That piece is kind of crap. The main reason is that the summer holidays are over. The kids are in school (and busy with clubs, homework and so on on the weekends) and the parents are working. And as most bathers are gone, so are the drink vendors, the equipment renters and so on.You'll still find people on beaches, just not many.

I would imagine that this will be the kind of a movie that will be in the five dollar bin at Walmart's within a month of release. And there will be a lot of them that simply stay in that bin.

At least they'll be neatly packed.

"Indeed, in Japan, only the western half of Honshu, Shikoku and Kyushu have enough sunny days to justify large-scale rooftop solar installations."

Which is why one of the largest solar plants opened near Sendai in northern Honshu a couple years back? What conditions are profitable depend on the technology you use, and the cost of production. And as solar cost decreases and efficiency increases more locations will be realistic.

There was some seismic activity for a period of time before it happened - but this is Japan; there's always seismic activity happening. According to interviewed researchers there really was no warning beforehand.

The video is deceiving; that trail is much steeper than it looks. Slowly stumbling downwards is pretty much all they could do. Also, most deaths from eruptions are either from poisonous gas or from heat. A small hut will shield you from neither. But both gas concentration and heat will disspiate by distance, so simply trying to get away from it may well be your best chance to survive.

So why are people generally using quadcopters for autonomous systems? What's the disadvantage of a single-rotor copter when you're doing autonomous flight? I can imagine that perhaps it's a size issue - quadcopters are lighter or cheaper or more efficient below a certain size or when indoors? Or is it much more difficult to write a reliable control system for a single-rotor system?

Could they handle that weight for such a distance, though? And in very rough weather, including rain and strong headwinds?

Here's some data on the hardware, from http://ca.reuters.com/article/...

* 65 km/h peak speed, and will cover the distance in about 15-30 minutes;
* It weighs 5kg, and can carry a payload of up to 1.2kg

With 1.2kg it can certainly carry a complement of medicines or even small, urgently needed hardware and parts (batteries or spare bits for medical equipment for instance). Not general use of cours, but it does look like more than just a stunt.

Why? Maybe I simply only review things I like. Why would that devalue my reviews?

This. One reason really low and really high reviews are much more common than they ought to be is that people only bother voicing an opinion if they feel strongly (positive or negative) about something. Another is that once they do, they'll tend to exaggerate their evaluation to really drive home how they feel.

My suspicion is that the only stable scale is a simple "really liked it/really disliked it" up/down system. Then somehow weigh that according to the proportion of customers or buyers that actually bother to review. That depends of having a decently good estimate of that proportion though. The likes of Amazon have that for their products; for restaurants it'd be hard to impossible,

I feel sorry for Rollo. He seems to get all the blame ever since he stated working for that website project.

Charlie Stross recently posted a very good take on this: This is a permanent change. Whatever happens during the first few years is basically irrelevant, compared to the long-term results. Did Norway separating from Sweden cause short-term economic upheaval? Does that matter at all a century later?

This is a long-term change, not a short.term one. Any voter should consider the probable situation twenty or fourty years from now, not whatever happens in a year or two.

"Also some of the science and tech courses are very demanding but the teachers don't simplify it leading to many whooshing sounds for the student throughout the courses. Such courses could benefit from a simplified overview of the course material."

How many employers would like to hire people that can't understand the actual content and need "simplified overview" to get a grade? If you really don't grasp it to the point where you can actually apply the math for new, novel problems, then you don't actually know it, do you?

MOOCs have a serious credibility problem already. The very last thing they need is to dumb things down. If it becomes common knowledge that, say, an engineering MOOC graduate can't even handle a system of differential equations in an intelligent manner, or don't understand the implication of Green's function, then the credits will become truly worthless.

It does depend on the size of the field as well, though, as well as the funding. I can well imagine astronomy having major problems; everybody has heard of astronomy, and lots of people dream of being astronomers.

A friend of mine is working in paleogeology. As you might imagine there's not a huge amount of money in the field. On the other hand, few people have heard of it either, and there aren't that many people dreaming of working there. There's no movies starring daring paleogeogists with hat and bullwhip in hand ducking poison arrows and swinging across pits of snakes in order to determine the local sea bed temperature during the cambrian. The end result is that funding is pretty stable and dependable. People that are qualified and willing find funding. I bet there's a fair amount of other obscure fields in a similar situation.

You could write an open source application in C++ rather than the much less mainstream R language and you'd have lots of people ready skilled to maintain it.

You may be right in general. But R is not a general-purpose language. It's a programmable tool for statistical computing; you'd have to spend a lot of time to reimplement a set of high-quality statistical libraries to do the same. Doing that correctly is very non-trivial and not quick. Very similar to saying you can replace Matlab with your own C++ code.
 

Wow, that's a lot. I wonder why they allow so much power per outlet?

Because of all those damn industrial vaccuum cleaners.