On the other hand, I've known programmers who are great at graph theory but can't debug their way out of a paper bag.

And I've worked with a great programmer who had an excellent pure math background (ABD from PhD a program with heavy discrete math component) and someone comparably good with a high school diploma who was entirely self-taught. I wouldn't necessarily set them to solve the same class of problems, but their core skill-sets overlapped quite a lot, as did their attitude toward correctness, good design, etc.

Programming is still an area where a good autodidact can excel, and many academic courses are less than impressive. It's a subject we are still learning how to teach, and so far I've not seen anything to make me believe any particular academic background is either necessary or sufficient to inculcate the desired skills.

Does it taste good? If not, you haven't made a real animal.

There is nothing deep or even particularly interesting about these questions, and just how stupid their breathless idiocy is can be seen by asking, "Does the newly created entity lack almost every interesting property of the entity some philosophy-addled idiot thinks we should 'wonder' if it is absolutely identical to in every respect?" The answer is always, trivially, "No."

So only an extremely stupid person or a shill trying to market something (fake wisdom?) would ask such an idiotic question.

There are more reasonable questions that people who are neither idiots nor philosophers (but I repeat myself) are reasonably well-equipped to answer. Like this: lacking anything remotely resembling neurochemistry, is it appropriate for us to impute to this model any of the effects of neurochemistry that may or may not be lumped into the neuron behaviour? Since we're pretty sure neurochemistry is independent of network architecture, it would be incredibly stupid to identify the entirety of the robot's responses with the neuronal architecture, rather than the neurochemical environment they behave in?

For example, if you starve a worm its behaviour changes because its neurochemistry changes. Hormone levels, cortisol levels (or their worm equivalents) change, and that changes behaviour, in some cases quite dramatically. So what happens to the robot when you starve it? And if you can't starve it, why do you think it is in any way identical to a worm, rather than just an interesting simulation of part of it?

And of course, simply because we can imagine a more complete model of a worm doesn't mean we can build one that is sufficiently similar to a worm in all respects to make any of these questions interesting. It would have to eat and excrete and so on. It would have to have environmental sensitivities. And imagining those things aren't important is stupid: what we imagine is not relevant to what is real. There is no basis for saying a mechanical worm is a "real" worm (what would an "unreal" worm be?) It is a "real" mechanical worm. You still can't eat it, so it isn't a worm. Saying certain properties "don't count" is pure magical thinking, unworthy of scientists.

Because they are all done by men, whose lives are not valued by our society.

Stirling engines are used in space, but only when there is a compelling reason to do so. The basic argument against them is two words: moving parts.

Mechanical wear is a huge problem, and thermal management is not a small one. Depending on the spacecraft a sustainable thermal regime may have to be maintained across very different environmental conditions (full sunlight, deep shadow) and very different operational phases. Just getting lubrication to work properly under such circumstances, over a decade in the case of Rosetta/Philae, is non-trivial.

Simpler is better in space, so batteries and solar panels are always going to win over RTGs, and RTGs with thermoelectric conversion are always going to win over RTGs with Stirling engines unless there is a compelling need for the greater output power density each step up in complexity gets you.

Steam turbines are just a non-starter. Water is a dreadful substance to deal with. Highly reactive, prone to freezing, capable of going wrong in ways we can barely imagine, but would certainly discover if anyone was foolish enough to put a steam turbine on a spacecraft. It might be made to work one day, and there are probably some really clever things we are missing, but the cost of each mission is so high that extreme conservatism rules the day, and rightly so.

More likely multiple graphs, since "skills" are abstract categories, and abstract categories are made things that ever knowing subject creates for themselves, with only approximate overlap between them. So what I mean by "interpreted language" and what you mean by "interpreted language" are going to overlap substantially, but we will draw the edges of our attention differently. Some borderline cases you will call "interpreted" and I won't, and vice versa.

For example: are just-in-time compiled languages interpreted or not? There are perfectly legitimate ways of drawing the edges of our attention that include them and others--equally legitimate--that exclude them.

The problem with all such attempts as this is they naively and wrongly assume that the world, which is necessarily some particular way, must also be divided into uniquely determined, rather than usefully constrained, abstract categories by knowing subjects. This is simply not the case, and we have endless examples demonstrating it.

Classical and Newtonian mechanics, although mathematically equivalent, use different and incompatible abstract schemes (one says the principle of least action causes motion, the other Newton's laws.) And so on. 99% of the time we are dealing with interior cases where such distinctions don't matter, but then we hit some weird edge case (quantum mechanics in the case of classical vs Newtonian physics) where one of them suddenly looks completely different from the other, and naive people say, "OK then, the world is really this way" (where "this way" means, "is organized uniquely according to this abstract scheme"). But it isn't: abstractions are ways we describe the world to ourselves as knowing subjects. They are objective in the sense that they arise out of the causal relationships that knowing subjects have with the rest of reality, but to be an object requires a subject (and to be a subject requires an object: there is no view of no-where.) Different subjects will divide up the same objective reality in slightly different ways.

Yup, but innumerate idiots are only going to hear "radiation" and "Fukushima" and claim that the entire west coast is going to be dead. This is an actual claim an actual person made to me just a few weeks ago.

The same person claimed to be deeply concerned about climate change.

Only via complete and utter innumeracy is it possible to be deeply concerned about climate change (not an unreasonable position) and opposed to nuclear power, since nuclear power is the only proven-to-work, proven safe (in precisely the same sense that airliners, which crash sometimes, are "proven safe") alternative to base-load coal.

But they are different skills, and more powerful tools necessarily imply that what used to be highly skilled jobs are now not so skilled.

Automation isn't making programers dumber, it's allowing dumber people to be programmers, or dumber programmers to do harder things. It's been this way forever. 99% of programers working today couldn't have toggled firmware into a 6502 and made it run. Fortunately, they don't have to.

I'm old enough to have known programmers who still were kind of suspicious of these new-fangled "compilers", and I've actually programmed on punch cards, and collected data on a machine that booted from paper tape (there were no working tape punches left, so the lifetime of the machine was dependent on taking really good care of the few remaining tapes...)

All of that has gone away, and the skills programmers needed in those days have gone with it. That's a good thing, although it kind of sucks for people who put in thousands of hours honing skills that are now irrelevant.

Part of expertise is knowing when to ignore fads. This may require faking it well enough to get along for a while.

Cool-hunting has been around forever and is done by all kinds of people, not just hipsters. Were hipsters in at the start with glam rock? Disco? New Country?

Yet all those things were "cool" (for a certain value of "cool") once upon a time.

So hipsters are at best a subset of cool-hunters, and not a very interesting set, because they differ from other cool-hunters in their stupidity, insularity and arrogance. Many cool-hunters want to find the cool and share it with others. Hipsters want to find the cool and keep it to themselves, to the point of denying that anything that has become popular is cool any more.

Furthermore, you don't understand futures trading, even a little bit. Futures trading is about hedging, not discovery. They literally have nothing to do with each other. Futures markets are not predictive, they simply represent the mean of trader's expectations. They are an essentially homogenizing force. So if you think hipsters are like futures traders you are saying they are trying to make everyone the same bland and boring type.

Another clue that hipsters have nothing interesting to say is their proclivity for using unconventional typography--such as eschewing capitalization--to draw attention away from the vacuity and falsehood of so much of what they say.

Hipsterism is the practice of misdirection. Hipsters are lame people who have learned that attention is the scarcest human resource, so they can hide behind a few attention-grabbing quirks. It saves them from having to do anything actually interesting, useful or productive.

It's kind of sad, really, but the hate they get is well-deserved, because they are socially useless people who are deliberating soaking up our precious, limited attention on completely pointless self-aggrandizement.

All of which makes the article's breathless touting of this "innovation" pretty funny.

Two of the most basic moves in engineering are:

1) Take two functions that used to be separate and integrate them into a single component. This increases efficiency.

2) Take two functions that are performed by a single component and split them apart. This increases robustness.

Which move is a good idea at any time depends heavily on technology. Wing-warping (lift and control both done by the same component) was a poor fit for wood-and-fabric technology, so ailerons (lift and control done by separate components) was a good move. Metal frames and skins were not much different from wood and fabric in this regard, but now we are making aircraft mostly out of plastic (https://en.wikipedia.org/wiki/Boeing_787_Dreamliner) it may be time to reconsider the problem (which I guess has been done for some military aircraft already).

But it's not like this is a super-innovative work of genius. It's a pretty standard move that any good engineer is likely to consider when faced with a problem of efficiency (although exactly why integrated flaps are supposed to be such a huge improvement is not at all clear from TFA).

What makes horses simple and zebras complex?

Answer: nothing.

The issue is not simplicity, but prior probability. All else being equal, horses are more likely in most locales than zebras. When you hear hoofbeats, the plausibility of "There is a horse nearby" and "There is a zebra nearby" go up by the same factor. Since horses were already more likely, horses are still more likely.

Ockham's razor works, in the very few cases it does, as a consequence of Bayes' rule, and invoking some ill-defined notion of "simplicity" rather than prior plausibility is misleading.
 

Science is the discipline of publicly testing ideas by systematic observation, controlled experiment, and Bayesian inference.

Because science is at heart applied Bayesian reasoning it is not in the business of certainty of any kind: theories become more plausible or less plausible, and are never "true" or "false", which would imply that they are immune to any further evidence whatsoever. This state simply cannot be achieved within the Bayesian formalism.

The quest for certainty is science's equivalent of alchemy: alchemists weren't wrong because of their investigative techniques (which were often quite good) but because they were pursuing the wrong goal (transmutation of the elements). The proper goal of science is not certainty but knowledge, which is inherently uncertain. Philosophers don't understand this, and will no-doubt continue to promulgate the model of Poperian method for generations to come.

Massive generalization has been good to the sciences.

In physics, we have used invariance principles to expand our definitions to the most general possible. This was the argument behind General Relativity, for example: Einstein wanted equations of motion that would be invariant under any smooth second-order transformation whatsoever, and when he put that constraint, and only that constraint on, he found the most general form of the equations of motion were uniquely determined (up to a constant of integration, which is the Cosmological Constant).

Biologists have generally shied away from this kind of approach to their field, but there is an argument to be made that there is an equally general definition of "life" as "anything that participates in a process of evolution by imperfectly heritable traits that result in differential reproductive success". It need not be tied to any particular concrete mechanisms like DNA and RNA.

This idea may turn out to be silly (which is why I wrote a novel about it rather than a scientific paper: http://www.amazon.com/Darwins-...) but well into the 1800's there was no general view of "energy" that unified all the disparate forms, to the extent that the fact that light had energy that was in any way related to mechanical energy was not really appreciated. The kind of unified view of energy we have today would have seemed bizarrely speculative at that time, in the same way that the notion of a unified, generalized view of life is purely speculative today, but it turned out to be amazingly useful, so it's worth considering the possibility in biology today (anti-science people will likely attack it as a waste of money, using computer technology that would not exist without a similar "waste" on ridiculous fantasies like quantum theory and "obviously useless" research into the basic physics of semi-conductors in the past.)

There's a joke attributed to Abraham Lincoln: Q: If you call a tail a leg, how many legs does a dog have? A: Four. Calling a tail a leg does not make it one.

Nothing in ordinary parlance (headlines, say) means an absolute absence, which is an empty concept, like the Philosopher's Stone or an honest politician. You don't get to decide for the sake a headline that it means something else. Only grasping little scumbag shills do that, people who are so fundamentally and thoroughly debased that they turn everything they touch--even language--into garbage.

"Pretty much nothing" is not "nothing". "In terms of mass and energy" is not a relevant restriction. The quantum vacuum is rich in properties. There is absolutely no basis to ignore those properties for the sake of a dishonest, misleading and confusing headline. It's like people who say "Before Europeans arrived North America had no people in it!" Which is true, for a certain value of "people". This is why precision in language matters, because there are people who actually say things like that to the considerable detriment of their fellow-humans.

"From a sufficient distance" would mean "infinity" if you want to talk about asymptotically vanishing properties of the vacuum, which would still leave all the other properties, so again: it isn't clear why anyone would dishonestly and stupidly restrict the discussion to one particular set of properties unless they wanted to dishonestly and stupidly made a false, dishonest and stupid claim that "the universe came from nothing!"

So other than being dishonest, stupid and wrong, there is nothing at all dishonest, stupid and wrong abut the claim "the universe came from nothing."

That's a really nice description. I wish this was better known. But...

The algorithm only works (in the sense of leaving the parties psychologically satisfied) if their preferences are transitive (that is, if they are not insane).

In reality, even sane people's preferences change in pseudo-non-transitive ways as possibilities become actualities. So when Caleb gets the car, Adam is going to wish he'd valued it more highly, and so on. Our inner monkey won't be happy until it gets more than everyone else.

There is also a considerable body of data showing that our ability to judge the value of stuff is very poor. Happiness research has been big on this, showing that most of what people think will make them happy is radically inferior to easily predictable things that will actually make them happy.

So while the algorithm is beautiful and general and ought to be used wherever appropriate, it is not going to satisfy people, and it will then fall out of use because no one is going to say, "I am broken" when they can say "The algorithm is broken" instead.