pixcavator

>"economists asserting the difference between modeling human behavior vs. the behavior of physical objects"

Is it really that different? Try to attach an object to several springs and see if it behaves "rationally".

And Amazon.

>>Half a mil to learn about Poetry Writing. Sorry, but that's just silly.

Except it's payed for by your taxes. Who is silly here?

>>This is largely a topological problem: If you're surrounded by mountains...

Topological or topographical? It's kinda both in your example.

They can't give course credit or award degrees.

There was a relevant quote posted recently by "tokenadult": “the diligence of public teachers is more or less corrupted by the circumstances which render them more or less independent of their success and reputation in their particular professions.. . . . The privileges of graduation, besides, are in many countries . . . obtained only by attending the lectures of the public teachers. . . . The endowment of schools and colleges have, in this manner, not only corrupted the diligence of public teachers, but have rendered it almost impossible to have any good private ones." -- Adam Smith, The Wealth of Nations, Book V, Part 3, Article II (1776)

But it's OK for doctors, lawyers and their clients to do that?

I find this kind of reports very encouraging. I think the future of education is professors becoming free agents in a free market. Really, why can a lawyer set up his own firm or a doctor start his own practice and deal with life and death situations but a professor can’t just teach a few calculus courses without being affiliated with a university?

>>My cofounder, Ryan, has spent the past couple of years helping developers at Columbia learn how to apply their theoretical computer science skills to actually programming. He’s taught hundreds of Columbia students the basics (and beyond).

Take notes, Mr Khan.

>>if it was funded by someone else (e.g. the public) then my view is that everyone who contributed to funding that research should have the ability to access and use the results.

If you show those results to "the public", they'll likely want their money back. I know I do.

Non profit != no charge

Why would it be more impressive? Because paying customers vote with their money. Money is a better validation than "you're great!".

I'd be much more impressed it these (or at least a part of them) were paying customers.

I'd take a good book over this any day.

>>a list of people's heights isn't going to follow the distribution.

Try centimeters.

Here's a relevant image: http://inperc.com/wiki/index.php?title=A_common_view_of_digi....

It's actualy the latter, followed by the former -- on paper.

Here’s a “Russian” problem: find the largest projection on the plane of a box with sides a, b, c.

"European"? Are you ashamed of your nationality?

Anything that requires extra work from students is no-go.

Two pieces are rotated, that doesn't make the whole thing symmetric.

I understand that they are "equivalent" in a certain sense, but "symmetric"?

Did it ever occur to you that maybe you were able to handle PCA, CG, portfolio theory, etc because "Hefferon was my textbook for 2 full semesters"?

Jumping through the hoops for a chance of $20,000 doesn’t sound enticing. Certainly, the amount won’t be enough incentive to spend several years to write a new book. They might get some really old books or more recent ones that have gone nowhere (otherwise they wouldn’t be available).

What a strange use of the word "symmetric": "Two grid paths are symmetric if they share the same start and end point and one can be derived from the other by swapping the order of the constituent vectors."

>>It's why I think companies like Google (disclaimer: I work for Google) should stay out the content business.

Isn't AdSense content business?

BTW, is "officially" the new "literally"?

>>Randomization means that the last percent is just not worth chasing...

True, but only in the case of complete randomization. In the case of randomization kicking in only when the election is close, one side will be chasing the last percent to get to randomization and the other to avoid it. Same thing...

Thanks for the explanation! My only concern now is the validity of the meaning of the principal eigenvalue: "λ determines how much influence people share with each other through their connections. If λ is small then the CEO has a lot of influence, if it is large then he has little." It seems to depend on λ>1 or λ<1. Also, has this method been applied in practice, if you know?

Measuring influence via this eigenvalue method seems to make sense. However, there is ambiguity. First, you have to decide which eigenvalue to pick. Do you choose based on its magnitude? Even when you choose one somehow, what if the dimension of the eigenspace isn’t 1? Unfortunately, there are no links to clarify this.

The article was oulined here: http://news.ycombinator.com/item?id=2902684.