Every OECD annual report disproves that the gov-sponsored or private-sponsored have any performance differences. In fact both produce on average equally qualified students all over the world -- no difference. So if you want to choose between gov or non-gov sponsored model, performance is not the main issue, both model is equal in that metric.
Also, this excluded middle-ground fallacy makes me mad. There are millions of hybrid systems that could work very well, still people only argue about the extremes. I would opt for a gov-sponsored system at start that gradually falls back to a student sponsored one in case of bad personal performance. Loan is a bad idea -- let people drop early, quickly and cleanly.
No, undecidable would mean, that no PROVABLY P-time algorithm PROVABLY solving a given NP-complete algorithm exists. A subtle, but important difference.
I could imagine an algorithm that provably solves an NP complete problem, but it is impossible to prove that it is in P. You cannot even decide by experimentation because being in P means that there is a k and n0 that c*n^k upper bounds the execution time, for all n >= n0. Now in practice, without theoretical guidance, it is impossible to derive n0 and c and k.
In fact this could happen if P=?NP turns out to be undecidable. All of the undecidable problems could be made decidable by adding a new axiom to the system (but of course, other undecidable problems will rise).
It does not really matter. You can still execute it and do a lot of useful analysis. Simply sweeping over the input space and plotting the output will help. If the output is really crazy (chaotic, whatever), then you can conclude that the asset is too complex and risky to buy.
Well, GSM codecs will wreak havoc on these modems, as GSM voice channels are not simply bandlimited like POTS, but they use an artificial acoustic model of human voice tract.
Slashdot Mirror
Every OECD annual report disproves that the gov-sponsored or private-sponsored have any performance differences. In fact both produce on average equally qualified students all over the world -- no difference. So if you want to choose between gov or non-gov sponsored model, performance is not the main issue, both model is equal in that metric.
Also, this excluded middle-ground fallacy makes me mad. There are millions of hybrid systems that could work very well, still people only argue about the extremes. I would opt for a gov-sponsored system at start that gradually falls back to a student sponsored one in case of bad personal performance. Loan is a bad idea -- let people drop early, quickly and cleanly.
I am a PhD and I 100% agree with you.
(In fact I went for PhD only because I want to teach at my University)
Well, this is all in the eye of the beholder:
- Gay couples pay extra tax
vs.
- Non-gay couples get tax benefit
Or
- Non-Gay employees are negatively discriminated
vs.
- Gay employees are compensated
The iPads are pretty satisfied.
Brilliant! All we need now is Inkscape or GIMP. And a lot of malevolence :)
iPhone noTouch
Er, a small correction:
No, undecidable would mean, that no PROVABLY P-time algorithm PROVABLY solving a given NP-complete algorithm exists. A subtle, but important difference.
I could imagine an algorithm that provably solves an NP complete problem, but it is impossible to prove that it is in P. You cannot even decide by experimentation because being in P means that there is a k and n0 that c*n^k upper bounds the execution time, for all n >= n0. Now in practice, without theoretical guidance, it is impossible to derive n0 and c and k.
"undecidable would mean no P-time algorithm for solving a given NP-complete algorithm exists"
No, undecidable would mean, that no P-time algorithm PROVABLY solving a given NP-complete algorithm exists. A subtle, but important difference.
In fact this could happen if P=?NP turns out to be undecidable. All of the undecidable problems could be made decidable by adding a new axiom to the system (but of course, other undecidable problems will rise).
I hope it is, Rule 30 would be a nasty surprise.
Then just append the following string:
"? Undecidable"
Too cliched for my taste. The problem is that some of the really interesting theorems are hard to summarize in a simple formula. Like http://en.wikipedia.org/wiki/Four_color_theorem.
Bayes Theorem would be also top on my list.
+1 Information Entropy
That's why you never should tattoo physics laws, just mathematical theorems -- they change rarely.
It is better to tattoo P=NP, because you still can modify it later, if the opposite gets proven (just strike trough the equality symbol).
The definition of information Entropy would be my choice.
A more appropriate name instead of Supreme Court would be Oracle of Delphoi.
That's why I teach ;)
It does not really matter. You can still execute it and do a lot of useful analysis. Simply sweeping over the input space and plotting the output will help. If the output is really crazy (chaotic, whatever), then you can conclude that the asset is too complex and risky to buy.
How is this different from automatic updates? Is it initiated by the phone (pull), or by a remote entity (push)? Is it usable by 3rd parties?
Well, GSM codecs will wreak havoc on these modems, as GSM voice channels are not simply bandlimited like POTS, but they use an artificial acoustic model of human voice tract.
Also, they may invite your paper presented on a conference to be included in a journal. This is quite common.
Add Hungary as well
I would be mobsters on mobsters action.