste to even make jokes about eugenics, but did not want to offend the original poster.
No it's not.
Eugenics is a program by which breeding is controlled. The Nazi eugenics program began like that, and then took on the killing people bit.
That's not to defend the Nazis or eugenics, it's just to clarify what it means.
Here's a bit of trivia. Eugenics is a classic socialist idea. Modern socialists usually leave that out of conversations, since the Nazis made it unpopular. Also, the Nazis were socialist. Nazi is short for German National Social Workers Party.
Again, not a dig on socialism, just a few observations regarding the roots of what you're discussing.
I thought that the whole point of the exercise was to submit to control.
You really should try it. It's quite liberating to submit to anothers control.
Don't you feel better whenever you hit the 'Submit' button on Slashdot?
SUBMIT DAMN YOU! I TOLD YOU TO SUBMIT!!
An unmanned spacecraft really wouldn't impress Hollywood... better throw a crew on it.
Honestly, I think that that is how some missions are put together anyway.
K, so, I RTFA... TrueSkill actually rocks...
You are only hurting yourself you know
Actually, they're only hurting their students. They probably stand tons to gain from this.
Interestingly, I met up with a chat buddy of mine from HS who saw me posting on /. Also, a bunch of Cornelians seem to post here. I've had a couple chit-chats with people who even posted other student's research papers here (perhaps a couple who posted their own, but said that someone else did ;-))
That actually rocks.
Bayesian Inferrence refers to a rather large class of algorithms. It would be nice if something were more specific.
To give a heads up as to what this all is. Bayesian statistics are based on the idea that a probability can be updated based on additional information.
So, perhaps you have a prior of 0.5. There is a 50/50 chance that whoever you're looking at is better than another player.
Ok, so, 0.5 is the prior. Or, perhaps he's one 90% of the games played, so 0.9 is the prior. Now, 50% of games against the 2nd ranked person, he won, but only 20% against the third... and so on. That would be one form of bayesian inferrence.
Other forms? Naive Bayes is a type of, fairly simple machine learning algorithm. There are also graphical models, which are rather advanced bayesian machine learning models.
Don't worry. I've been using her email address when I sign up for spam supported junk for years!
(just kidding)
Very cool. Good luck with everything. Don't let school drive you up the wall!