We already can't really 'deconstruct' or 'understand' the decisions of the simplest neural networks acting as classifiers. A model of something difficult to understand is likely to be difficult to understand. Certainly this would be the case for any sort of advanced AI. But there is nothing magical in the human brain. It's all mushy cell parts and salty fluids. There's no reason at all you couldn't break it down into all its components and say "aha – it chose vanilla because the connection weights + prior state + given sensory input pushed the system into this basin of attraction as opposed to that one." Any AI worth its weight in whatever the hell it gets made out of will be a learning machine, and trying to understand 'why' its connection weights are configured in a particular way won't be any easier than it would be for a human...

If by 'failure' you mean razed to the ground by either communists, republics, or both – then you're probably right. The US went to 'nam to fight direct democracy (so did the communists). Same in Italy. And Guatemala. And Cuba. And, well... you probably get the point. It's notoriously difficult to maintain a functional society when you are being murdered all the time...

Close. Most stimulants are dopaminergic, which is a catecholamine and only has a single ring; (nor)epinephrine are also in that family. The indole backbone is in the tryptamines, including serotonin, and in psychotroptics like psilocybin. Also present in LSD, although given how monstrous that molecule is, it doesn't figure as prominently as a 'backbone'.

That said – I'm not sure we really want to seed our water with a substance that converts a broad family of small organic molecules into psychotropic compounds... Much as it certainly sounds like good fun, I suspect there might be unintended side effects...

Except none of this is really about technology, per se, or even really about spying. It's about image and reputation. The same reason Chelsea's diplomatic cables got everyone so pissy. Much as superior force and technology are certainly big factors on the world stage, they aren't the final determinants. If no one likes you and no one trusts you, no one will cooperate with you. Your trade will become costly, your citizens will find it increasingly unsafe to travel, your people will revolt. You will have to resort to war and extortion on an indefensible number of fronts, inside and out, and the rest of the world will be increasingly likely to align themselves against you. Modern politicking is, in bulk, about propaganda and message control. The National Security State is losing that battle, and it is important to keep striking while they backpedal. They will never capitulate, and in fact will only redouble their efforts, but trust and respect cannot be rebuilt with more defence spending. They have to be earned, and the US govt is haemorrhaging whatever stock they have left...

Also, why would anyone think that you would hear about political dissidents being jailed arbitrarily? The media ain't too fond of dissidents either. But believe me, if you're part of the wrong ideology they'll get a warrant to raid your house for books and put you in front of a secret tribunal...

It shouldn't be too surprising...

Ah yes - the ever dangerous 'smoking vehicle'. Real masterminds there. We sure dodged a bullet on that one. Thank god the NSA, I mean CIA, I mean FBI, I mean police, oh wait I mean street vendors were able to foil that plot with their sophisticated 'eyes'... I haven't stopped shaking ever since.

You might revisit the definition of 'exactly'...

My reading of the scare quotes was, within the context of the post, that the national security state tells us that the people it kills are our enemies, when this is often not the case. The reading you favour, that it implies that Taliban and AQ aren't enemies, makes the presumption that the people being killed are in fact Taliban and AQ. We know this often is untrue. Hence the "enemies" scare quotes. The "our" scare quotes presumably alludes to the fact that US foreign policy is generally geared towards specific benefits for specific stakeholders – and those stakeholders are rarely 'the people of the united states' or 'the people of the world'. The fact is that 'Taliban' and 'AQ' are used as catch-all labels for a tumultuous mass of politics, ideologies, and capabilities – and while some of those subsets have indeed aligned themselves explicitly against the interests of the 'average joe,' many are only enemies of capitalist imperialism – which is itself an enemy of the people. Telling us that they are fighting "our" enemies, is among the simplest of possible manipulative deceptions...

The issue of definitions is pivotal, and for this reason I identify as igtheist and I invite you and all other atheists, tired of being accused of faith, to join me!

The fundamental premise is very simple. The question "do you believe in God?" is actually meaningless gibberish unless you also provide a definition of 'god'. No one ever does this, and if you corner someone on it, they'll most often be reduced to absurd statements like "well god could even be this blade of grass," at which point you can comfortably agree with them. "Ok - I also believe in that blade of grass."

The secondary aspect is that, if a definition is provided, it must be a definition which, at least in principle (even if the tests are wildly impractical), has some measurable consequence for the universe. If it fails this test, than the definition reduces to the empty set (i.e., the only set satisfying A = ~A).

No faith involved!

His description wasn't exactly... eloquent – but he is basically correct. I'd say it's still impressive – in the same way that any successful application of bleeding-edge technology is impressive – but there's no real theoretical advancement here at all. They combined two well-established techniques: EEG-decoding (the stuff that's so well established you can buy your own crappy sets and play boring games 'with your mind'), and TMS (invented circa 1910 by this badass - Silvanus Thompson) which now-a-days can be applied to replicate the 1950s era work of Wilder Penfield in electrical stimulation.

In other words: we were already able to trigger motor impulses externally, and we were already able to record and decode the neural signatures of sufficiently different thoughts. What they did here has huge media value, but it isn't particularly ground breaking. Even TFA states this:

The problem with this as support for 'living on' or a 'soul' or what have you, is that the encoding is arbitrary for each individual. If you are willing to allow arbitrary encoding, then the only constraint on what is 'recoverable' is the number of bits. So this kind of a system could also 'recover' a great number of things which never existed. I think it's a bit disingenuous to claim that this is any kind of preservation or storage mechanism. And it certainly isn't a runtime-supporting environment. Humans aren't built of particularly interesting materials – it is the shape/configuration/topology/what-have-you that gives us our complexity. Saying that you can in principle reconstruct a human after it has deteriorated isn't an argument for the soul. It's a necessary conclusion for deterministic materialism.

Well, considering that most organisms do in fact change shape throughout the lifespan, whether from development, diet, exercise, or injury, it's both adaptive and pragmatic to simply map your sense of agency to objects whose observable properties correlate with your sensations, or with your intended actions. This is how children learn to pick things up, to walk, and to talk. It's also the mechanism that subserves the extension of receptive fields during tool-use. This is just an elaborated instance of the practically-ancient rubber arm illusion (show someone a rubber arm, and hide their real arm. simultaneously stroke the fake and the real arm for a little while, and the person will come to view it as their own. hit the rubber hand with a hammer, and the person will get a gsr response greater than that for observing the same action without entrainment). Nothing new here.

No way that'll stay legal. Fortunately you can foil these things with the right makeup...

First, to be clear – I'm not going to suggest that MSNBC isn't biased – but I do have some reservations about the study. I'll just list the four big ones, for me:

1) Funding: The source of the report – the Project for Excellence in Journalism (PEJ) is funded by the Pew Charitable Trusts. The funds for this charity come from the children of Joseph N. Pew, the founder of Sun Oil Company (Sunoco). Half of the current board of directors are from the Pew family. Does this mean everything they do is biased? Nope. But you sure don't see a lot of left think tanks financed by big oil...

2) Methodological Problems: They have two pages that list details of their methodologies, but nowhere is it indicated how 'fact' vs. 'opinion' was determined by the coders (nor do they report inter-rater reliability for these ratings). This isn't a simple task. For instance, if one newscaster takes a disputed datum and says "It seems that X" while another says "It is the case that X" – the latter will sound more factual, whereas the former is more factual. This sort of ambiguity is troubling. Further, from the report it would seem that the fact-opinion ratios scale with the proportion of air time given to interviews. This shouldn't come as much surprise – but the problem then is in giving primacy to the fact-opinion bias, rather than to the programming selection. The opposing bias would be something like "msnbc provides a greater breadth of viewpoints through more time devoted to interviews with guests". They should have given independent fact-opinion scales for each of the program types, or at minimum provided a program-normalized score instead of an aggregate.

3) Methodological bias: The following statement in the methodology was of particular interest to me: "Early Evening and Prime time (6 PM - 11PM) together as a unit, rather than separating out talk and news or early prime and late prime. Within this five hour period, we included all programming that focuses on general news events of the day. Basically, this removes three programs: Fox's Greta Van Susteren, which is more narrowly focused on crime, CNN's Larry King which as often as not is focused on entertainment or personal stories rather than news events and MSNBC's documentaries program." So they removed two highly opinionated programs from CNN and Fox, respectively, while removing documentaries from msnbc... The choices are odd, though the justification sounds reasonable – but it seems like another strong case for normalization at the analysis stage.

4) Rhetorical bias: From the report, we have the following two 'opener' paragraphs under the 'Comparisons by Cable Channel':

And then we have the following gem that I stumbled across in another PEJ/PEW report: A First Look at Coverage of the 2008 Presidential Campaign. After noting various metrics of coverage time and coverage tone, the report asks:

Subsequently, with seeming reluctance, they acknowledge that there may be other explanations. For instance:

So – to reiterate – I'm quite sure that msnbc is a wildly opinionated more-left-than-right outlet. But there should be innate skepticism when someone suggests that Fox is not a wildly opinionated more-right-than-left outlet. The linked study, while interesting, doesn't quite go the distance for me...

If I were "told the story" I would absolutely ignore it. People tell stories all the time. It doesn't make for very compelling evidence. If you have a reference to documented evidence surrounding such a story, then we are in a different area entirely. If you can do so, I'd appreciate it.

As to whatever a 'fundamental atheist' is, which somehow differs from an 'atheist' in any sense other than rhetorical – I'm not one. I was actually quite please when a friend introduced me to the term igtheist which captures my beliefs perfectly. In any event – when did God enter into this discussion?