Monday assorted links


Bitcoin and the death of the nation-state?

Really? Come on man.

It's a logical conclusion from that article.The points that apply to a firm also apply to a nation-state. Of course Garbage In leads to Garbage Out.

Basically the article is shallow with no deep thoughts or insight behind it.

#4 So that's what America has bexome: a diminished economic power, which has doreign policy implictions. How it happened?!

Free lunch economics aka voodoo economics aka Reaganomics.

Slash labor costs to create wealth to create jobs.

Stop paying workers to build assets driving scarcity inflation of asset prices creating wealth.

5. Rising debt and rising inequality are two sides of the same coin, as rising asset prices fueled by rising debt are what contribute to high levels of wealth inequality. My view is that excessive wealth inequality contributes to financial and economic instability; hence, anyone concerned about financial instability (in China, in the U.S., etc.) should also be worried about rising debt, rising asset prices, and rising wealth inequality. Of course, financial instability/financial crisis and the collapse of asset prices corrects the excessive inequality but at great cost to everyone not only the wealthy. I'm not sure how China can control this, as everyone wants to get in on rising asset prices even if prices are already very high (absolutely, relatively, etc.). The recent crackdown on Chinese billionaires may be the government's effort to discourage others from copying the high flying ways of the billionaires - if they are vulnerable to falling asset prices, then so is everyone else. It could be that China and the rest of us are trapped in a cycle of rising asset prices and rising wealth inequality followed by a financial crisis and then extraordinary efforts of central banks and governments to stop the collapse of asset prices and then to re-ignite rising asset prices. It could be worse: it could be 1929 et seq. all over again.

#2 It is interesting how much headway they've managed to make but the supposition of how these bit-currencies will change the economic landscape and replace real government backed fiat is ludicrous. There is a cosmic gulf between what these are now and what cheerleaders for bit-currencies what them to be. I don't foresee them going away but they will suffer a correction - likely serious - and then consolidation and then a serious confidence problem moving forward in perpetuity. Addtionally, should the fixed supply of these bit-currencies ever expand, forget it, they're no better than government fiat but not backed by the legal and coercive power of any government. Might as well be monopoly money.

#6 "A Brief History of Time" had an interesting thought experiment regarding the first few seconds of the universe. How a perfectly structured system balanced for heat, gravitation and containing no entropy was "kicked" by one particle that, for inexplicable reasons, set off a domino of effect of matter and gravitational conglomeration. Thus, all matter observable and existent in the universe is the result of the imperfection of one particle, and that gravitation and matter as we know them are an entropic force. In a perfect system - so ordered - none of this would be here.

Intelligent design maybe?

An intelligent design being would enjoy contemplating static perfection, forever.

But a mildly curious intelligent design being would say, "enough with the static perfection already! Hold my beer and let's see what introducing one little imperfect particle can do."

"There are more things in heaven and earth, Horatio, Than are dreamt of in your physics [philosophy]." - Hamlet (1.5.167-8), Hamlet to Horatio.

I want to see the physical/video evidence from the first moment. Some believe that God the Father Almighty created Heaven and Earth. That was before they discovered His (sorry girls) more widespread creations.

If you really think the Judeo-Christian God has an actual gender you really don't understand anything about It.

Maybe when the now-obligatory genuflections to She-the-Creator finally equal all the references to God the Father, we'll achieve that stasis again. In the interim the tedium will at least be perfectly static.

#7) I thought liberal, left-leaning economists already figured out how to make things better: "...societies are at a fundamental crossroads with one fork leading to political and economic chaos and the other way leading them back to being a prosperous part of the world economy....workers and entrepreneurs, not government, create prosperity....government's job is to create a framework for growth, namely stable prices, a convertible currency, private ownership and legally enforceable private contracts." These aren't the words of some libertarian right winger; they're the words of left-leaning development expert Jeffrey Sachs [].

I look forward to Stiglitz et al's coming wisdom about how to achieve stable prices, convertible currencies, private ownership, and legally enforceable private contracts. After all, these are the keys to prosperity and turning away from political and economic chaos.

You said 'Jeffrey Sachs'. Quit waving red capes in front of Barkley.

Wow, Art Deco, you find the weirdest things that you think I am going to get upset about. I do not particularly have anything against Sachs, although right after 1991 he did give some not very good advice to some governments in Eastern Europe, something he has since admitted although does not dwell on understandably.

The more interesting aspect of this story is not this quote from Sachs or anything like it, but the fact that the third Nobelist in that group Is not part of it, George Akerlof. My guess is that he is probably sympathetic to it, but has to lay low given that his wife is not only the current Fed Chair but actually still under consideration for reappointment, so he really needs to lay low.

Do you cuck much Barkley?

Actually, AD, I am not quite sure what this means for all you alt-rightos, which I know is a word you all really like to use a lot. Why don't you explain it to all of us, since you are such an expert on it.

Barkley, that comment (at 5:49) was from some poor soul who likes to boringly and sinfully (slander, lying) imitate others. Art Deco - the real poster - posts a thousand posts a year, and does not follow up on all of them. The poor soul who posted at 5:49 was not Art Deco, and, as a deceitful imposter, needs our prayers: my best guess is that he thinks, with his hateful little repetitions of words he thinks are insulting (poor little fellow has no idea of what really happens in this world), that he is being a good little minion to Vox Day or someone like that (and Vox Day should not be blamed, by the way: none of us should be responsible for the poor little souls who do sad nasty things to get our attention). God bless everyone who prays that those who are afflicted with a mental illness may someday recover. I, for one, hope that the phony Art Deco of 5:49 will be, 5 or 10 years from now, a man whom any father would be proud of. It will happen: that is my prayer: it will happen. And, 5:49, if you are reading: man up. Never say the word cuck again: you too can be a real man. Just never say the word cuck again: that is my advice to you. A week from now, if you stop saying that word, everyone - including your future wife - will have forgotten that you loved so much to say that word: a word you never understood, by the way. Remember - every once in a while someone gives you good advice. 5:49 - as a kind-hearted alpha male who wants you to some day be a real person: I ask you this one thing: never use the word cuck again. If you do, there will never be even one female in this world who will not suspect you of being a sad nasty little man: if you don't, you have the chance to know what it is to have a woman respect you. God bless you: you can do better in the future. I am happy to give you the last word on this insignificant little thread, deep in the basement of the internet: tell me I too an a cuck (an unimaginable lie, if you knew me well) tell me that you have won your "rhetorical" battle: I don't care: but be a man, be the man God wanted you to be.

For God's sake. man, stop sucking up to the people you imagine are happy to hear your stupid invective. They are not your friends: I, on the other hand, want you to be a real man. Alec Guiness once told some poor fanboy child that, for the love of God, he should never watch Star Wars again. That boy grew up to be someone who deserves respect (and he did not watch Star Wars again). 20 years from now, you will be thankful for this advice: don't use the word cuck as an insult: just don't. You demean yourself. You don't really think anyone respects your viciousness, do you? Well, there is hope for all of us. Be a man. Stop the passive aggressiveness. Find a woman who loves you, and treat her right.

And finally, you little fucking idiot who thinks he is cool because he repeats and repeats the word "cuck" - you - you - you are the cuck. You wasted my time. Again and again. But I am an indulgent alpha male who knows the women who love me laugh at losers like you - and I forgive, and wish the best for you (you remind me of one of my sons). We all know you will never steal a good looking woman from another man, you poor little fucking foul-mouthed idiot. But in the goodness of my heart, I want you to be better. Women need good men. You can be a good man one day. But only if you stop wasting people's time by saying "cuck cuck cuck" from your lonely sad little place. Believe me, my friend: it is easy to be a decent person. I forgive you, a thousand times over, for wasting my time, again and again, you sad little fuckwit , with your sad little battle to say cuck cuck cuck as much as you can. There is no way, my sad little friend, that any woman you have ever met would rather be with you than with me: but from my heart of hearts, I can say this: I want you to find a woman that loves you. You, too, are a human being: and that is important. And I, for my part, will - if you straighten up - never hint that you are the real cuck, because, when you straighten the fuck out, you won't be (the real cuck that, at this moment in time, you obviously are) (I am a leader of men and you are not. Just saying. Some day maybe you too will understand leadership.I want that to happen for you - believe me.) Hate on me as much as you want, you little fucking loser: but I have said what you need to hear. Leave Barkley, a good man, alone. Leave Tyler , another good man, alone. Repent your desire to suck up to Vox Day - he would be ashamed of you, by the way - and repent of your foul mouth. A week from now, or less - I pray for this - and God answers my prayers, you poor little idiot, whether you believe it or not - you will begin to understand what it is like to be a man who lives in a world where men do not slander and lie and gossip. God bless, you, my little friend, and you are welcome. Again, feel free to have the last word. But, whatever you say: you can move on from your foolishness to a better world, where you are a decent person whom people respect. Peace, brother.

Good luck, my friend.

Which do you dispute, that the world has any problems, or that anyone left of (American) center should try to help?

For reference, "the median American takes 4 conservative and 6 liberal views."

So you know, there is a chance that some of this stuff could be, or end up, mainstream.

I dispute that a self-selected group of economists has either the answers to any economic problems or the right to subject others to them.

That is not our system of government .. anywhere on earth.

The question is more about how to engage with expertise. "I reject the people even before I hear their ideas" strikes me as an overly defensive approach. But here we are.

2. "So if Bitcoin has demonstrated the positive characteristics of a firm along with all of the added benefits of a decentralized org, why isn’t the death of the firm obvious?"

Does it demonstrate all of the positive characteristics of a firm? The article mentions structure and stability as a benefit of the firm, but it doesn't seem to argue that Bitcoin provides these. One of the benefits of a decentralized org is its lack of structure.

Yeah, that article provides exactly one concrete, actually existing example of bitcoin facilitating productive work, and it's just a TaskRabbit competitor. That's not a bad thing, but it's certainly not a "new" thing. But if Bitcoin is going to enable new business models or obviate existing firms, that article doesn't contain any examples or notable explanations of why.

Every time I hear an explanation of this stuff, it sounds like a neat way to do a few kinds of financial transactions with fewer fees and all that revolutionary rhetoric sounds like lofty nonsense. Can someone actually describe a new business model Bitcoin enables?

Actually the bitcoin transaction fees are hidden and higher than in the traditional banking system.

Oh really? I thought I'd read it was cheaper to use bitcoin for transferring to certain countries with limited banking infrastructure. But I guess if it was, banks probably responded quickly enough so they could recapture most or all of that market.

Not so much with Bitcoin, but I think smart contracts (programs that run in public, are outside anyone's control, and own and control real assets that matter) are a genuinely new thing in the world. You can kinda-sorta do simple ones with Bitcoin, but later blockchains (notably ethereum) are designed for them.

The downside of smart contracts is almost exactly equal to the upside--there's no human intervention. That means it can operate very quickly and won't be swayed by social or legal pressure to change its behavior, but also that it will follow what's written in its code even when that's obviously not what the writer intended (as with various attacks that emptied smart contracts of all their money thanks to finding a coding error).

With enough eyes, all bugs are shallow. Unfortunately, for open source software, that doesn't guarantee anything, because there aren't all that many skilled eyes looking at the code in depth. Smart contracts where you can get real money by finding a flaw, on the other hand, create the incentive for a whole lot of eyes to find their bugs....

Yeah, I thought about mentioning them, but I'm dubious that that really opens up something new. If a contract provision can be decided objectively, it can already be handled through a quick and easy arbitration procedure if that was spelled out in the contract to begin with. But even something as simple as my lease with my landlord -- a very standard, highly-regulated contract -- would require a court for any of the substantial stuff that might actually come under dispute. Sure I can think of ways an automatic contract would be useful in certain activities, but as you note, there's a downside too, so I'm dubious that they will unlock major economic potential.

there's no reason a smart contract can't also include some human-based arbitration processes as part of its algorithm. The trusted third-party arbitrator would of course need to be selected by the contracting parties at the start.

In a way it has already happened. The high tech industry has structures itself into a service industry for purchasing innovations from smaller more agile firms. Sales and financing companies.

#7. I guess this is another group that will try to get Koch Brothers funding.

Nobel Laureates Stiglitz and Spence to lead/announce new group to make things better.

By mistaking social and cultural problems for economic ones and (without a doubt) prescribing solutions to the economic 'problems' which incorporate more employment in public agencies for people like Joseph Stiglitz. If he wants to locate an actual source of trouble, he might look in the frigging mirror.

#1 a rather strange way to assess AI/machine learning advances. People have a damned Siri on their pockets, amazon and similars optimize warehouse operation, nvidia produces hardware for autonomous driving, bioinformatics depends on it, automated trading........and the author worries about chess & go.

I think the author is right to focus on fundamental tasks and games. Any technology became radical when it surpassed human (or animal or nature) capability by orders of magnitude. At the moment I do not see Alexa or Siri systematically better than a human being. They are just more convenient interfaces than a smartphone screen to some people.

+1 The majority of current AI applications are pretty unimpressive. I don’t expect that to change any time soon.

Scroll through to get the crux:

"General AI systems, as far as I know, do not exist today, and the million-dollar question is whether they can be built with algorithms similar to those used today, or if there are further fundamental algorithmic advances that have yet to be discovered. So far, I think there is no empirical evidence from the world of deep learning to indicate that today’s deep learning algorithms are headed for general AI in the near future."

There are tons of people who expect (implicitly) general AI in the near future. You need that for empty-cab, or go-to-sleep, self-driving. But we are nowhere near.

Maybe that is why I worry about the performance of the group mind so much. Networked humans in social networks are not covering themselves in glory either.

In some ways computers make us smarter, in some ways not, and in some ways they make us more vulnerable.

#3.The interviewer in the podcast has the most extraordinary voice I have ever heard.

Oops, I meant #4.

Not a bad voice for this day and age. But radio and television listeners of the 1940s and 50s were exposed to many male voices of this quality and better, with a wide range of speed, resonance and what I call "spread": the ability to seemingly coat the whole face of the speaker with their vocal presence (obviously the audio engineer played a role in boosting spread). Most of those male old-time radio announcers were heavy smokers, so their voices had a warm, appealing buzz, possibly caused by the microscopic shredding of larynx tissue. These confident, energetic voices have mostly gone the way of the castrato.

#1. I'm impressed with AI progress, but also still a skeptic:

Today, many consider ImageNet solved—the error rate is incredibly low at around 2%. But that’s for classification, or identifying which object is in an image. This doesn’t mean an algorithm knows the properties of that object, where it comes from, what it’s used for, who made it, or how it interacts with its surroundings. In short, it doesn’t actually understand what it’s seeing. This is mirrored in speech recognition, and even in much of natural language processing. While our AI today is fantastic at knowing what things are, understanding these objects in the context of the world is next. How AI researchers will get there is still unclear.

Exactly. And I'm not convinced that the 2% error rate for objects is real either. Just for the hell of it, I uploaded an image to the Google Cloud Vision Api page:

It was an photo I thought would give the API problems and that turned out to be true. It's a photo from a back country cabin of a bulletin board with various postings over a set of shelves containing books, playing cards, board games and other objects. On the wall next to it are a thermometer and fire extinguisher. You can see part of a doorway with rocks outside and, in part of a covered window, you can see the reflection of mountains out windows behind. What did Google's image algorithm come up with? "Window, table, furniture, shelving, door, glass". Very basic. And there's no table. All in all, still massively behind human image recognition capabilities -- and that's just for recognizing all the objects in the scene, let alone understanding what it's about. There's *so* far still to go.

#1. This is from the section where A.I. learned to play Atari 2600 video games. These were the bottom five where A.I. performed the worst where it smashed humans in Pinball and Breakout and scored higher at Space Invaders.

Zaxxon 50%
Venture 30%
Ms. Pac Man 10%
Astroids 10%
Montezuma's Revenge 0%

This must have been an old ranking, and it turns out the Nature letter that described this was published in 2015. A year later, an algorithm was introduced whereby the A.I. blew past the best human players to get the first score of over 1,000,000 in Ms. Pacman and a "creativity" algorithm got the A.I. out of loops in Montezuma's Revenge to score a perfect 100%, up from not getting a point a year earlier.

The progress has been quite impressive. Alphago zero, in particular was quite the accomplishment.

Yes. But many people mistakenly project from success in game playing to expectations of success in other, much more difficult domains -- domains that don't have simple, well-established spaces, action rules, win/loss criteria, etc.

#3 The chances of them not screwing they up and this not becoming a white elephant/slum are almost zero.

+1. The idea of giving anything connected with Google the chance to plan part of a city gives me the heebie-jeebies. "...all of these things that we could do if someone would just give us a city and put us in charge." Yeah, let's put you in charge, since you are so much smarter than the rest of us. That's all that's required to fix everything: some smarter people. And here comes Utopia.

Maybe after this becomes a disaster it’ll shut up the Alex Tabarrock/Tech booster idiots for a while. But I doubt it.

Yeah. Speaking just for myself, I'm sure I'll call it a minor glitch on the road to utopia no matter what the outcome.

Of course you would you Zuckerberg Cuck.

#7 - Guess he fixed up Venezuela - what could go wrong ?

#6 Yet another grossly misleading science story headline which also illustrates the misguided arrogance of a lot of folks at CERN who think that if Nature doesn't confirm to their preconceptions of it that this is an "unsolved problem in Physics" instead of a hint about how Nature just happens to be.

#6. Link does not work.
But we already know we're living in a simulation, right?

#1 The comparison for the rate of increase between human and computer might not be fair. Human champions are known to be aversed to challenges and without those challenges the contestants are slow to advance. Thus there are max time limits that the champions can defer the challenges. Are the human wins and loses to computers counted in the official ELO ratings? If they are counted then it could be easy for people to manipulate the official ELO scores. If they are not counted plus a sizeable financial rewards most human will accept the challenge from computers and hence the rate of increase for computers could be higher.

Comments for this post are closed