IsaacL

Implying my opinions are manipulated by shadowy forces is a very low form of trolling.

Yes, China's legal environment is unfriendly to building a large business as a foreigner. But no, foreigners in China can start tech businesses with "near-zero regulatory compliance requirements" - I knew 3 white guys in Chengdu who started a mobile marketing consultancy in their apartment.

China is basically anarchistic in some areas, authoritarian in others, is what I'm getting at.

Anyway, yeah, it's funny how often you meet experts on Chinese business who haven't actually done business in China but can go into elaborate detail about why it's a bad idea. I met lots of Western entrepreneurs out there - there are tons of foreigners running bars, restaurants, outsourcing firms , language schools, etc without much hassle from the government. You can even get around the minority ownership rule by incorporating in HK, or at least you used to be able to. I wouldn't start a big internet company out there, or a factory, but small companies in certain sectors seem to get left alone. It's probably more anarchic than the West, actually, which was one reason I liked it out there, though I've never seen myself settling there long term.

I did like living in China though, and after 2 years working in the UK considered moving back. Spent 2 weeks in Beijing in 2014 and checked out the local startup scene a bit. Unlike this guy I didn't really optimise my trip, I just went to a coworking space and asked people there if they knew of any big tech meetups (London has multiple meetups every night, with one big one (100s of attendees) at least once a week). Things seemed much quieter in Beijing - there were maybe 2 or 3 small tech events each week listed on meetup.com, for example. Maybe locals are meeting up with some other site, idk.

The other big reason is there seems to be on average, a lower level of investment in people and processes. In the West I've seen companies that were clearly in for the long haul, where the management really tried to put good systems in place, attract good people, train them, etc, and companies which were more opportunistic, weren't picky about who they hired, had a band-aid approach to problems, etc. Chinese businesses seemed skewed towards the latter (this seems like a developing country thing, it will be interesting if they become more like Japan as they develop).

Oh, and the latest ruling clique appears much more Marxist than their recent predecessors. That's not a promising trend.

However, other problems have even larger state spaces. Imagine writing an AI which read project Euler problem descriptions (in English) and output working code (in some given programming language). Keep outputs limited to 100-line scripts, max 80 characters per line.

There's roughly 100 usable characters in ASCII, so the possible space of 100-line programs is roughly:

(10^2)^(80 * 100) = 10^16000.

You could simplify this by having the AI work with predefined tokens rather than individual characters, but it's still a vast amount of combinations. Then consider 1000-line or 10000-line programs, and you see how high a mountain AI still has to climb. Humans are able to "compress" this state space via conceptual reasoning, which is much more complex than the "pattern recognition" many deep learning researchers are chasing.

(See "Introduction to Objectivist Epistemology" for more on how humans think in concepts - I'm planning to write more at some point on how this book shows where the practical limits of AI lie).

The counter to the reduced volumes is the fact that bitcoin has been enjoying a sustained price rally for the last 5 months.

It also seems like bitcoin is enjoying a rally, possibly in anticipation of the reward halving in 10 weeks (in theory the reward halving should already be priced in, but the bitcoin market is not the most rational).

It might be a superior means to store value. The theory is that is as the amount of dollars (or any fiat currency) constantly increases, anyone holding large quantities of dollars will want to exchange them for something with a more fixed supply. Usually this is stocks. Oil futures, property, and gold are also used as stores of value. (In countries with weak currencies, people also hoard dollars as a comparatively stronger source of value).

So the purely monetarist argument says if there are only 21m bitcoins, and an ever-growing amount of dollars, one would expect that the dollar price of bitcoin will only be stable at 0 or infinity. E.g., either the network crashes and bitcoin becomes worthless, or the dollar collapses and bitcoin reigns supreme.

http://nakamotoinstitute.org/mempool/im-hoarding-bitcoins-an... - see here for more on this theory as regards bitcoin.

The counter to this argument is that a. despite seeming logical, money supply inflation doesn't always seem to occur in reality. E.g., quantitive easing hasn't noticeably increased consumer prices.

b. Bitcoin is volatile, which means it isn't actually a safe store of value. One persuasive argument I heard (from goldbugs) is that bitcoin, having a fixed supply, has no mechanism to stabilise the price, and so booms and busts become self-reinforcing. With gold, on the other hand, the amount mined varies depending on the market price, which smoothes out booms and busts.

I'm expecting both gold and bitcoin to increase in value over the next ten years as the EU and US both lurch from crisis to crisis. This will bring a debate over the future of fiat currency into the mainstream, after which things will get very messy and political. (You can't run a welfare state on gold or bitcoin, for one).

I cured myself of these beliefs when I looked at how reality actually works (and when I stopped smoking weed). When Louis Gerstner turned around IBM as the company's first outside CEO, that was a result of his decisions and vision, not the emergent decision-making power of IBM. More fundamentally: people can think about companies, companies can't think about people. There are no collective minds, only individual minds working together.

The actual article in interesting. It sounds like Russian/Chinese/non-Western multinationals are growing in power, and with significantly less scruples than their Western counterparts.

I think that yours is a reasonable point of view but I do still think it's in error and that it would lead to massive rights violations in practice. E.g. how would the Google subdivisions be forced to compete if Page and Brin still owned a controlling stake of each?

Have a read of this if you want an alternative point of view (yes, I'm a randroid). http://itech.fgcu.edu/faculty/bhobbs/Capitalism-The%20Unknow...

States have justified their actions by many different codes of ethics. I know my answer to the legitimate role of the state: to protect individual rights.

In every human society in history, there have been people who choose to rob, rape and kill. There's no pleasant way to deal with such people, but it is the role of the state to deal with them, or you have anarchy. It isn't the role of the state to prevent "emotional damage", "bad wives" or "bankers".

Stafford Smith's position seems to be a twisted version of Christian morality where all problems are caused by meanness. Since the justice system is an institutionalised form of meanness, it must be wicked -- indeed, it's probably responsible for crime and terrorism.

This is a really evil worldview.

By robbing selected Peter to pay for collective Paul;

But, though we had plenty of money, there was nothing our money could buy,

And the Gods of the Copybook Headings said: "If you don't work you die."

[...]

And that after this is accomplished, and the brave new world begins

When all men are paid for existing and no man must pay for his sins,

As surely as Water will wet us, as surely as Fire will burn,

The Gods of the Copybook Headings with terror and slaughter return!

(http://www.kiplingsociety.co.uk/poems_copybook.htm)

The essential point I want to make is that just because you have a right to something, does not mean you have a right to the material implementation of that right. (Because that would mean you had a right to the labour and property of others, i.e. a right to violate their rights).

You have a right to free speech. You do not have the right to publish your opinions in any outlet of your choice (because that would violate the publishers' rights). Even if you lived in a small town with only one newspaper, and they wouldn't print your letters to the editor, it still wouldn't be right for you to compel them to print your letters by legal means. They built the audience, its their right to communicate with them as they see fit. (This isn't censorship, because you're still free to hand out flyers in the street, start your own newspaper, sue the existing newspaper if they commit libel, etc).

Your argument is essentially that because Google is "the world's largest information system", we shouldn't be "letting" them put whatever content they want on their pages, for whatever reason they want. The implication is that if someone builds something really useful, that thing should become public property. (People are saying the same thing about Twitter).

AdWords is absolutely moral because a privately-run service needs a source of funding. (I should note that I do think many PPC ad networks are exploitative and immoral, but that's not the case for Google).

Unless someone sets up a nonprofit search engine run by donations, the only alternative is a public, state-funded search engine -- that's the only way to get an "information system not biased by money". On the unlikely possibility that the state could build and maintain such a system, it would be the most powerful tool for potential censorship imaginable.

E.g. if you have a set of things which vary in many dimensions (even if each element is no more than a set "distance" from normality), very few elements will belong in the "centre" (normal on all dimensions) and most will belong on a spike (wildly extreme in 1 or 2 dimensions).

Rand is pro-government where it's needed. She's even pro-environmental regulation - she says it's justified where there's an objective threat to human life (as with air pollution), or violation of property rights (as with noise pollution).

(For what it's worth: I don't hate poor people, nor do I think all rich people deserve their wealth, nor do I think the free market is a universal panacea for all ills.

What Rand actually says is that there are good people and there are bad people, and the defining characteristic of good people, the essence of virtue is not altruism, but productive work. People rarely mention that there were truck drivers and manual workers in Galt's Gulch, and that half of the villains in the novel are unscrupulous businessmen).

Rand didn't actually say that much that's original about economics per se. This leads people on both left and right to dismiss her as a simplistic thinker. This ignores what is original is what she's saying.

Unlike many libertarians, she doesn't argue that the free market is good because it leads to good results for society. She says that a capitalist society is good because it's the only society which fully protects individual rights. It's not a coincidence that capitalist societies are more prosperous -- however, when this is used as the justification for capitalism, the logical conclusion is to have a semi-free market where the government tries to manage the economy and optimise different variables (GDP, employment, inflation, etc), i.e. the system in almost every western country today. I'd argue the constant financial crises and ballooning sovereign debt show that this system cannot possibly be sustained.

Also: the Journal of Ayn Rand Studies is not respected in official Objectivist circles. It's one of a number of organisations which attempt to water down Rand's message to appeal to more people, but as a result put off both Objectivists and non-Objectivists.

I'm genuinely curious, if any of his fans are reading this, what is his appeal? Is it the political messages and well-drawn near future settings?

The first two comments on this thread were definitely snarkier and more aggressive. Part of me thinks there were advantages to the earlier culture of HN, though. People were more likely to be challenged on their statements, and forced to defend their beliefs.

The timeline is roughly (apologies if dates are inexact, but this is roughly correct from memory):

AD64: Nero perseuctes the Christians in Rome, who at that time were a tiny community

AD 65-249: Christianity is largely ignored and left to grow in peace. During this time, Rome enjoys a golden age under five virtuous emperors, then begins to disintegrate under the reign of various tyrants and opportunists.

AD 250-260: Rome is falling apart under various civil wars and there's two major state-led persecutions (by Decian and Valerian) of Christians. (The motivation being "rid the Empire of this impious cult which is weakening our social fabric").

AD 260-300: The "long peace of the Church", where Christianity is officially tolerated.

AD 300-310: Diocletian comes to power and starts a brutal series of persecutions to wipe out Christianity for good. It's from the Diocletian persecution that most of the images of brutal Roman suppression come from.

AD 310-325: Various civil wars, leading Constantine to come to power.

AD 325: Constantine establishes Christianity as the state religion.

So though there's one small persecution in 64, then two medium-sized persecutions in the 250s, then a massive persecution in the 300s, between these Christianity is either ignored or tolerated by the Roman state. There's evidence of mobs of Roman citizens attacking Christians independently of any state backing, but aside from that, Christians were largely left alone for a 200 year period.

This is the age of the 5 "enlightened" emperors - Trajan's attitude is informative. The Romans saw Christians as impious atheists (for not worshipping the Roman gods) and superstitious cultists (for following a novel religion), and they were troubled by the growth of this sect; Trajan took it for granted that Christians who refused to worship the Roman gods should be punished. But, he also states that actively seeking out Christians would be "out of keeping with the spirit of our age"- a weird Roman kind of tolerance, but there you go.

"You observed proper procedure, my dear Pliny, in sifting the cases of those who had been denounced to you as Christians. For it is not possible to lay down any general rule to serve as a kind of fixed standard. They are not to be sought out; if they are denounced and proved guilty, they are to be punished, with this reservation, that whoever denies that he is a Christian and really proves it--that is, by worshiping our gods--even though he was under suspicion in the past, shall obtain pardon through repentance. But anonymously posted accusations ought to have no place in any prosecution. For this is both a dangerous kind of precedent and out of keeping with the spirit of our age."

(Trajan's letter to Pliny here: http://faculty.georgetown.edu/jod/texts/pliny.html)

"This paradox is actually very similar to Banach-Tarski, but involves a violation of additivity of probability rather than additivity of volume.

Consider the case of a finite number N of prisoners, with each hat being assigned independently at random. Your intuition in this case is correct: each prisoner has only a 50% chance of going free. If we sum this probability over all the prisoners and use Fubini’s theorem, we conclude that the expected number of prisoners that go free is N/2. So we cannot pull off a trick of the sort described above.

If we have an infinite number of prisoners, with the hats assigned randomly (thus, we are working on the Bernoulli space {\Bbb Z}_2^{\Bbb N}), and one uses the strategy coming from the axiom of choice, then the event E_j that the j^th prisoner does not go free is not measurable, but formally has probability 1/2 in the sense that E_j and its translate E_j + e_j partition {\Bbb Z}_2^{\Bbb N} where e_j is the j^th basis element, or in more prosaic language, if the j^th prisoner’s hat gets switched, this flips whether the prisoner gets to go free or not. The “paradox” is the fact that while the E_j all seem to have probability 1/2, each element of the event space lies in only finitely many of the E_j. This can be seen to violate Fubini’s theorem – if the E_j are all measurable. Of course, the E_j are not measurable, and so one’s intuition on probability should not be trusted here.

There is a way to rephrase the paradox in which the axiom of choice is eliminated, and the difficulty is then shifted to the construction of product measure. Suppose the warden can only assign a finite number of black hats, but is otherwise unconstrained. The warden therefore picks a configuration “uniformly at random” among all the configurations with finitely many black hats (I’ll come back to this later). Then, one can again argue that each prisoner has only a 50% chance of guessing his or her own hat correctly, even if the prisoner gets to see all other hats, since both remaining configurations are possible and thus “equally likely”. But, of course, if everybody guesses white, then all but finitely many go free. Here, the difficulty is that the group \lim_{n \to \infty} {\Bbb Z}_2^n is not compact and so does not support a normalised Haar measure. (The problem here is similar to the two envelopes problem, which is again caused by a lack of a normalised Haar measure.)"

I found thinking about the two envelopes problem he mentions (https://en.wikipedia.org/wiki/Two_envelopes_problem) a more accessible way to wrap one's head around the paradoxes that arise when you compare a (potential) infinite quantity to a (known) finite quantity.

(The crux of the envelope paradox is that each envelope has an infinite expected value, but in reality only contains a finite amount. I can sorta-kinda see how the infinite hats problem is another example of the same paradox).

:D - I'm spending this sunday afternoon learning abut geodesic grids (https://en.wikibooks.org/wiki/Geodesic_Grids) as part of a project to teach myself d3.js.

Otherwise - I agree with the author. It was strange seeing programming become a huge thing about 2/3 years ago, but realistically, most people don't have the mathematical/logical abilities to ever get really productive. Which is fine, they should focus on what they're good at, rather than beat themselves up for not mastering a quite specialised skill.

I gather the growth in coding bootcamps has lowered the demand for junior devs, but there's still a huge demand for more experienced people.

The metals and plastics used in their manufacture is pretty minimal compared to cars or home appliances.

Their power consumption is low compared to heating or lighting (a macbook consumes power equivalent to about two lightbulbs, and most devices use less).

Earth metals represent probably the biggest environmental impact of consumer electronics. They're not actually rare, just expensive to mine, so there isn't an issue of sustainability. (As in, the earth metals aren't going to run out any time soon. There's some nasty environmental damage in the vicinity of the mines, but that's limited to a particular area).

I'm trying to fathom the reasoning behind this comment. Is it that because electronics are really useful and beneficial, they must have a proportional cost?

American readers tend to assume the worst about government regulation of the web, but a charitable reading suggests a pro-copyright ruling might not be completely ridiculous. They'd have to draw a tight boundary to ensure they only cover blatant cases (like the blog which linked to leaked photos), and avoid leaving all users open to accusations of copyright infringement. If they

- limited it to cases where it was clear the intent was to link to copyrighted content the owner expressly didn't want to be linked to (assuming that if the owner hasn't expressed this, content posted online is, by default, OK to link to)

- added exclusions for a) old links, b) user-submitted links, c) links to content which changed

it wouldn't be an unreasonable law. I don't know if the final decision will be that sensible, but I wouldn't rule it out. The cookie regulation wasn't especially onerous and probably helped educate users about the legitimate role of cookies.

I should also point out that this is a judicial decision, not something created by the legislature. The court has to decide how to apply the existing law to this case, and in doing so establishes new case law.

(I made this comment as I don't think the EU can be criticised on the grounds of producing ill-thought-out regulations. As technocratic bureaucracies go, the EU is efficient and works well. Instead, I'm opposed to technocracy on principle, and I think 30-50 years out the EU will have evolved into inefficient Brezhnevism).

I know that deep learning systems can recognise, e.g., different animal toys held in the hand, at different distances from the camera. How about shapes which are rotated? (I assume shapes rotated with the same face showing the camera are fairly trivial to recognise, I mean the other two rotations - 'pitch' and 'yaw').

I have a hunch that humans have a ton of perceptual hardware devoted to precisely this task, and that deep neural nets are going to have a hard time cracking it. Is my hunch accurate?

The only things on that list I have a real problem with are freedom of speech and LGBT rights. Capital punishment and caning shock Westerners, but Singapore's legal system is actually fair and efficient; it's not like the procedural trainwreck which is the US justice system where countless criminals walk free and countless law-abiding citizens are rotting in jail.

Conscription is pretty much necessary for a micro-state sandwiched between two nations which historically hated their guts. Migrant workers aren't forced to be there (and unlike Qatar or Dubai, they're not generally lured in under false pretences). Even the freedom of expression rules are very similar to the de-facto/de-jure hate speech rules which exist in the West, and they exist for a similar reason - Singapore wants to minimise Chinese/Malay tensions.

EDIT: I just remembered that this HN user - https://news.ycombinator.com/threads?id=frenchieinsg - initially piqued my interest in Singapore. Might be of interest to others of similar mind.

Even lower-middle class professions like teaching, nursing, policing, office admin, etc are much less routine than people realise. (Tech is obsessed with the idea of, e.g., replacing teachers with e-learning software, but what usually happens is that the teachers end up using the e-learning software as another tool to do their job).

Vehicle automation is one large area I can genuinely see becoming automated (possibly the success in this area is because the AI doesn't need to manipulate the environment, just navigate through it). There are still many blue collar jobs which are really hard to automate - cleaning, gardening, building maintenance, etc, etc.

(Honestly, I see the most likely outcome being that first world countries develop a grey-market, third-world economy, where low-income people work cash-in-hand, below-minimum-wage jobs running cheap food stalls, cleaning laundry, or offering other convenience services on the cheap. It's not as utopian as everyone getting a basic income and using their unlimited free time to paint watercolours, but it works out as a more robust social safety net).