Monday, November 19, 2012

Why the Shangri La Diet Cannot Work as Supposed


A man should treasure his doubts:

If consuming tasteless calories lowers your set point, then anyone who loses their sense of smell should lose weight like crazy.

I do not know, but I think that this is not true.


The Shangri La Diet book was a best-seller. Many people must have tried it, even though it sounds crazy and I can quite see why people would not go around admitting that they were trying it. But if it works for most people, then surely people who tried it and had it worked would tell other people. People are usually happy to go around spouting the worst craziness once they are convinced it's true.

So it should have spread by word of mouth and taken over the planet by now.

More arguments please. We can be like Greek philosophers, finding the truth without troubling ourselves to ask reality.

The Master (Film)


Absolutely riveting. Joaquin Phoenix, Philip Seymour Hoffman, Amy Adams all give performances as clear and simple as the truth.

I like Lancaster Dodd and his goodness and folksy charisma. I wish his Cause well. I like his psychotic sidekick and his scary kind wife. I think their epistemology needs work.

I have no idea what this film is about. What it is trying to say. Why it was made. I would like to know. I would like someone to explain it to me. I wonder who is master.

Sunday, November 18, 2012

The Shangri-La Diet Needs Our Help


A while ago, Eliezer Yudkowsky, a man whose thinking and writing I very greatly respect,
mentioned a crazy-sounding diet in a positive light. He said effectively: "It sounds like it should work, but it doesn't work for me." (http://lesswrong.com/lw/a6/the_unfinished_mystery_of_the_shangrila_diet/)

I decided that I'd try it out on myself. And to my surprise, it does seem to work.

It seems that Seth Roberts has found a hack into the process that controls how much we eat, and (he says) that it works for most people. I am reasonably confident that it works for me. There are a fair number of experiments on rats that show that the various mechanisms work in rats.

If so, then Dr Roberts has found a cure for one of the great scourges of life in the first world.

Obesity is a great problem for many people, and made worse by the incredible lack of sympathy shown by people who never have the slightest trouble with their weight.

But anyone who believes that anecdotal evidence is more than a suggestion of a good place to look, will assuredly end up going to aromatherapy sessions to cure whatever problems they have. And that may even work.

So it seems that we need to do a randomized controlled experiment to find out whether this hack works, and also from a less practical but more curiosity-satisfying point of view, whether it works for the reasons Dr Roberts says it does or not.

I have been swapping e-mails with Seth Roberts, and he reports that there is barely any interest in academic circles in testing his theory.

I quote with permission:
I disagree that a trial would be dead easy. Let me explain some of the difficulties. 1. Who would do it? Not any grad student: after you get a Ph.D. associated with a quack diet, who would hire you? The only people I could hire would know zero about this sort of research, which is very rare. 2. Where? 3. What comparison group? Nothing, placebo control, or best alternative treatment? 4. If placebo control what would it be? 5. Enormously difficult to measure compliance and equate compliance between the two groups. 6. There are maybe 10 labs in the whole world experienced in doing this sort of expt. Unclear why they would want to do this. I would need a lot of money to convince them.

This is clearly insane. There is a very great benefit to be had if this is true, so the fact that it looks a priori rather unlikely is no reason not to test it, if the test is cheap by comparison to the obesity crisis.

But I have also been talking to academic friends, and they find it all too plausible that what Dr Roberts says about nobody being interested in testing his theory is true.

From which we can conclude that academia is very badly broken.

So we need to do it ourselves.

I would like to come up with a way of testing this, that will not cost me very much money, and yet will provide some more reliable evidence than 'try it yourself and see if it works'.

But of course, I have absolutely no idea how to go about this.

My initial intuition is that we should make a vast number of little gelatin capsules, some containing 15g of extra-light olive oil, and some identical-looking ones containing something harmless with no calories in it.

And we should get around twenty volunteers to promise to take one first thing every morning, and not to eat or drink anything for an hour afterwards, and to record every day their weight at the time they got up, and whether they felt hungry or not during the day, and to keep this up for a couple of months.

We should send each volunteer a selection of these capsules, with a pattern like 'one week of olive oil, then one week of placebo, then one week of olive oil, then one week of placebo'.

And each volunteer should get a different pattern, which pattern is recorded secretly by me.

If Seth's theory is correct, then the secret patterns should be reflected back clearly in the data, with perhaps some sort of time-lag.

I have no idea whether this is a sensible sort of experiment to run, or how to analyse the data. I have just made all this up.

And I wonder if anyone else can make suggestions, or explain why this sort of experiment is so badly flawed that we will learn nothing useful from it.

One suggestion that has been made is that there should be a third group, or a third part of the random patterns, where no capsules are taken, but the same data is recorded.

That's probably a good idea, but of course it makes the trial half again as much work. So how do I tell whether the extra information from that is worth the extra effort?

Anyone know how to do this sort of thing?

Feel the Spirit (Wolfson College Choir)


A lovely hour of accompanied solos, instrumentals, and songs from the full choir. The music was uniformly good, although sometimes too slow.

Lee Hall was far too bright, and the performers in the harsh light too obviously the postgraduates and fellowship of Wolfson College. Something should be done to dim the lights for future concerts, but with eyes closed the sound of the choir was bright and loud, with several fine voices over a skillfully interwoven background. If I had a recording of this I would listen to it often. I wished it longer.

The choir of an elite institution of the race responsible singing the songs of slaves in crystal accents could have been incongruous and off-putting, but it was touching to think that the descendants of the perpetrators remember the horror and sympathize with the victims. Has that ever been true of an atrocity before?

It made me imagine the scholars of a future society remembering this doomed generation, the first that will die, knows it is going to die, and stares that fact in the face unflinching.

I hope they sing our songs.

The Paradox of the Envelopes

Oswald Mega, the always welcome, has two envelopes and some money. He divides the money into three equal stacks.

He puts one of the stacks in one envelope, and the rest of the money into the other. He seals the envelopes, takes them to the usual watering hole, describes what he has done, swaps them around behind his back a few times, and then he hands you one of them.

The envelope contains £1000.

"Would you like to swap envelopes?" he says.

The Paradox of the Swans


A wise man once observed that "All swans are white." is true if and only if "Things that are not white are not swans".

So one day, you see a green mini. That is a thing that is not white, and lo, it is not a swan.

So the green mini is evidence that things that are not white are not swans, and so it is evidence that all swans are white.

If you do not buy this argument, why do you believe that swans are white?

The Paradox of the Children




Suppose you meet a mathematician in the street. Around here that is not such a far fetched thing.

And it comes to pass that she mentions that she has two children.

"Are either of the children boys?" you ask.

Yes, she says, with a cryptic smile, and speaks no further word.

And later, you meet another friend, who is a betting man. And you tell him your story.

He says "I bet you £50 that the other child is a boy."

Should you take the bet?


Consider: Had you asked "Is the older child a boy?", then the bet would be fair.
And had you asked "Is the younger child a boy?", then it would be fair.

And the boy must be either the older or the younger.

Related: How likely is the king to have a sister?

Music in Cambridge


Someone's gone to the trouble of collating the various sources of information about local concerts. There is some variety of classical music going on in Cambridge every day.

http://www.cambridgeconcerts.com/

The best bit is they make the data available using google calendar as well, so if you've got a google account you can see all the dates and times in your diary.

It looks wonderful so far.

Monday, November 12, 2012

Shangri-La Diet, Month Four, Success, I'm Convinced


I declare victory for the Shangri-La diet. My belt now prefers to be on its fourth notch more often than on its third, and I think that counts as a win. My appetite still seems suppressed.

Updating for (down a notch/appetite loss)
Priors H87 : W2 : S11
Likelihoods: H 25/203: W 5/134: S 75/134
Posterior: H63: W 1: S36

My original prior was H60:W39:S1, so what's happened over these four months is that the fad diet looks more likely by a couple of orders of magnitude while the conventional wisdom has proved to be utterly wrong.

According to my analysis, I'm now licensed to believe that the theory "Shangri-La Diet causes belt size to contract by one notch a month" has a half chance of being true.

But actually, of course, what's happened is that after an initial quite dramatic loss of body fat, I've been slowly losing weight for four months, and my belt measurement is now about 3 inches less than it was.

At this point, I think I can claim that this crazy diet works for me exactly as it's supposed to. Slow weight loss without any use of willpower.

And I'm also claiming that the prediction almost everyone made that if I deliberately ate an extra 300 calories every day I'd get fatter is just wrong. Badly, hopelessly wrong.

That's the end of my experiment, because my belt won't come in any more. Its diameter is now fixed by the size of my hipbones. All the clothes that I found I couldn't wear four months ago fit again, and I feel lighter on my feet.

I've still got a fair bit of abdominal fat that I'd like rid of, so I'm going to keep on with it, but I don't think my belt measurement will go down any more. I predict that it won't go back up as long as I keep up with the olive oil. Even over Christmas, where my Mum usually manages to leave so much lovely food lying around that I become visibly fatter over the fortnight.

The formula that seems to have worked for me is:

Drink 300 calories worth of extra-light olive oil first thing every morning and immediately wash the taste away with water. Let nothing except plain water pass your lips for one hour after that. Apart from that, eat whatever you like, whenever, for whatever reason. It doesn't matter in the slightest if you miss the odd day.

According to Seth Roberts himself, this diet works for lots of people but by no means all.

I am convinced that it works for me.

I recommend you try it if you're overweight and would like not to be. If it works for you, you'll know within a month. I'd also recommend you read Seth Robert's book 'The Shangri-La Diet', for his explanation of why he thinks it should work, which is helpful if it stops working and you need to work out why and what you can do to fix it.

If you're of a technical bent then http://sethroberts.net/science/ is interesting as it reviews his theory of why his diet works, and lists lots of published papers that demonstrate various components of his theory. To be honest, I don't quite buy the mechanism proposed, but it's really not my field so I'm not competent to judge. On the other hand, it's definitely much closer to real science than most psychobabble and self-help. I'm pretty sure the guy's sincere.

My only worry here is that it does have to be Extra Light Olive Oil, not the tasty Extra Virgin sort. ELOO is produced by a nasty manufacturing process that pulls out the last drops of oil from olives that have already been squeezed, and it's not much fun to eat as it's nearly tasteless. However the consensus opinion seems to be that it's fairly healthy stuff.

Someone needs to do a proper randomized controlled trial on this.

D K Till, Cambridge Coal Merchant, 01223 232947

Getting a lorry onto Midsummer Common is an arcane and exotic process, which changes every year at the inscrutable will of Cambridge Council.

I'd like to recommend D K Till, our local coal merchant, who has managed this feat three times in order to deliver 15 bags / £150 worth of coal directly to the roof of my narrowboat. He even lifts it up there, which considering I once put my back out throwing 25kg coal bags around is a very fine thing for him to do.

They're a very friendly family firm, father and son-in-law. All their smokeless coal is £10/25kg bag and they sell Taybrite and Supertherm, the most popular brands for boat stoves, as well as several other brands. It's hard work to get it any cheaper than that round here. Their coal is fine quality and comes in their own heavy waterproof bags.

I'm told, although I have not yet experienced this for myself, that they also sell wood, kindling, and gas bottles.

They have no web presence, so I thought I'd try to give them one with this post. Their phone number is 01223 232947 and I recommend them without reservation.

Tuesday, November 6, 2012

Personality Metrics (November 2012)


I've spent the day doing web-based personality tests. There is no subject more fascinating than oneself. Anyone who isn't me may want to skip this post.


The first one was for political opinions. It's at:

http://www.politicalcompass.org/test

and it tries to classify your politics along two separate axes. I scored:

Economic Left/Right: -2.62
Social Libertarian/Authoritarian: -2.97

Apparently this makes me a libertarian lefty, a political ally of Ghandi and the Green Party.

The libertarian bit seems fair. Fuck off government.

But the left-wing bit strikes me as a bit unlikely. I do often vote Green, but that's because I love trees and hate cars, and want to send that message to the larger parties. I think the Greens' economics is loony, and I'd cheerfully die in order to prevent them getting their hands on power.

My ideal government would have been be the old Liberal party, before it merged with the lefty Social Democrats.

But out of Labour and the Tories, I feel happier when the Tories are in power. They seem to mess up less and interfere in my life less. The current Lib-Dem/Tory compromise strikes me as about right economically.

But the political compass has me as an anarcho-syndicalist, with all the major British parties well to the right of me economically, and way more authoritarian than me. One of my beliefs must be wrong. I wonder which it is?


Next one was the so-called Big Five test:

http://www.outofservice.com/bigfive/

Apparently this is taken quite seriously by psychologists, and when I first heard about how it was designed, I thought "Yes, that's the way I would have liked to design a personality test".

I'm Open 95% Conscientious 41% Extravert 91% Agreeable 17% and Neurotic 4%

So, crazy open to new ideas, slightly less conscientious than average, very extrovert, very calm in a crisis, and rather disagreeable and rude.

All that seems terribly plausible. However I took the same test 6 months ago, and it came out:

Open 96% Conscientious 46% Extravert 93% Agreeable 79% Neurotic 11%

So it looks as though four of these traits are stable, but sometimes I'm very rude and disagreeable and sometimes I'm very tactful and nice. Or more accurately, since this is measuring my own answers to various questions, sometimes I think I'm nice and sometimes I think I'm nasty.

I'm quite impressed by the stability of the test over six months on four of its axes, but really, what sort of measurement is it that can go from 17% to 79% on the exact same questions answered six months apart by the same person?


An IQ test next (I wanted a second opinion):

http://www.iqtest.com/

On which I scored 148. That sounds about right, I always used to get somewhere between 150-170 on IQ tests. The scores are a bit variable at the ends of the scale. And I'm aging and have probably poisoned my brain a bit with alcohol over the years, so it's no surprise if I'm in a bit of a decline. Also there's the Flynn effect, which means that modern IQ tests have to be harder to make up for the fact that people are getting better at IQ tests. No one knows what is going on here, although I'm inclined to believe that people really are just getting cleverer. This means that an IQ of 165 in 1980 is about the same as an IQ of 150 in 2010.

The weird thing here is that I'm pretty sure that I got all the answers right, so I cheated and did the test again. The questions were the same, so I was able to do them a lot faster. On the second pass my score was 151, which isn't much of an increase. On the third pass I had all the answers memorised, and so I filled them in about as fast as it's possible for a human to do. That was worth 152.

I don't see how it could be done better, so unless I've got an answer wrong, this test probably tops out at 152, and speed seems to be worth very little.



And a second IQ test:

http://www.free-iqtest.net

On which I got 147. Again, I think I got every question right, so I think I'm just topping out the test, like a fat bloke finding that his scales hit the bar.


Then there's the Myers-Briggs personality score, widely used in industrial personality profiling.

http://www.humanmetrics.com/

ENTJ
Extravert(56%)  iNtuitive(75%)  Thinking(12%)  Judging(1%)

So as far as I can tell from reading the explanation on the site, the Myers-Briggs type indicator is just some incoherent rubbish thing mainly designed so as not to offend anyone.

It does seem nice and stable though. About six months ago, I took this test and got ENFJ with strengths for each factor of 67/50/12/1. That evening, I tried the test again and got ENFP 22/38/25/11, and now apparently I'm: ENTJ 56/75/12/1

So it looks like I'm pretty normal on the last two factors, but am strongly extroverted and prone to abstract theorizing.

I think that's probably what you'd conclude if you talked to me for five minutes.



Next, the question that everyone who likes numbers must ask themselves: Am I Autistic?

http://www.wired.com/wired/archive/9.12/aqtest.html

Score 13 ( A bit less autistic that the average person)

So happily not. And this despite having spent the day doing personality tests on the web.

Sunday, November 4, 2012

What is going on here?

Spoiler warning: This post contains a picture of one of the Raven's matrices from the IQ test suggested in the Less Wrong reader survey.

There's rather a nice little IQ test app here:

http://iqtest.dk  (If you want to go and take it, do so before you look at the picture below!)

I scored 130 on it, which makes me think that the leetle grey cells may be falling apart under the influence of alcohol and old age. In youth, I usually got between 150 and 160 on these things (they're not terribly accurate near the ends of their scales).

Anyway, my powers are obviously declining. What's worse is that even when not under time pressure, I can't figure out what the answers to the questions are. Can anyone figure out which of the eight possible choices here completes the figure? And explain why? It's completely opaque to me.



If anyone has a go at this, the questions I can't do are 26, 29, 33, 34, 35, 36, 37, 38, 39. And I'd be really grateful for an explanation of what any of those are about.


Aha! http://www.jperla.com/blog/post/how-to-ace-an-iq-test gives an explanation of how these things work. After reading it I went back and can now complete the test in about 10 minutes (which gives me a score of 145+, apparently that's as good as you can do.). Most of the answers are indeed obvious in retrospect, although there are a couple where I can't decide between two choices, and one where I'm a little sceptical I could have worked out the pattern having never seen it before.

The particular one above works by adding the dots. dots outside the figure cancel with dots inside. As far as I can see, the dot positions are irrelevant and random. In terms of crossword clues, this would be OK in the Telegraph but not meet the standard for the Times.

In fact, I've now got some thoughts about the analogy between IQ tests and cryptic crosswords. They seem closely related in terms of what they're measuring, although obviously these visual patterns are language neutral, whereas I can't see how you could hope to do a cryptic in English unless you were a native-standard speaker. I might try to work this up into a blog post later.

Followers