textor

I agree in principle that learning things can be tedious at times, but... learning, memory retention and reward system are very tightly integrated. It's almost laughably ineffective to learn despite pain, relative to learning with joy. You know there are people who could complete your song-memorizing ordeal in 1/10th of the time, not feeling any major frustration? Because most of the work is happening effortlessly, the song being sung in their heads on autopilot, reconstructing and optimizing?

I won't claim that toughing it out isn't worthy or that there always is a better way, however it seems clear that top performers in every field tend not to have to tough it out very much. This is unfair but not smart to ignore.

Not really. Cats do "spaz out" episodically but a typical human with the same amount of exercise would probably be a flabby sack of fat with cardiovascular issues. The activity isn't predictive of the rate of muscle atrophy to the same degree across species, because loss of muscle is mediated by a specific, variably effective biochemical cascade and is not a consequence of inactivity per se. Similarly, loss of memories or cognitive skills is an active process that may have negligible speed even without much utilization.

But I'm nitpicking.

> We don't have the computational power yet

Certainly you do realise that this has been a moving goalpost for half a century? It seems that lately people started to avoid giving a concrete estimate of the power required, though. It was so easy in the 90-s! «Human visual/verbal system processes gygabytes per second/has n flops to the xth» or the like. Well, now we have that and more; how come a deeper modeling, a finer processing, a more complicated network has come to be needed?

And your examples are incorrect, for example Navier-Stokes equations plus some general physics knowledge always have allowed us to estimate how much data we need for a certain fidelity of a finite-term weather forecast. Certainly we need more for a complete climate model, but we know what we need. No such thing about the brain.

It's an easy way to score some rationality points by voicing rejection of "magic", but it's a strawman. Nobody will bother arguing for a mythical homunculus in the seat of the soul, nor even for a concise formula summing up the workings of the mind. Pick harder targets. "It's just big" or "it's just a bunch of heuristics cobbled together" is a non-explanation. The brain is not a Rube Goldberg machine that manages to produce any sort of work simply due to its excessive complexity – it is energetically economical, taking into account that neurons are living cells that need to sustain their metabolism and not merely "compute" when provided with energy. Its discrete elements aren't really small by today's standards, nor are they fast. The number of synapses is ridiculous, but since they aren't independent, at a glance it doesn't add that much complexity too (unless we abandon reason and emulate everything close to the physical level).

Yet we have failed to realistically emulate a worm. By all accounts we have enough power for 302 neurons already. There's no workload to give to overwhelm available supercomputers. It's knowledge and understanding that we lack, and it's high time to give up on the delusion that more power, naturally coming in the future, will somehow enable a creation of predictive brain model, for this would truly be magic.

Non sequitur. You're having this conversation because the current political climate in the USA leads to conflicting opinions on this subject.

How come you're so confident that "that’s not how things work"? Do you have any idea about epistemic standards in social sciences? And if qualifications are not completely measurable, then how much more qualified is "slightly more qualified" and thus disposable? Lastly, on what grounds can people that promote diversity claim to have enough expertise to regulate this process of affordable exclusion?

This isn't about the number of examples. It's plainly disingenious to say that "TD has been referred plenty, not under that name". Basically, in such way one demands the right to reduce supply side theories – or, in popular practice, to reduce every argument in favor of any taxation rate lower than proposed by the opponent – to a strawman of choice, as if the strawman was the original idea that's "covered up" or "dogwhistled" or whatever. Like this, in extreme:

A: We need to impose 100% tax on the top 10%, to achieve economic growth and lower inequality.

B: This isn't even wrong, this is asinine.

A: Oh, trickle-down much?

Why do this? Naturally, because genuinely debubunking the specific arguments against high tax rates is a lot harder than joking about the bourgeoise that promote TD because of their greed.

Well I know people who are perfectly fine with hinting bugs while laboring for another, who find some obscure pleasure in repetitive work that is normally considered boring. It's not incomprehensible. But yes, they are rewarded handsomely. It's a rate trait, after all.

Not trying to invalidate your larger argument, but it's worth noting that there is no such thing as "trickle-down theory"https://blogs.spectator.co.uk/2015/04/sorry-but-trickle-down.... Never was, never will be. It's a strawman through and through, originating in mockery, amounting to nothing more than "screw the poor" slogan that no self-interested economist could have endorsed. Its only purpose is to ridicule the perfectly valid – and hard – question of optimal tax policy that achieves long-term budget prificit without stifling development of businesses. Hilariously enough, some less than savvy people with conservative inclinations have defended and adopted this strawman as if it was a legitimate model, but this has more to do with meme dynamics than with economic policies.

As for your larger argument, well, this is a just-so model. There are many different ones, in favour of different policies, of course. But, realistically speaking, constant growth of a developed economy was never possible. However well you reduce inequality or deal with automation-related job loss, there are limits to attainable market expansion. Stagnation or, at best, negligible growth are something we need to learn to face.

At the risk of sounding banal, I believe it's worth checking to see if you have a severe depression. Because that's how depression feels. There are people out there who derive great pleasure from writing uninspired code. Not only doesn't work need to be the main source of meaning in your live (in addition to providing income), its perceived value is heavily dependent on your psychological and physiologial condition. In fact, I'm afraid that, thae way you are now, you might not be able to enjoy yourself even if your situation changed for the better.

This is a bleak message, yet it's true: most people do not like their jobs very much. They can just deal with it because they're not totally depressed. And the reason is, surprisingly often, simple brain chemistry. At least try some more roads before talking of suicide.

Hard to put into words how reassuring it is to see so many sane, grounded and critical reactions (both here and on hackernoon) instead of gullible hackneyed accolades that the author's tone implicitly demands. I feel like thanking everyone for calling him out, but this is a bit tasteless.

I wonder, though, if Sam himself believes his narrative. Startup CEOs tend to be borderline delusional in their optimism, one could even argue this is a job requirement. Still. Is there a subtler interpretation, such as this being a smart cover-up for personal irresponsible enrichment, or does he really think they did their best?

This is wrong. It's not needed because people are getting older, it's needed because the increase in productivity is not enough to cover expenses for long-lived retirees and other groups. If Sweden was 20% more productive, perhaps they could afford to just have longer and happier retirements. And if it was, say, 50% more productive, they could cut work week to 15 hours or decrease retirement age. Alternatively, they could cut other expenses (for example, stop bringing in more refugees) and still afford fixed retirement age. This "simple math" is not inherently linked to longevity of the Swedish elderly, it's the budget constraints that matter.

Edit: given that bringing up refugees invites downvoting, I want to clarify that this may not be among the major causes of Sweden's decision to raise retirement age (and is not the core idea of my message), but it's not negligible either: according to this guy http://voxeu.org/article/fiscal-cost-refugees-europe , current immigration policy costs Sweden 1,35% of annual GDP.

Ironically, the article doesn't really discuss any legitimate reason as to why Sweden is – or indeed will be – fine despite automation. It's all about what Swedes expect and how much they trust their present society to hold. Also, labor union propaganda, but that's tangential. So, the takeaway is that Americans believe business to be cutthroat and competition to be effective at weeding out the least optimized actors, whereas Swedes feel that employers will take care of employees because this is a nice Sweden thing to do, or something. Beliefs, expectations, feelings, self-affirmations. «Everything is fine, we can provide everyone with jobs, we can even support a large and growing number of uneducated refugees and help them get jobs, we just need to innovate faster than anyone to be sure that our market share won't ever diminish, it'll all work out in the end". Well. Guess it's pleasant to be so positive. But how is any of this even relevant to the issue? The issue being, mind you, automation-induced joblessness, not people's anxiety about it.

It's funny how nowadays (for a few iterations, actually) Apple's phones have bottom of the barrel battery life, even facing severe degradation in a few months, yet the reputation from 4S and earlier lives on. I know many people who still believe that Android phones suffer from faster discharges.

Life is not a playing field full of fun and games, even though you could frame it in terms of game theory. People want to be successful, not to evaluate their "innate" performance against other people. Moral justification for doping is unimportant, because it's arbitrary. In reality, however many competing actors make me miserably insecure by their superiority, whatever is their percentile relative to the general sample, I wish to win, I wish be more than what I am all the same.

Please read the article.

There's also another, intrapersonal scale at play: if you get 20 "free" subscriptions, in 3-4 cases you'll definitely forget to unsubscribe before it starts charging, since it becomes impossible to track and optimize every process. Attentional impact of each one is miniscule, so many people don't notice going way above 2% or any other hard limit they could probably deal with semi-efficiently.

The whole situation with attention economics is very illustrative of the neglected power of small things. Facebook can't really hack your dopaminergic pathways, however much they improve their feed AI. AdSense doesn't brainwash you to be a consumerist drone. Smartphones aren't exactly subtle with notifications, and it's easy to disable them. There's no violent, obvious infingement of one's agency, so the opinion that you still make your free choices is not without reason. But more often than not, people don't actively counteract these influences. They go with the flow, they allow the scales to be tipped a little. The more they do so, the more suspectible to distractions they become: that is, less prone to consciously deciding if the next stimulus is worth responding do. So even the tiniest starting amount of influence, as long as it isn't annoying and doesn't raise any internal alarm, is a seed of a serious behavioral change. This is how vulnerabilities work: it's not always intuitive that some small hole can compromise an entire system which is otherwise robust and adaptive. But you don't argue that this is pathetic, you go and patch it before the damage is done. Hard to do with humans.

On a more practical (if unrelated) note, I think all browsers should block website notifications by default, i.e. when you get a prompt about push notifications, Deny button has to be highlighted. And it shouldn't be ever shown on mobile.

Sometimes beautiful hardware design, questionable performance (in Mediatek models), great audio, poor battery life and camera, their OS is full of neat gimmicks. Meizu isn't on top of their game lately, their best phones were perhaps MX3-4, now they seem a bit pricey for the offered specs compared to Xiaomi.

Incidentally I have their headphones (HD50), very satisfied (except faux leather is tearing too easily).

It makes no sense to discuss objectification in regards to the level of societal development that is implied by polygyny. Nobody there is seen as an individual being in a sense that no one's human rights are guaranteed or even promised to be. This is only about basic needs. Of course, the men at the top are winners, and winners take all. Women are inequally treated by them but get their basic needs covered (safety and resources, reproduction and opportunity to care for their children etc). Majority of men, though, simultaneously have to fend for themselves AND are deprived of sex. They are worse off, biologically and personally. They have less to lose, too: no offspring, no stability, no relationship, however inequal; which makes it so much easier to challenge the status-quo. Moreover, men are more likely to respond to violence in just about any context, especially this. No wonder that their frustration is a greater factor in societal disturbance induced by polygyny.

What is this "rest"? Are you really convinced this is a correct generalization? What about Mi 5, Mi Mix, 4C/A/X, can you show what they copied? Their first well-known phones barely had any design to speak of, and the same could be said for modern Redmi Notes etc.

Loose sim cards, really? I've had experience with 6 Xiaomi phones to date and never noticed that. Shaky GPS? Yes, this happened on one bottom-line Redmi, but it's hardly pervasive as far as I can tell. Software is debatable – it's not Pixel experience, I'll concede, but also definitely not TouchWiz. Battery is on average better than for most Chinese brands or for Apple, though inferior to, say, LG and Samsung. (One actual issue: camera performance is varying, typically it's bad without customization).

I won't argue that at least some models are clearly very closely copying a major brand (for instance, MiA1 is openly compared to iPhone, Mi Note 2 has Samsung S6 design), but this does not appear to be a crucial part of their business strategy (unlike the more abstract "Chinese Apple/cheap Samsung" sales pitch), and comments like yours strike me as repating a vague stereotype. In the post-iPhone era, striking design originality is rare to find, and Xiaomi doesn't seem more prone to copying than most others. They jump on the bandwagon as soon as the new technology becomes available/popular, and that's about it.

Xiaomi dynamics are about as unpredictive of its future success as a given cryptocurrency's exchange rate. At least there's less buzz here.

Xiaomi has a reasonable average quality and a few consistently interesting product lines (I myself enjoy using Mi 5s and will probably change it for Mi 7), and a ton of... strange... gimmicks. When they were being enthusiastically buried by various news outlets, I didn't consider it a real thing, neither do I put any stock in this renaissance. I would be a little surprised if they go bankrupt in 5 years, somewhat less so if they displace Huawei. They already have earned a certain place in market.

Prior to AZ, some competent people had doubts if self-taught NN could compete at all with a mature chess engine at its strongest setting. You know how that went. The improvement rate of subsequent Alpha revisions is incomparable to what software like chess engines shows. Hence, the tendency to evaluate each one as a breakthrough. Personally I believe that "breakthrough" is a non-technical word that's hard to define, but "improvement rate surpassing most optimistic expectations for a given field at a given time" could probably qualify as a criteria.

It could perhaps be said, then, that the Alpha series as a whole is a rolling breakthrough? Each new generation introduces changes that do not seem major compared to the ideas in the foundation of Deep Learning, but this actually allows to surpass state-of-the-art, improving performance on all relevant parameters (and not by a few percent). Dismissing these updates as technicalities is a dubious position.