Your analogy is good for most hosting sites but doesn't work for Megaupload. I hate to cite reddit as a source but this guy did a pretty good summary of why MU was shut down. I think the others are quite safe for now.

For what it's worth, we also pay a tax on memory on almost anything that can store a few mp3's in France but we still get screwed if we get caught downloading.

You missed my point. I agree that there needs to be some sort of class to offer to high level concepts like the functioning of a network of computers, basic OS concepts and intro to some basic programming, but the notion that we should skip on "bits & bytes" just because it is bound to change is ridiculous.

Same applies to maths. Mathematical theories are abstract constructs that can (and sometimes are) replaced with more efficient ones, but that doesn't mean we should stop teaching the old ones, because knowing where we come from help assessing the relevance of new theories. Just as possibility theory hasn't wiped out probabilities, ternary logic hasn't wiped out boolean logic (and ternary computer have existed since 1870). The underlying logical models do matter.

I tend to think that we haven't actually left those times. The only difference is that back then there was a valid explanation (mainly economy of space - writing tools and support were expensive) while now it is just an expression of people being 2 lazy 2 bothr wrting all te ltters & pncttion.

I am not debating whether it is a good or bad thing, I am just sick of people thinking this is somehow a "new thing" caused by [technological innovation] and that all people are more stupid compared to [insert era before major technological innovation]. It is not new. It is probably as old as the act of writing itself (and I say probably only because I am too lazy to search for a citation).

Using abbreviations to save time or express words phonetically is as old as the Romans and the ancient Greeks, it has nothing to do with twitter.

Have you even RTFA ? Nowhere does it say anything about atheism, it talks about whether scientists think there are conflicts or not.

Came here to post this - I think it was a poor choice a word. One could very well have written that "85% of scientists think that there are conflicts between religion and science" and it would have made more sense. Now what frightens me is the 15% that think that there is no conflict at all.

Eww - I would tend to agree wholeheartedly. In my research institute we have a few Agile nutjobs that try to evangelize their methodology - despite the fact that the academic settings mean that anything "Agile" makes no sense at all.

www.ai-class.org, by Sebastian Thrun and Peter Norvig

Textbooks are the base on which you build before you go on checking the current research on a topic - they are not only useful but also quite often essential, both as a reference and as an introductory text. What you just said doesn't make any sense: what if the team which happens to leads in my field is tens of thousands of km away ? What if I study multiple subject dominated by, say, one institute in Belgium and the other in the US ? What if, say, I want to study general architectural theories for common sense reasoning systems, do you think I can go bother Pr. Minsky ? Maybe I should even go see Don Knuth for my advanced algorithmics class. And as another commenter pointed out, you can be a great researcher and a sucky teacher. Or even a so great researcher that you don't necessarily have time to teach. Or... you could just find a good textbook. A textbook written by people who happen to be knowledgeable about the subject and not suck at explaining things (I'm not saying all textbooks are like that - a lot of them are terrible - but there are some pearls that are usually not hard to find to get into a subject quickly).

"This was repeated by Facebook. You will find endless experts claiming you could never scale either PHP or MySql to be a serious site, yet one of the largest sites in the world runs on those two." Except that Facebook: - compiles its PHP to C, so in the end it really has nothing to do with PHP anymore apart from the syntax they code in. So the "endless experts" were right: PHP is an unscalable turd. - runs like a piece of shit anyway. Numerous problems of disappearing messages, posts that can't be seen for hours after they were supposed to. The only reason it isn't a real problem is because nobody does real work with Facebook, it is used only for leisure so nobody gives a shit if a message appears right now or in a dozen of hours. So... Once again, the "endless experts" were right.

Not my field at all so this is a real question: wouldn't that percentage depend on the student ? It was my understanding that people respond differently to different ways of teaching (some learn more with visual content, some prefer audio stuff, some are better with books, some respond better to a combination of multiple ways), so wouldn't there be a certain "kind" of students whose favourite method of learning would be through an intelligent tutoring system ? I know this is purely anecdotal so only for illustration purposes and not as proof, but for instance in my case I know I always fear "looking dumb", even though I know I'm here to learn, so it stops me from asking questions that would make me look like a complete moron. When confronted to a computerized tutoring system (even a basic one such as the Open Learning Initiative courses) I don't have that kind of anxiety so I am not afraid to make mistakes and do the same exercises over and over until I get everything right, which in the end helps me learn more efficiently (or so I like to think).

Care to point out where it is said that it is a "breakthrough" in anything ?