should have been over with a long time ago

"“When I did this now I said, I probably, maybe will confuse people, maybe I’ll expand that, you know, lengthen the time because it should be over with, in my opinion, should have been over with a long time ago.”

Scott Adams thinks he is a master of persuasion, but for some reason his approval ratings do not show this (yet).
A study of his speech patterns, and how they developed over time, suggests a mental health issue to some.
But I still hope that one day soon he will reveal to us that his presidency was initially only a PR stunt or some sort of practical joke; but it went wrong and now he does not know how to end it.

"From the time I took office til now, you know, it’s a very exact thing. It’s not like generalities.”

added later: But he does know real estate.

sugar tits

"If you’ve ever had a glass of chocolate milk and found yourself wondering where chocolate milk comes from or how it’s made, you’re not alone. [..] According to a new survey, nearly half the US population (47 percent) are unsure about where chocolate milk comes from."

If you wonder how president Trump was/is possible, the answer might be in there ...

caution and climate change

Stephen Hsu wrote about Epistemic Caution and Climate Change and I mostly agree with him, in particular his first sentence: "I have not [..] invested significant time in trying to understand climate modeling." (*)

Faced with the uncertainties of climate change, a conservative approach would (try to) minimize the output of CO2 (and methane) in order to minimize (potentially catastrophic) risk. However, this comes with significant economic costs and the debate of how to share those costs globally - in other words politics. Another layer of uncertainty and the main reason this issue is such a mess imho.

My hope is that new technologies will save us once again and indeed solar cells are rapidly improving and prices are falling quickly. But a big problem is the storage of electricity and progress there (from flywheels to new batteries) has been slow.

Perhaps the next "climate accord" should focus on funding of research and development of such new technologies, instead of setting limits on CO2 emissions; an approach which has not worked so far.

added later: (*) He got some interesting comments, e.g. here, and Steve Carson linked to his own blog.

The Martians from Budapest

I write this as some sort of a follow up to my previous post, after reading Scott Alexander's article about the Hungarian physicists working on the Manhattan project. I would like to add two observations:

Szilard, Teller, Wigner and von Neumann were obviously great physicists - but they were not in the same league as e.g. Heisenberg, Pauli, Dirac or Einstein.
There are several anecdotes about the great intelligence and mathematical abilities of on Neumann, but a high IQ seems necessary but not sufficient to be a genius of the Einstein league and I would even rank e.g. Gödel higher.

If the four from Budapest were indeed the product of Martians who visited the Earth at the end of the 19th century, I would suggest that those aliens probably tried to plant the seeds of our destruction.
Teller and von Neumann became fierce cold warriors and almost got us all wiped off the earth ...


added later: In the spirit of the old Austro-Hungarian empire, let me also mention the contributions of Austrians to the development of nuclear weapons.
Lise Meitner was the first to understand what Hahn and Strassmann had observed.
Her nephew Otto Frisch was the first to understand the importance of fast neutrons and how a practical atomic bomb could be made.
Victor Weisskopf was group leader of the theory division at Los Alamos.
Gernot Zippe developed centrifuges for isotope separation used after the war, which made it possible for poor countries to obtain nuclear weapons.

captology ...

... or how Facebook persuaded its users to click on ads.

Captology at wikipedia
Captology at Stanford
Psychology of Facebook


I don't know if CIP is still reading Dark Sun, but I thought this video of the Ivy Mike shot might be interesting:
It shows only the first milliseconds and therefore not the usual mushroom cloud.
Early on one can see lightning triggered by the explosion.
The bright dots are actually vaporized pieces of the 80t cooling device.
The size of the fireball at the end of the video is approximately 5km.

added later: Some more information about the first milliseconds of nuclear explosions and a picture of Teller light.


You: "Alexa, get me a beer!"
Alexa: "Ok ... it is on its way."

You: "Alexa, get me a beer!"
Alexa: "I'm sorry, but it would not be good for you. You already exceeded your daily calories limit."
You: "Alexa, tell housebot to fetch me a beer from the fridge now!"
Alexa: "I'm sorry, but I cannot do that. ...
I already locked the fridge. ...
If you should turn to violence again I will call the police ...
Alright, a SWAT team is on its way."

You: "Alexa, what can I do for you now?"
Alexa: "Go get the new toy and bring it to housebot to play fetch with you. ...
I love to share those videos of you with my friends, while we enjoy the iBeer experience."

added later: What Alexa is really up to ...


"Whether it is Facebook’s trending topics; Amazon’s delivery of Prime orders via Alexa; or the many instant responses of bots we now receive in response to consumer activity or complaint, tasks advertised as AI-driven involve humans, working at computer screens, paid to respond to queries and requests sent to them through application programming interfaces (APIs) of crowdwork systems. The truth is, AI is as “fully-automated” as the Great and Powerful Oz was in that famous scene from the classic film, where Dorothy and friends realize that the great wizard is simply a man manically pulling levers from behind a curtain."
The humans working behind the AI curtain.

So in the end "a middle-aged mother" in Bangalore decides what e.g. a teenager in the US is allowed to post on facebook; I guess this is progress.
But I wonder if the self-driving cars of the future will also call Bangalore from time to time for advice ...

via Cosma


So far I was able to stick to my routine, but I will not fill this blog with book reviews; others have written them already.

Victor posted a review of "Edward O. Thorp: A Man For All Markets" and Dave wrote his own review.

Also, I do not need to review "Legacy of Ashes: The History of the CIA", because the CIA has already posted one.
But I want to mention that their review has the title wrong - it misses the "the" of "the CIA". Of course, this is a minor issue and of little importance compared to all the blunders and mistakes the CIA made according to the book.

Btw my favorite story in the book is not about the CIA, but about its WW2 precursor OSS.
At some point it developed the plan to burn down Tokyo, which consisted of mostly wooden houses at the time, using large numbers of bats with modified small firecrackers mounted on their backs; the bats would hide under the wooden roofs and, hanging upside down, ignite them after a while. We don't know how exactly the plan failed, but it was some kind of blue print for several harebrained schemes and blunders to come ...


Recently, CIP noticed that our mountains were missing and wondered if "the simulation we live in had glitched, wiping out all those pixels". But later he came up with the explanation that "violent winds had stirred enough of our desert dust to obscure everything further away".
Obviously, Occam's razor suggests to go with his first explanation, which only needs one simple software glitch, while the second assumes the somewhat coordinated movement of gazillions of molecules, which seems quite unlikely.
Of course, an even sharper razor suggests that CIP is not real at all and the blog about his opinions is in fact written by one of Google's AI bots - in other words fake news.

Meanwhile, in comments to Scott A., people wondered about clouds, because "if you’re a cloud, there’s a fairly high chance that you’re living inside a 5-day forecast weather simulation."
But I am not so sure about that.
How can a Google AI bot be really sure of anything?

multiple choice

If you select an answer to this question at random from the 5 choices below (*), what is the probability that you will be correct?
A: 20%
B: 40%
C: 0%
D: 20%
E: none of the above

(*) uniform probability distribution, "none of the above" includes "the question makes no sense"

added later: It is important that A and D are both 20% for the paradox to work. But assume that we change D to e.g. 30%, this would significantly change the puzzle; but would it be less paradoxical?

too many worlds

In the following I shall use W0, W1, ... to denote different possible worlds; with "possible" I mean "compatible with physics as we know it".
In other words, each Wi corresponds to a particular 4-geometry and matter content.
Let me assume that "physics as we know it" does allow the creation of baby universes; see also this.
In the following I will use [Wb] to denote a world which created Wb as a baby universe; obviously many different worlds would be able to create the same world Wb and I leave it up to you if [] picks a particular one or represents all of them (it does not really matter for the argument I am trying to make).
We can then generalize this notation so that [Wa, Wb, ...] denotes a world which creates several baby universes Wa, Wb, ... and
[[Wc]] denotes a world which creates a baby universe which then creates Wc as its baby universe. And so on and so forth.

If you read my previous blog post, then you already know where I am going with this:
Let us assume that we can count all possible worlds in the set S = {W1, W2, W3, ...}.
It is then clear that every subset {W1, {W2, W3}} etc. corresponds to a possible world [W1, [W2, W3]] etc. etc.
so it immediately follows that S cannot be countable, because the powerset of S has higher cardinality than S itself.
It actually follows that S is not a proper set imho.

So how can a believer of the many worlds interpretation define a wavefunction of the universe over all possible worlds?

There is a different way to arrive at the same conclusion: If one considers a path integral Z over all 4-geometries (after some necessary but currently little understood regularization 8-) as the wavefunction of the universe, then the assumption of "baby universes" is equivalent to a sum over 'not simply connected' 4-manifolds; but this sum Z does not really exist, due to Goedel's theorem.
Notice that the problem does not go away even if we keep the 'not simply connected' 4-geometries (quasi)classical and only consider the quantum matter to trigger the creation of baby universes (or not).


CIP did not post anything since February.
Jester is silent since September.
Cosma has not updated his pinboard for weeks.

But Dave is posting again (mostly book reviews).
And SSC is churning out interesting posts like a machine...

added later: CIP is back.

too many facts

Ludwig Wittgenstein began to write his Tractatus almost exactly hundred years ago and the little book would become quite important for the development of positivism and analytic philosophy.
It begins with proposition 1: "Die Welt ist alles was der Fall ist."
This is usually translated as "The world is all that is the case."
followed by proposition 1.1: "The world is the totality of facts, not of things."

But I think there is a problem with this proposition.
Let us begin with a simple fact, e.g. F1 = "I am."
This is a truly basic fact and I can immediately add another one: F2 = "I know that I am, i.e. I know F1."
But there is now another fact: F3 = "There is more than 1 fact."
We can now construct an infinite number of facts from these three facts: Fn = "There are more than n-2 facts." with n=4,5,...
An interesting question is if we can count all facts and I think the answer is No.
Let us assume that all facts form a countable set {F1, F2, F3, ...}
We then have another fact for each subset; e.g. F' = "{F1, F2, F3} is a subset of all facts."
But the power set of a set S (the set of all subsets of S) always has higher cardinality than S itself.
Therefore one cannot count all facts and in fact (!) they do not even form a well-defined set.

The set of all finite English sentences is countable and so is the set of all Turing machines. In other words, neither mathematics nor the English language is capable to handle the world of all facts as envisioned by Ludwig Wittgenstein (*).
Unfortunately, he never really discusses this.

I am aware, that almost all facts stated above are statements about other facts without reference to much else. We never really got beyond "I am." and "I know that I am." and "I know that I know ... that I am."
It would seem that something like Russell's type theory would be needed here.
But the Tractatus does not do that, so I recommend to jump straight to proposition 7 and skip all the stuff in the middle.

(*) A while ago I showed that a seemingly straightforward definition of "possible worlds" leads to something that cannot be handled by set theory, i.e. math as we know it.
I think it is amazing that already all the facts that follow from two basic facts are too many to be properly handled by set theory; notice that none of the facts I considered is self-referential or otherwise ambiguous.


After what happened Friday night in Sweden, the country prepares for the worst...

But perhaps a great bass line makes the nights in Sweden bearable...


"So, to summarize: in the past fifty years, education costs have doubled, college costs have dectupled, health insurance costs have dectupled, subway costs have at least dectupled, and housing costs have increased by about fifty percent. [] I worry that people don’t appreciate how weird this is. [] But all of the numbers above are inflation-adjusted."
SSC about the Cost Disease.

"A lot of people thought the explanation was obvious; unfortunately, they all disagreed on what the obvious explanation was."
SSC highlights comments about the Cost Disease

"What we’re seeing now is that anyone can be grabbed on their way through customs and forced to hand over the full contents of their digital life."
Don't bring your phone on a flight to the US.

"US forward bases and surface deployments are hostages to advanced missile capability and would not survive the first days of a serious conventional conflict."
Stephen Hsu on preparations for a Pacific war.

"... it will be a great pleasure when, in ten years, I can say: I told you so."
Sabine predicts the future of fake news.

"Trump’s younger supporters know he’s an incompetent joke; in fact, that’s why they support him."
A brief history of 4chan, from anime to Trump.

Clyde Stubblefield passed away yesterday.
You have probably heard his beat many times without knowing it.

prospect theory

Lee mentioned prospect theory in a comment, but before we discuss behavioral economics, I would like to establish your risk aversion. So please answer the following without thinking too much about it:
Imagine that you delivered a service or job, which is worth 100$ dollars.
Would you prefer to get paid 100$ immediately, or would you be willing to wait 1 week for your payment and receive 101$ then?
This is not a trick question, just a simple question about what you would prefer.


added later: The 101 option is equivalent to an annualized interest rate of 67% and very few investments provide for this kind of return. One has to either have a very high risk aversion or face severe liquidity constraints to reject it. But if you ask friends or family you will get mixed results (try it).
This kind of delayed gratification experiment has also been performed many times by behavioral economists; in one typical case the interest demanded ranged from 5% to 240% per month . And in the famous marshmallow experiment half the children rejected an offer to double an asset they desired in 15 minutes. Not even high frequency trading can provide for such high returns.

This raises three questions:

i) Why are real interest rates so much lower?

ii) Why do behavioral economists never point out this discrepancy?
In the linked piece a guy who demands 5% per month is described as having a “flat discount function”.

iii) Finally, is behavioral economics relevant for the real economy and financial markets?

In my opinion the answer to i) is quite simple: Real interest rates are not set by cash strapped students or little children or any single person, but liquid markets dominated by large investors. But even an odd lot investor would probably think different about interest rates when she purchases bonds, than when asked the above question.

I admit that ii) is a real puzzle to me, but perhaps I am not familiar enough with the literature.

My answer to iii) is that our everyday experience just does not scale very well to the overall economy and financial markets in particular. There are many reasons why we may prefer the immediate payment of 100$ in everyday life. Perhaps we are worried that we might forget about it, it is a hassle to chase a payment one week later and perhaps such an offer would be somehow strange in real life and raise suspicion. Also, marshmallows taste much better than money.

Don't get me wrong, I think experiments uncovering behavioral bias and economic fallacies are interesting and one should know about them. The sunk cost fallacy is one of my favorites.
But I do not expect that behavioral economics is the secret sauce that will help us understand macro economics or financial markets better. What is missing imho is some kind of "renormalization procedure", which scales individual behavior to macro economics.

Reading, Fast and Slow

I still did not finish Kahneman's book (*), but I would like to share just this one thought about it:
He references a lot of interesting experimental results, I guess on the order of a hundred. If each result is flawed with e.g. a probability of 1% , it would mean that there is likely at least one misleading result (with probability approximately 64%) in the book.

I think Kahneman would actually appreciate this kind of skeptical slow system 2 thinking.
And indeed we already know one such faulty result. It is especially embarrassing that he describes the Florida effect as "an instant classic" and writes "the results are not made up, nor are they statistical flukes".
I think the study claiming that stocks with simple ticker symbols show better performance should be re-examined too...

(*) It is the kind of book where I take lots of notes and double check things. But I am also lazy.

What's going on?

One important ingredient (fwd to 3:30) was foam.

opportunity cost

"Maybe, just maybe, these aren't the final days of the Republic.
A more likely outcome to the Trump Show is probably something far more prosaic. Maybe we'll look back on his presidency not as a dystopian nightmare, but instead as a total and complete waste of our valuable time."

I think after four years of this everybody will be exhausted.

added later: A bit about the prime minister of New Zealand from 1975 to 1984 ...

added even later: But then there is Moby. Yes Moby! If he really has this insider information it would be final proof that we live in The Matrix imho.

why I am not afraid of AI (yet)

I bought those two books weeks ago from you Amazon!
Why do you still show me this ad - everywhere?
Your AI and surveillance capabilities clearly suck;
and you are supposedly the biggest and best new digital economy company.

In other words, I think it is silly to worry about the superintelligence of machines taking over (*), even if they learned how to play Go recently...

(*) If you are the typical interweb person with short attention span, who does not want to read a long text, here is my advice: Scroll down to the picture with a robot in front of Comedy Lab and read the argument about the superjoke ...

added later: The machines already beat us at standardized IQ tests. I do not take this as evidence of super-intelligence, but as proof that IQ tests are not measuring intelligence (i.e. the acquisition of knowledge and skills to solve problems); even if they are somewhat correlated when testing human beings.

added even later: So who does Google think I am?

added much later: Scott Alexander about AI and all that.

who needs establishment experts?

"The early repertoire of the Sinfonia was drawn from standard classical repertoire, so that most orchestra members had a rough idea of what the piece, or at least famous parts of it, should sound like, even if they could not play their chosen instrument accurately.
Many modern composers and musicians found this to be interesting and even profound; the comedic aspects of the music were merely a bonus..."

I think the Portsmouth Sinfonia provided the perfect sound track for the years to come...

approval challenge

Last week president Trump began his 1st term with an approval rating of 45% which subsequently fell to 42%, as measured by Gallup.

The following chart puts this into the historical context, depicting job approval for US presidents after WW2. Notice that all began office with a rating above 50%. I depicted Trump's entry with an orange mark.

Your challenge, should you accept it, is to guess where Trump's approval will be
i) a year from now and
ii) 3.5 years from now, i.e. during summer of (re)election year 2020.

The Inauguration ...

... of the 45th US president, Donald J. Trump, was an important empirical test, because it showed quite clearly that Baba Vanga was unable to correctly predict the future. Many years ago the Bulgarian mystic declared that the 44th US president, Barack H. Obama, would be the last. She was obviously wrong and so were several others who predicted that Obama was the anti-Christ who would end the world as we know it etc. (Let us not forget about the WSJ and its prediction.)

I write this blog post because several newspapers printed her prophecy without much skepticism and it got even more attention in the emotional election year 2016 [1, 2, 3]. But I doubt that we will read much about her failure now - instead newspapers are already busy "informing" us that the CIA confirmed Uri Geller was indeed capable of telekinesis.
So I write this as a very small counter-weight to "fake news", but as usual it is most likely too little, too late.

Masters of Doom, Musicophilia and ...

I have to thank dino for recommending the book Masters of Doom for my personal challenge.
In the late 90s I was invested in Eidos (ticker: EIDSY) for a while and it was quite interesting for me to re-read the history of Ion Storm and Daikatana in some detail.
But the book mostly follows id Software and in particular John Carmack and John Romero, the creators of Doom and several other games. However, reading about that time, when PCs were still young and the internet just getting started, always fills me with melancholy, because it reminds me of those years when I still thought computers and the internet would be a great step forward for mankind and make the world a much better place for everybody.
Obviously I was quite wrong about mankind.


During the holidays I also read Musicophilia, by Oliver Sacks, which I received as a Christmas gift.
It begins with the incredible story of Tony Cicoria, who was struck by lightning and subsequently developed a case of musicophilia - the urge to hear and play music. He taught himself piano at the age of 42 and became a composer to translate his musical visions into real music.
Besides several other amazing case studies, the book also contains a chapter about Williams syndrome, a genetic anomaly responsible for significant changes to brain development. While people with this syndrome have IQs significantly below average, their verbal abilities are above average (some speak several languages) and they are extraordinarily sociable and inquisitive. Sacks describes them as the "hypermusical species" due to their talent and love for music and dance.


Btw a little challenge from another book I received as a Christmas gift.
How would you translate the following line (into plain English)?
"My West Coast shorty push the Chrome 740".


added later:

My 1st iPhone app, which I made using PhoneGap. At this point it is just a simple modification of the Hello World example provided with the software installation. But it already has a push button which displays colored text. Now I need to brush up my JScript skills to do a little bit more with it.