Almaviva

"So it’s 99.5% that it occurred due to warming over the industrial era"

There are at least a couple of statistical fallacies in this conclusion. And there isn't a lack of agreement about that.

One problem with p-value tests is precisely that people misunderstand what the p-value means, which is where basic education comes in. It could save people from believing a lot of things they shouldn't. (Like many health and fitness crazes over the last generation, for instance) Or at the very least, we could train science journalists.

I played a game of Monopoly where I started with a roll of 12. There was a 97% chance I correctly rejected the hypothesis that the dice were fair. Maybe this is true in some sense? But it's still somewhere between misleading and nonsensical to say.

The 0.5% quote is fine. It's turning it into a 99.5% that is not fine.

It's essentially like saying that a pair of dice that rolled a 12 has a 97% chance of being loaded, because there's only a 1 in 36 chance of rolling that high.

> Why is it wrong?

https://xkcd.com/882/

If you look at 200 rivers, it would not be surprising to find something that naturally occurs 0.5% of the time. It is not correct to say that there is a 99.5% chance that this is due to non-natural causes.

"The calculations put chance of the piracy having occured due to natural variability at 0.5%. “So it’s 99.5% that it occurred due to warming over the industrial era,” said Best."

We need to train children in school to understand what statistical confidence means, so that we stop saying wrong things like this.

> Also, is js really even the best tool for this?

For visualizing the Mandelbrot set? No. For writing an article on the web about incrementally building it up? Yes.

Presumably "Machine Language" which is what we called Assembly, back when the translation between what you wrote and what the CPU actually did was pretty transparent.

Lots of people visit links on the clear web showing illegal and horrific acts. If the full extent of a crime is filling out an http form with a fake email to see some pictures and video, it's still not clear that this is so far beyond the pale that years of prison for hundreds of thousands of people is the best solution.

If the site mentioned in the article had 150,000 members as they say, that's a lot of people to lock up for years for pointing a browser at a url.

The problem is fundamental and serious, what to do when a growing number of people have literally nothing valuable to contribute to an economy. But horses aren't a good analogy to shed light on it.

Horses are more analogous to steam engines than people, in their historical function on the economy.

Horses haven't shown the same level of adaptation to different skill sets that humans have.

> you cannot pick this up in any meaningful way in a "few months" of after hours/weekend study

You can't pick up coding like this either. See Peter Norvig's famous "Teach yourself programming in 10 years" article. The delta in the wisdom you obtain, between a few side projects over months and battle hardened experience with real products and code bases over years, is immense.

> whole grains

Yet the glycemic index of whole wheat bread is higher than Coca Cola. (Look it up!) I.e. the same amount of calories of bread will spike your blood sugar more quickly than Coke.

Metabolically, starchy foods including grains are a lot more like sugar than people who think they are making rational health decisions want to believe.

Now consider that in human history, about 40% of men were able to breed. So 30% of the population goes extinct every generation without passing on genes, assuming all women reproduce. So with a few hundred iterations of this kind of 70:30 split, you could see how evolution could happen quite quickly even under normal circumstances.

Most of the smartest humans spend a significant part of their lives in learning and mastery of skills that go beyond survival.

If it has any general goal at all that it takes seriously, it will be better achieved with self-improvement.

We know that AlphaGo uses a Monte Carlo tree search, and presumably contains innovations and refinements to techniques applicable to turn based board games with perfect information and a clear binary win condition.

We also know it uses a Deep Learning algorithm to imitate play from the best humans. And presumably, being stronger than any human now, it could also recursively refine and train on its own games.

Do we know how important the Deep Learning component is? Would AlphaGo be just as strong or nearly so without that part?

I know that over the last few years chess engines have become dramatically more powerful, even on the same hardware, to the tune of hundreds of Elo points. They still stick to the classical techniques though (Alpha-Beta pruning, minimax is still the core). This is not to understate the refinements, but the general ideas are (I think?) only applicable to turn based games with finite moves and a binary mathematical criterion for victory.

Is it possible that the improvements AlphaGo has made are mostly of this type, and the Deep Learning parts are not that important to its actual play strength? Or can we totally rule this out? Is there an in depth discussion of this question somewhere?

There's one huge incentive I can relate to: a product for the n percent of lonely men each generation who don't have the social skills, status, or otherwise cannot attract a mate. And afterwards, the men who would find a perfect and sentient companion to be superior to dating and relationships. I'd bet that logs of Siri's queries support this idea quite well.

> Evolution had billions of years, working in a massively parallel way to work this out after all.

And it also got to work on it with a hundred billion stars with earth-like planets, in each of a hundred billion galaxies, and (maybe) in a basically infinite space of universes. All in parallel with infinite time to play with.

Is it so far-fetched to postulate that humans will hit our limits before getting this far?

> If you think there's something else, that is by definition supernatural.

First, how others perceive your words usually matters more than the precise definitions you have in mind. If I describe myself as believing in the supernatural, it has a lot of connotations I don't want.

Second, the possibility that consciousness (in some sense) could be built into physics themselves. So it doesn't contradict that the brain is made of subatomic particles, but the workings of the universe have more correlations built in that could work at different levels. Something I think is comparable is that many cosmologists think that the arrow of time, which events come before which, can be explained by a universe with initial conditions of extreme low entropy. So there's nothing within the universe that explains why the initial conditions were like that, but it's a real property of our universe.

Maybe consciousness could be an end state that the universe is progressing toward. Is that really so much weirder than working from an initial state of extremely low entropy?

> If we re-distribute our wealth, population growth will subside.

I don't get this argument. People want to fuck and reproduce, and the next generation will contain the genes of precisely the people who most want to, and so on. Something needs to check this other than wishful thinking.

> a way of asserting that you think the probability of a technological civilization arising is vanishingly small

I have no idea one way or another. I wonder why many scientists are confident that extraterrestrial intelligent life exists though. Viewing through the lens of one example in a random (otherwise) unremarkable multiverse, it's not clear to me why this should be so.

Say a 1080p monitor display is completely randomized, until it shows a clear visible five-pointed star. Given I'm looking at a star, what is the likelihood there is another one on the screen? Practically 0, I'd think. Why is the emergence of an intelligent civilization in the universe different, assuming I take a multiverse and the anthropic principle seriously?

(In other words, given that a universe could have multiple intelligent civilizations, why is it the case that if I'm in a random sample with at least one, there is likely to be another?)

I think words are very powerful, in particular "microservices" vs "monolith". By accepting those words, we imply the conclusion: microservices sound sexy and lean and elegant - who can argue with separation of concerns? And a monolith is a big unchangeable rock.

I think we need a better word for apps that are single tight self-contained systems than "monolith". You can design elegant interfaces, and avoid creating a sloppy mess, with function calls or objects too.

Do the test run quickly, are they clear and readable, do they help find regression problems, is it reasonably easy to tell what went wrong when they fail, does the time and effort and peace of mind saved exceed that invested (over the life of the project, amortized, sometimes, for the value of iterating quickly)?

These are important questions, whether you call a test integration, functional, or unit, is not an important question.

People seem to have a lot of faith that a drive or urge to reproduce is not something that can be selected for and transmitted through generations, either genetically or culturally, in human beings. If it can be selected for, it will be, and very rapidly, by simple math.

> If there are elements of consciousness that are completely separate from the physical world (in that they have no effect on it), we cannot talk about them, nor even think about them.

Yet here we are, doing both, talking as well as thinking, right? If I talk about a universe where you'd never "woken up" as a subjective individual, and the present time was just like (to both of us) the previous 13.7 billion years, you know exactly the concept I mean, whether it's expressible in terms that have meaning beyond a single reader who knows (in the Cartesian sense) that their (subjective) existence is something special that only they can observe.

I wasn't clear, it's vitally important to predict behaviour in practice at a poetic naturalism level, but there seems to be no reason to adopt strong emergence: consciousness doesn't introduce any new explanatory power that couldn't, in principle, be explained by physics, or even chemistry, biology and so forth.

See Dennett's interview with Bob Wright starting at about 30 minutes: https://www.youtube.com/watch?v=Ss0aCWpNzSM

I'm just not ready to stop there, despite this all being true, I think it seems not parsimonious to believe there is nothing fundamentally mysterious about how I feel like something subjectively.

Rationally I feel like we have to accept a Daniel Dennett view: with what we know about the brain, all human behaviour is consistent with the laws of physics, and changing the brain very much changes "what it's like" to be - it doesn't take very many experiences with drugs, even alcohol, to realize this. Consciousness, then, has no explanatory power whatsoever. A lot of people argue at this point, but they have nothing substantial to stand on. Maybe they just want to believe that Free Will is something special.

However, I think the Hard Problem is a knot that can never go away. I notice physicists are willing to explain much of the properties of the universe (like time having a direction) just by saying there was a very specific initial condition, a very low entropy state that (by definition) is extraordinarily unlikely. So what about saying there's another constraint on conditions of the universe, where we have physical laws that create consciousness. Is this just an anthropic principle of some sort?

I was listening to Sean Carroll's "The Big Picture" audiobook. He argues that I can't really be sure I'm not a p-zombie. But I think I can: I'm with Descartes on this one - it's the one thing I actually _can_ be sure of. I'm sure I have an inner experience (that explains nothing scientifically to anyone else) but I can easily imagine not having it, with no change to the laws of physics.

Does anyone know of a good read that isn't full of jargon that fleshes out these ideas some more?

That's not a good comparison. Red and blue have very objective definitions given that we have three types of light receptors in our eyes, and a given kind of light can be measured to activate the "blue" or "red" kind without any subjective input.