Frankly, I agree wholeheartedly that the university should have publicly dissociated themselves from him, censured him, held a big press conference where they denounced him as a big fat jerk, taken away his library card, egged his house, given his email address to the Spammer's Union Local #97, sent him dead flowers, whatever.
But a PhD is a certificate of an accomplishment. It doesn't make sense to say it's rescinded unless it turns out that his actual dissertation had falsified data.
Frankly, I feel somewhat sorry for him. Reputation is everything in academia. We have a status system that makes Hollywood look like a socialist commune by comparison. I am just starting my own scientific career, and I can tell you, when you are considered "hot talent", you are treated quite well, but the pressure to produce is tremendous. Everyone is expecting great things of you, and when the promising lead you were chasing doesn't pan out, it can feel like *you* have failed.
Indeed I am. Perhaps you could enlighten me on what it is you do? Then I'd have a better idea where you're coming from, here.
Rather than basing our development model on mathematical rigor, we should instead look to the most complex thinking system in the world: human beings. We go blissfully through life dealing with contradictions, incorrect and incomplete information with ease. Our software designs should emulate this behavior, and be as fail safe and robust. What do humans depend upon most often? Heuristics.
Ironically enough, what you have just described is the field of AI, which is all about how we approach problems to complex to solve computationally/mathematically.
But, as you say, most commercial programmers do not solve problems like that. They solve problems which fall squarely into the realm of the mathematically tractible, and that's what the stuff you seem not to like (algorithm design, complexity analysis) is all about.
My point is we have a tendency, due to the culture of the Mathematics department, to put more emphasis on proving our algorithms are correct - rather than getting any practical tools in user's hands.
Now I must confess I am puzzled. What exactly are you advocating?
Should we change the way we teach computer science? Should we teach SQL instead of O(x)-notation, HTML instead of the pumping lemma?
Or are you talking about how programs get written?
What would be the benefit that we think we would derive from writing our programs more inefficiently?
Imagine if it he had a medical PhD and was working in a hospital - would you see this issue diferently then?
Not quite the same. What he would stripped of then would be his license to practice medicine, not his M.D.
I mean that You can't strip someone of knowledge
Indeed but if you are fabricating data you are proving that you didn't have that knowledge in the first place.
I disagree. The knowledge he received his PhD for the "knowledge" he fabricated are two different things.
Oh, don't get me wrong. His scientific reputation is, and should be, in the toilet permanently. He should never work in academia again. Period.
But trying to withdraw a PhD sends a misleading message about what a PhD means. It's a certification of having fulfilled certain requirements, not a grant of endorsement.
We cannot pretend to alter the past, and say someone did not accomplish what they did, even if we later decide we do not like him. It sets the troubling precedent that we may strip people of their academic credentials at will. That's a bad idea, even our reasons for doing so would be good ones in this particular case.
They should have stuck with an announcement censuring him.
I assume you are referring to the difficulties involved in factoring large numbers. While that problem (deciding composites) is a tough one, I don't think it has been proven to be NP-hard
No, I am not. I am refering to (G to the X mod P). Nobody bases encryption on prime factorization any more.
IIRC, it's a bit of an odd one by being polynomial-time reducible to its complement problem of deciding primes.
Actually, you don't recall correctly. Prime factorization is not polynomial-time reduceable to deciding primes...unless P=NP. This is because deciding primes is in P.
Try to reduce an NPC problem to primes, you'll see what I mean.
This whole thread is precisely why Computer Science should have never been allowed to fall into the Mathematics Department.
It wasn't "allowed to". Computer science came from mathematics in the first place.
How many practical computing problems have I run into in my carreer that have been NP complete? 0 - in 10+ years
This isn't very meaningful to us unless you tell us what it is you do.
In fact, NP-hard problems are not abtruse things that exist only in the minds of researchers. They are quite common. Most problems in the field of AI are at least NP-hard, if not PSPACE-hard.
99.999% of the computer science graduates will not have to deal with this issue - which is mainly concerned with or cutting edge theoretical issues (for example, how to do ray tracing in real time in a video game).
I'm afraid that's a very poorly chosen example. Rendering is in P.
In fact, issues of complexity theory are dealt with by basic tasks in a great many subfields of comp. sci. The reason most commercial programmers (folks who got their BSCS and got out) never see them is because they are dealt with in system calls and library routines which they call, but do not write, and seldom even read.
To give you an example, many programmers encrypt things. Many of them are not aware, while doing so, that the whole private key/public key system is based upon the characteristics of NP-hard problems. No NP, no asymetric encryption.
In many instances rigorous mathematics isn't needed at all, and fuzzy logic
The way you use this term is a bit unclear. Are you entirely sure you know what it actually means? I find it difficult to imagine that you solve most problems by coding Bayesian reasoning agents...
rules of thumb can be used effectively to get the job done.
I am afraid this is the very attitude that leads to the production of a lot of very, very bad code.
However, due to computer science being tied at the hip to mathematics, people are getting educations which don't mesh with the reality they see in the business world (where 85% - give or take - of the graduates will end up).
Now I'm going to have to be blunt and just say you're dead wrong.
You seem to be under the impression that academic research isn't "real", and that business is "reality". In actuality, computing in the business sector is a hothouse flower, which could not have sprouted, and would not continue to grow and thrive, without the tending and care given to it, for free, by research/theoretical/academic computer science.
The everyday benefits of this allegedly "abtruse" research surround you every day; they permeate your world. Without them, you have no operating system, no compiler, no VLSI-designed CPU, no heirarchical cache, no network computing, no internet, no Google, no...
But I digress. The point is that academic computer science has done its job almost too well, so well that the people who use its largesse every waking hour of their lives are largely unaware of whence these magical benefits came.
You are taking the stance of a man who kicks the tree whilst munching on an apple.
The definition of NP-Complete has nothing to do with whether the problem can be solved in polynomial time.
Not quite correct, because:
The set NP is defined as all problems whose solutions can be verified in polynomial time
...is one of *two* equivalent definitions of the class NP. The other is:
"Set of all problems which can be solved in polynomial time by a nondeterministic Turing Machine."
So the question "Does P equal NP?" is also the question "Does an NTM have the same computional power(in time) as a TM, or does it have more?" (It is already known that as far as decidiblity is concerned, TMs and NTM are equivalent.)
Because of the way the proofs have been constructed, if you can solve any of the NP-Complete problems in polynomial time, you can solve all of them in polynomial time.
Some more detail here:
An NP-hard problem is a problem to which any problem in NP can be reduced in polynomial time.
(Essentially, it can be used as a subroutine for any NP problem, with only a polynomial number of calls. Thus a solution to it is a solution to any problem in NP.)
An NP-complete problem is one that is:
1. NP-hard
2. In the set NP.
Thus if a polynomial-time solution exists to an NP-complete problem, then P=NP, because a polynomial number of calls to a function that terminates in polynomial time is O({polynomial}*{polynomial}) = O({polynomial}) .
Please note, however, that not all NP-hard problems are NP-complete.
How about a little common sense here? If you are a "bad guy" (and frankly, I'm not sure I trust the FBI's moral values enough not to assign this description to *them*), and you do not wish to get caught, you had better know enough about security not to rely on your OS's default install to protect you.
Most anything *can* be made secure, a fact that most people tend not to believe simply because it so seldom *is* made secure.
The brain just doesn't work that way. The "storing" of information in the brain is an emergent property that comes from a vast array of correlated neuron activations.
Think of a massive Hebb-rule neural net.
So how do we "forget" things?
Well, as in a neural net, old patterns, if not reinforced, are slowly lost as the "writing" of new patterns degrades them. But every time we think about something, it reinforces that pattern.
So how can these patterns be directly and quickly erased?
They can't. You'd have to be able to identify all the minute changes in neuron co-activation that represented *this* particular memeory and no other, and directly (physically) reverse them.
This would require a technology so advanced that it would be indistinguishable from magic.
To the politically correct, porn is not sexual entertainment but rather a political manifesto. A manifesto arguing in favor of hedonism; a demonstration of how enjoyable lack of inhibition can be.
Well said.
Porn is not, of course, a manifesto of any kind. What it actually is is a simple wish-fulfillment fantasy, much in the same way that trashy romance novels are. Railing against porn because it depicts women from a standpoint of male needs and priorities is about the same as objecting to romance novels because the men in them tend to behave in a way that appeals to their entirely female audience.
But of course, for modern feminism to admit that would be tantamount to admitting that male desires can possibly be legitimate, important, and worthy of consideration.
And that's not what feminism is about. Like any political cause that has outlived the achivement of its major aims, feminism seeks to locate or fabricate "injustices" to correct.
Since porn caters primary to male tastes, it's a primary target. Never mind that some women might enjoy porn, or enjoy being straightforwardly and enthusiastically sexual, because by the classic feminist circular argument, they must ipso facto be brainwashed or excessively influenced and their feelings therefore don't count.
Since the "real" needs as desires of women are simply defined by a bunch of activists, and anything that runs counter to those is decried as "unfair" or "exploitive", conflict can be maximized and any inconvenient compromise or compatibility can be avoided.
Slashdot Mirror
You're too kind.
Frankly, I agree wholeheartedly that the university should have publicly dissociated themselves from him, censured him, held a big press conference where they denounced him as a big fat jerk, taken away his library card, egged his house, given his email address to the Spammer's Union Local #97, sent him dead flowers, whatever.
But a PhD is a certificate of an accomplishment. It doesn't make sense to say it's rescinded unless it turns out that his actual dissertation had falsified data.
Frankly, I feel somewhat sorry for him. Reputation is everything in academia. We have a status system that makes Hollywood look like a socialist commune by comparison. I am just starting my own scientific career, and I can tell you, when you are considered "hot talent", you are treated quite well, but the pressure to produce is tremendous. Everyone is expecting great things of you, and when the promising lead you were chasing doesn't pan out, it can feel like *you* have failed.
You guys must all be in research or universities.
Indeed I am. Perhaps you could enlighten me on what it is you do? Then I'd have a better idea where you're coming from, here.
Rather than basing our development model on mathematical rigor, we should instead look to the most complex thinking system in the world: human beings. We go blissfully through life dealing with contradictions, incorrect and incomplete information with ease. Our software designs should emulate this behavior, and be as fail safe and robust. What do humans depend upon most often? Heuristics.
Ironically enough, what you have just described is the field of AI, which is all about how we approach problems to complex to solve computationally/mathematically.
But, as you say, most commercial programmers do not solve problems like that. They solve problems which fall squarely into the realm of the mathematically tractible, and that's what the stuff you seem not to like (algorithm design, complexity analysis) is all about.
My point is we have a tendency, due to the culture of the Mathematics department, to put more emphasis on proving our algorithms are correct - rather than getting any practical tools in user's hands.
Now I must confess I am puzzled. What exactly are you advocating?
Should we change the way we teach computer science? Should we teach SQL instead of O(x)-notation, HTML instead of the pumping lemma?
Or are you talking about how programs get written?
What would be the benefit that we think we would derive from writing our programs more inefficiently?
Imagine if it he had a medical PhD and was working in a hospital - would you see this issue diferently then?
Not quite the same. What he would stripped of then would be his license to practice medicine, not his M.D.
I mean that You can't strip someone of knowledge
Indeed but if you are fabricating data you are proving that you didn't have that knowledge in the first place.
I disagree. The knowledge he received his PhD for the "knowledge" he fabricated are two different things.
Oh, don't get me wrong. His scientific reputation is, and should be, in the toilet permanently. He should never work in academia again. Period.
But trying to withdraw a PhD sends a misleading message about what a PhD means. It's a certification of having fulfilled certain requirements, not a grant of endorsement.
We cannot pretend to alter the past, and say someone did not accomplish what they did, even if we later decide we do not like him. It sets the troubling precedent that we may strip people of their academic credentials at will. That's a bad idea, even our reasons for doing so would be good ones in this particular case.
They should have stuck with an announcement censuring him.
I assume you are referring to the difficulties involved in factoring large numbers. While that problem (deciding composites) is a tough one, I don't think it has been proven to be NP-hard
No, I am not. I am refering to (G to the X mod P). Nobody bases encryption on prime factorization any more.
IIRC, it's a bit of an odd one by being polynomial-time reducible to its complement problem of deciding primes.
Actually, you don't recall correctly. Prime factorization is not polynomial-time reduceable to deciding primes...unless P=NP. This is because deciding primes is in P.
Try to reduce an NPC problem to primes, you'll see what I mean.
Been there. If we could do that, then P=NP=coNP.
This whole thread is precisely why Computer Science should have never been allowed to fall into the Mathematics Department.
It wasn't "allowed to". Computer science came from mathematics in the first place.
How many practical computing problems have I run into in my carreer that have been NP complete? 0 - in 10+ years
This isn't very meaningful to us unless you tell us what it is you do.
In fact, NP-hard problems are not abtruse things that exist only in the minds of researchers. They are quite common. Most problems in the field of AI are at least NP-hard, if not PSPACE-hard.
99.999% of the computer science graduates will not have to deal with this issue - which is mainly concerned with or cutting edge theoretical issues (for example, how to do ray tracing in real time in a video game).
I'm afraid that's a very poorly chosen example. Rendering is in P.
In fact, issues of complexity theory are dealt with by basic tasks in a great many subfields of comp. sci. The reason most commercial programmers (folks who got their BSCS and got out) never see them is because they are dealt with in system calls and library routines which they call, but do not write, and seldom even read.
To give you an example, many programmers encrypt things. Many of them are not aware, while doing so, that the whole private key/public key system is based upon the characteristics of NP-hard problems. No NP, no asymetric encryption.
In many instances rigorous mathematics isn't needed at all, and fuzzy logic
The way you use this term is a bit unclear. Are you entirely sure you know what it actually means? I find it difficult to imagine that you solve most problems by coding Bayesian reasoning agents...
rules of thumb can be used effectively to get the job done.
I am afraid this is the very attitude that leads to the production of a lot of very, very bad code.
However, due to computer science being tied at the hip to mathematics, people are getting educations which don't mesh with the reality they see in the business world (where 85% - give or take - of the graduates will end up).
Now I'm going to have to be blunt and just say you're dead wrong.
You seem to be under the impression that academic research isn't "real", and that business is "reality". In actuality, computing in the business sector is a hothouse flower, which could not have sprouted, and would not continue to grow and thrive, without the tending and care given to it, for free, by research/theoretical/academic computer science.
The everyday benefits of this allegedly "abtruse" research surround you every day; they permeate your world. Without them, you have no operating system, no compiler, no VLSI-designed CPU, no heirarchical cache, no network computing, no internet, no Google, no...
But I digress. The point is that academic computer science has done its job almost too well, so well that the people who use its largesse every waking hour of their lives are largely unaware of whence these magical benefits came.
You are taking the stance of a man who kicks the tree whilst munching on an apple.
The definition of NP-Complete has nothing to do with whether the problem can be solved in polynomial time.
...is one of *two* equivalent definitions of the class NP. The other is:
Not quite correct, because:
The set NP is defined as all problems whose solutions can be verified in polynomial time
"Set of all problems which can be solved in polynomial time by a nondeterministic Turing Machine."
So the question "Does P equal NP?" is also the question "Does an NTM have the same computional power(in time) as a TM, or does it have more?" (It is already known that as far as decidiblity is concerned, TMs and NTM are equivalent.)
Because of the way the proofs have been constructed, if you can solve any of the NP-Complete problems in polynomial time, you can solve all of them in polynomial time.
Some more detail here:
An NP-hard problem is a problem to which any problem in NP can be reduced in polynomial time.
(Essentially, it can be used as a subroutine for any NP problem, with only a polynomial number of calls. Thus a solution to it is a solution to any problem in NP.)
An NP-complete problem is one that is:
1. NP-hard
2. In the set NP.
Thus if a polynomial-time solution exists to an NP-complete problem, then P=NP, because a polynomial number of calls to a function that terminates in polynomial time is O({polynomial}*{polynomial}) = O({polynomial}) .
Please note, however, that not all NP-hard problems are NP-complete.
How about a little common sense here? If you are a "bad guy" (and frankly, I'm not sure I trust the FBI's moral values enough not to assign this description to *them*), and you do not wish to get caught, you had better know enough about security not to rely on your OS's default install to protect you.
Most anything *can* be made secure, a fact that most people tend not to believe simply because it so seldom *is* made secure.
Yes, it's been said before, but it needs saying again.
Yes, anything windows does, Linux *can* do. But some of those things don't work nearly as well, and require a lot more effort.
Yes, the underlying structure of Linux is far superior to windows, but the user interface still needs much, much work.
Yes, Linux (or some OSS OS) will eventually replace windows althogether, but not for a while yet.
In the meantime, let's not be sophomoric about our OS preferences, yes?
Infinite minus one.
The brain just doesn't work that way. The "storing" of information in the brain is an emergent property that comes from a vast array of correlated neuron activations.
Think of a massive Hebb-rule neural net.
So how do we "forget" things?
Well, as in a neural net, old patterns, if not reinforced, are slowly lost as the "writing" of new patterns degrades them. But every time we think about something, it reinforces that pattern.
So how can these patterns be directly and quickly erased?
They can't. You'd have to be able to identify all the minute changes in neuron co-activation that represented *this* particular memeory and no other, and directly (physically) reverse them.
This would require a technology so advanced that it would be indistinguishable from magic.
To the politically correct, porn is not sexual entertainment but rather a political manifesto. A manifesto arguing in favor of hedonism; a demonstration of how enjoyable lack of inhibition can be.
Well said.
Porn is not, of course, a manifesto of any kind. What it actually is is a simple wish-fulfillment fantasy, much in the same way that trashy romance novels are. Railing against porn because it depicts women from a standpoint of male needs and priorities is about the same as objecting to romance novels because the men in them tend to behave in a way that appeals to their entirely female audience.
But of course, for modern feminism to admit that would be tantamount to admitting that male desires can possibly be legitimate, important, and worthy of consideration.
And that's not what feminism is about. Like any political cause that has outlived the achivement of its major aims, feminism seeks to locate or fabricate "injustices" to correct.
Since porn caters primary to male tastes, it's a primary target. Never mind that some women might enjoy porn, or enjoy being straightforwardly and enthusiastically sexual, because by the classic feminist circular argument, they must ipso facto be brainwashed or excessively influenced and their feelings therefore don't count.
Since the "real" needs as desires of women are simply defined by a bunch of activists, and anything that runs counter to those is decried as "unfair" or "exploitive", conflict can be maximized and any inconvenient compromise or compatibility can be avoided.