erostrate

It would take about 30 years to get there, assuming constant acceleration & deceleration of 5g, so I hope you won't be dying too fast :)

It's not perfect, but it levels the playing field a lot more than other systems. And it's good at making sure talented but poor students have access to the best programs. At those the level is high, for example the best math program probably has the highest Fields medal per student ratio of any university in the world.

The downside is that universities and engineering schools are small and underfunded so they don't stand out in worldwide rankings.

How do rich people game the system? They can to some extent game it earlier, putting their kids in better high schools. But the kids will still have to take the same exams. So generally they just avoid this system and put their kids into business schools, which are not state funded and where admission is based on much softer and game-able criteria. Or they send them to US universities.

I still believe it is important to not be overly cynical. My mental model is that most people are neutral or good. It might be proved wrong say 5% of the time, when people are indeed put in MBS-like position. But the vast majority of the time it will be correct, and it will be much more useful than the cynical model where your normal friends are potential murderers that just lack opportunity. A more correct mental model is not always more useful.

[1] https://en.wikipedia.org/wiki/Situationism_(psychology) [2] https://plato.stanford.edu/entries/moral-character/#MorChaEm...

It's also worth remembering we're talking about someone who assassinates his critics and cuts them up into pieces. I would say the "bone saw" aspect outweighs the "playing AoE" aspect and he is very different from us.

Big companies, a rent-like business, no pressure to do things well, own lots of very sensitive personal data, clients giving the data away without realizing it, little regulatory oversight on the data front, etc.

I would expect them to sell or leak poorly anonymised personal data, leading to huge privacy issues.

Why hasn't this happened?

Sure, it could happen somewhere down the line but saying that blockchain is "destroying the moat" today is a big stretch. Blockchain barely just started to realize that the moat is actually much larger than previously thought. And that parts of the moat exist for a reason. It's barely starting to understand the moat, still far from attacking it, let alone destroying it.

Hopefully it will but I'm not holding my breath nor my bitcoins.

To me the Turing test gives a falsifiable experiment that defines what "thinking like a human being" means.

It does so by focusing only on the input output. To do that, it puts the system under test out of view of the experimenter and mediates the experiment with written text.

So yes, the input output is the important part, the Turing test, albeit flawed, is interesting, and any claim that machines can't think based on a definition of "think" that does not lend itself to a falsifiable experiment, such as Searle's, is uninteresting.

I will leave it at that and make a quick exit, as Adler suggests.

The way I think about this (from Feynman) is that entropy measures "how special" a macroscopic state is, on average. When you apply many microscopic independent fluctuations to a given state you are unlikely to end up with a "more special" state. That's how states are ordered at the macroscopic level giving the arrow of time a clear direction.

https://philosophynow.org/issues/46/Newtons_Flaming_Laser_Sw...

http://www.informationphilosopher.com/solutions/scientists/f...

First, Facebook is already used by pedophiles despite the lack of E2EE. 2500 arrests for child abuse last year according to the letter. An example is given of a pedophile identified through his messages with an 11 year old girl. With E2EE enabled he would not have been identified at that time.

Secondly, the letter highlights that a major problem is the combination of open profiles of children with E2EE messaging. This does not apply to any service, mainly Facebook and similar platforms.

Don't get me wrong, I am overall in favour of E2EE, and I share the usual cynicism about the government's intentions.

You can make the argument that although absence of E2EE does help identify pedophiles more easily, in the balance it is more important to enable E2EE to protect privacy.

But you should not make the argument that absence of E2EE on FB would not help identify pedophiles more easily, because this letter provides evidence that it actually does.

"Even in an airplane the number of vacuum tubes now approaches or exceeds a thousand. Other machines, containing up to 10,000 vacuum tubes, up to five times more crystals, and possibly more than 100,000 cores, now operate faultlessly over long periods, performing many millions of regulated, preplanned actions per second, with an expectation of only a few errors per day or week. Many such machines have been built to perform complicated scientific and engineering calculations and large scale accounting and logistical surveys. There is no doubt that they will be used for elaborate industrial process control, logistical, economic, and other planning"

"Basically, it's a perfect spying platform. You can't control it. You can't patch it. It can completely control your computer's hardware and software. And its purpose is remote monitoring. At the very least, we need to be able to look into these devices and see what's running on them."

https://www.schneier.com/blog/archives/2013/01/the_eavesdrop...

Some other interesting references:

- The waterfall argument in paragraph 6 of the mind-boggling "Why Philosophers Should Care About Computational Complexity". https://www.scottaaronson.com/papers/philos.pdf

- "NP-complete Problems and Physical Reality" https://www.scottaaronson.com/papers/npcomplete.pdf

There are non-quantum physical systems that exhibits positive and negative amplitudes and interference.

Can I factor large numbers by throwing rocks in a lake and measuring the water height at the right place? Why not?

There are also interesting differences in how economists view themselves compared to other humanities, see this (biased) paper: Fourcade et al, the Superiority of Economists http://pubs.aeaweb.org/doi/pdfplus/10.1257/jep.29.1.89 Economics tend to be a more insular and hierarchical discipline, and views itself as non-normative.

Economists are not intentionally supporting multinationals interests over people's interests. They do so as a side effect of their belief in the correctness and practical relevance of microeconomics and neo-classical theories. This behaviour is in turn encouraged and rewarded by multinationals and investment banks, leading dissenting voices to be marginalized. Of course there are notable exceptions, Stiglitz, Krugman, Rodrik, and a few others. But the general point is that if you want to get an economics PhD from a red brick university you will have to support this viewpoint.

Before deep learning people would manually design all these extra features sin(x_1), x_1^2, etc. because they thought it was necessary to fit this swiss roll dataset. So they would use a shallow network with all these features like this: http://imgur.com/H1cvt8d

Then the deep learning guys realized that you don't have to engineer all these extra features, you can just use basic features x_1, x_2 and let the network learn more complicated transformations in subsequent layers. So they would use a deep network with only x_1, x_2 as inputs: http://imgur.com/XBRjROP

Both these approaches work here (loss < 0.01). The difference is that for the first one you have to manually choose the extra features sin(x_1), x_1^2, ... for each problem. And the more complicated the problem the harder it is to design good features. People in the computer vision community spent years and years trying to design good features for e.g. object recognition. But finally some people realized that deep networks could learn these features themselves. And that's the main idea in deep learning.

Do you have a reference for that please? I'm interested in the subject.

It looks more like black magic than code to someone like me who doesn't know awk. It reminds me of these APL/J/K solutions I see on projecteuler.net. I imagine this is what happens when people completely ignore learning curve steepness and optimize for maximum productivity. Very impressive!

I wonder if they have tried to simply explain to commuters why it makes sense to stand on both sides. Treating people like intelligent and responsible adults often works much better than treating them like intellectually disabled children.

Quoting the text:

(1) Some critics concede that the man in the room doesn't understand Chinese, but hold that nevertheless running the program may create something that understands Chinese. These critics object to the inference from the claim that the man in the room does not understand Chinese to the conclusion that no understanding has been created. There might be understanding by a larger, or different, entity. This is the strategy of The Systems Reply and the Virtual Mind Reply. These replies hold that the output of the room reflects understanding of Chinese, but the understanding is not that of the room's operator.

https://letsencrypt.org/2015/10/29/phishing-and-malware.html

[1] http://www.staff.science.uu.nl/~gadda001/goodtheorist/index....