The "Napoleon Dynamite" Problem

Yesterday I saw the movie Napoleon Dynamite. And what did I think of it? If you've read my blog you might think that you can figure out what I thought about the movie based on other movies I liked and did not like. But as it turns out Napoleon Dynamite is a very particular movie in this respect. It is particularly difficult to predict what people think of this movie, even when their movie tastes are well known. And that made me really curious, I just had to see this movie.

Netflix is an online company that lets you rent as many movies as you like against a monthly fee. For them, it's important with happy customers that wants to keep renting movies. And one way for them to make their customers stay is to give them accurate recommendations. In fact, making accurate recommendations is big business for them. Netflix have had people working on improving their recommendation system for many years. But after some time they hit a well, they didn't make any improvements anymore. Or if they did, they were minuscule. But they still had their mind set on improving the recommendation system so they took drastic measures: in the spring 2007 they announced a competition. Any programmer who could improve upon Netflix own recommendation system by 10% would win the competition. The prize was set to one million dollar.

Given that kind of prize money this has naturally attracted a whole lot of programmers world wide. And initially they were making good progress towards the 10% goal. They were making leaps of several percentage units coming up to seven, eight and now even over nine percents improvements. But as they crept closer to the goal it keeps getting harder to make new improvement. And as it turns out there is a small class of movies that turn out to be extra hard to predict. And the hardest one seems to be Napoleon Dynamite. This movie has some special quality that makes it singularly unpredictable. It is this movie and a small set of others that are fooling the algorithms to make anymore progress.

You can read about all this in more detail in this interesting article in New York Times.

So, what did I think about Napoleon Dynamite? It was not my kind of movie, I didn't like it at all. It was actually a pain to watch it until the end but I forced myself though it. But I can see why some people like it. The movie is set in an 80's style and the protagonist is an incredible looser. I would have maybe appreciated the retro style if I liked the rest of the manuscript but as it is now it didn't help the film a bit for me. The people in the movie are quite angry with each other and that's one thing I certainly don't appreciate in a movie, especially if it's over small stupid things. The only redeeming things for the movie is the intro and the fact that the ending is a happy one.



I've mentioned a couple of times that I use Wine to run applications on Windows on my Ubuntu Linux laptop. So it's high time I give it the praise it deserves: Wine, I love you! Well, in a platonic way anyway. I'm really impressed with what this application does and how well it achieves it. Sure, I've tried a whole number of Windows applications that it doesn't handle but it doesn't matter. For me it's more like for each program that actually run it's a bonus, a happy surprise. Today I tried one of my favorite games Happyland Adventures and it ran perfectly, although I had my screen set on such a high resolution that window of the game was a mere dot in the middle of my screen as it runs in what I believe to be 640x480.

Earlier today (well, technically yesterday) I also managed to find AppDB, a database which records how various Windows programs work under Wine. The interface is a little cluttered but apart from that this is an extremely helpful thing. Just go there, look up your favorite Windows program and see how it works for others. Even if it doesn't work right out of the box some people might have found out a little trick to make it work. Pure awesomeness.

For me Wine makes life on the Linux side quite a bit brighter.


Eye of the Tiger

I've just hear the most brilliant cover ever! I was browsing around for music in Spotify and came across a title I knew well,Eye Of The Tiger. But the artist was a big surprise: Paul Anka. He has made an album called Rock Swings which contains some surprising covers. And his version of Eye of the Tiger sounds nothing like the original, it's a big band swing version. The only thing left from the original is the lyrics, it has none of that pumping electric guitar. Yet I couldn't help laugh when I heard it, barely recognizing the original behind the cover.


Batman suing Batman

I've blogged about Batman before. No, not the guy in the cape living inside comic books and recently on the movie screen. No, it's the city in east Turkey. I find the name incredibly funny. But the fun doesn't stop there. It turns out the city mayor of Batman is now suing Warner Brothers for their use of the name Batman.
What can you say? The guy is obviously greedy. But it's also the case that the city has gained more reputation from the movies than it had ever could have managed for itself.
I read somewhere that the city has a bad reputation in Turkey for, among other things, the high rate of suicides among young women. So maybe they're looking for to improve their image. But suing Warner Brothers won't help I'm afraid.

Oh, and by the way. The Dark Knight is a really good movie. My DVD is arriving in two weeks.



Right now I'm sitting in my sofa and watching the second episode of CSI (Las Vegas that is, the original series) from this season. And no, I have not downloaded it. I think this is some kind of record for Swedish television. Never before have we been so close behind a show that's imported, at least as far as I know. One month and eight days. But I guess we will see more of this now that people are starting to download stuff from the internet. It's only the translation that's going to hold it back from being shown the same week.

I've followed CSI pretty much from the beginning. And I really like how the show is aging. Many shows fall in the trap of trying to ramping things up. But CSI has remained low key, focused on what it does well. It has kept to the weird cases and the scientific curiosities. I like that. It's too bad that all my favorite characters are taken off the show.


Sometimes I come across a quote that very succinctly captures the essence of some thought that I have had for myself but never quite managed to distill to my satisfaction. Todays quote is such a quote. Once you read it it seems so obvious.
A large enough quantitative difference
makes a qualitative difference.

Paul Hudak



Which is worse, ignorance or apathy?
Who knows? Who cares?

Murdoch James Gabbay, on the TYPES mailing list


Word of the Day: Troglodytic

troglodytic adj.
Of or pertaining to troglodytes, or dwellers in caves.
(From AllWords.com)

Thanks to Mr. Willingham on the Haskell-cafe mailing list for making me aware of this word.


J.J. Abrams Wants To Write Interactive Fiction

Occasionally I play interactive fiction games. If you're not familiar with the genre it used to be called text adventures back in the 80's when it was popular. The sole form of communication (at least in the pure games) is text, the game produces text for you to read and your actions in the game are controlled by typing commands with your keyboard. These games aren't produced by any companies anymore but there is a vivid and very much alive community which not only plays these kinds of games but also create new ones.

The reason I'm mentioning this is because I just saw a short video snippet with an interview with J.J. Abrams (of Alias and Lost fame). He got the question if he would like to write computer games and what kind it would be. It turns out he would want to write interactive fiction.

I find this interesting. J.J. Abrams' scripts often feature some kind of great mystery which makes people really hooked on his series. But as the series progresses it's very difficult to keep the excitement of these mysteries alive. This happened most notably with Lost. But I think this concept could work a lot better in a game and in particular in interactive fiction. There you don't have the problem of keeping the mystery alive for umteen seasons. You don't even have to reveal the hole thing, only enough to keep the game play exciting.

I would love to play games by J.J. Abrams, I hope he get around to write some one day.

Word of the Day

Todays' word: discombobulate


Real life Tron

If you, like me, is a fan of the 80's movie Tron I recommend you read the following blog post about how a programmer not only recreated the light cycles game but also accidentally managed to reproduce the way Tron manages to escape into the computer. The blog post is: Real Life Tron on an Apple IIgs


Moon Stories

Moon Stories are a couple really different but somehow very sweet little games. "Little" is the word here. They're experiments in game design and should only take a couple of seconds to start playing and a few minutes to master. I recommend trying them out.


Shooting yourself in the foot

The other day I came across a page which I've been looking for quite some time but haven't been able to find. It's about how you shoot yourself in the foot in various programming languages. I saw this list for the first time in the mid 90's. Now that I re-read it I find it rather disappointing. It's simply not that funny. The idea is that you shoot yourself in the foot in a way that parodies the weaknesses of the language. But I don't think they've really managed to find the essence of the frustrations with the languages on the list. Or maybe it's just me, you can go read for yourself.


Sweet website

Sometimes you go to a website you've visited many times before and you find that it's not answering. So, what's the problem? Is the site down? Is there a problem on my side? How do I find out? Well, now there seems to be an answer; the site downforeveryoneorjustme.com. You just got to love that name.
The question now is, if downforeveryoneorjustme.com is down, is it down for everyone or is it just me?


Awesome picture

I'll just let this picture talk for itself.

From Alex Brown

The Monty Hall Paradox

The Monty Hall Paradox is a classic problem to illustrate how unintuitive probabilities can be. The other day I came across a blog post which summarizes the problem nicely but most of all, the comments contain some of the best explanations to the problem that I've ever come across. But first let me introduce the problem with a quote from the movie 21:
Suppose you're on a game show, and you're given the choice of three doors: Behind one door is a car; behind the others, goats. You pick a door, say No. 1, and the host, who knows what's behind the doors, opens another door, say No. 3, which has a goat. He then says to you, "Do you want to pick door No. 2?" Is it to your advantage to switch your choice?

The blog post continues:
And the correct answer is: yes, by switching doors your chance of picking the "right door" goes from 33% to 66%, provided that the host knows which door contains the car. If the host doesn't know which door contains what and is just picking randomly, then the answer is no: your chances of picking the right door are 50-50.

Not all people agree with this reasoning, including people with a PhD in mathematics. If you want a longer expose on this conundrum I suggest you read the whole blog post. What I would like to do is to present some of the explanations for how to understand the solution to the Monty Hall problem that appeared in the comments section of the blog post.

The first explanation I want to highlight is the most common one I've seen and I think it is pretty helpful:
Imagine that there were a million doors. Also, after you have chosen your door; Monty opens all but one of the remaining doors, showing you that they are “losers.” It’s obvious that your first choice is wildly unlikely to have been right. And isn’t it obvious that of the other 999,999 doors that you didn’t choose, the one that he didn’t open is wildly likely to be the one with the prize... regardless of alien intervention?

(Monty is Monty Hall, a game show host which featured such a game in his host).
For some people this explanation doesn't help though. I think their is that they don't see that the game hosts actions ought to change probabilities. Another explanation is to see the game hosts actions in another light as in the following:
I always thought a simple way of explaining it was "Would you trade your one door for the better of the prizes behind the other two doors?" If that was the question, everyone would trade up. And that's essentially what you are doing, it's just that Monty has opened one of the other two doors - the one with the goat - to make it obvious which of the other two doors is the right one.

To me that's one of the most elegant explanations that I've seen.

Of course then there is always the alternative is to run a simulation on your own with, say, three playing cards, two 2's and an ace to represent the doors with prices. If you do in the order of a hundred runs you should find that if you choose to switch door/playing card your winning percentage will be close to 67%.


Cave Story

I had scheduled the past weekend for total relaxation. Sleeping long, watching TV, order pizza and play computer games. When doing a bit of surfing Friday night I found a game which I then spent quite some time during the weekend playing: Doukutsu Monogatari, or Cave Story as it's called in English.

Cave Story is a platform game where you run around shooting enemies, searching for items and talking to people and creatures. All these actions are driven by the games well fleshed out story. And it's decidedly addictive. I normally don't play this type of game but this particular one is so well crafted and everything, the graphics, sounds, story and so forth, fits together so well. I suppose it doesn't hurt that it has quite a bit of 8-bit flavor, just like the games that I grew up with.

Click on the image above to get to a really good page for the game where you'll find all you need to know about it, including download it. I should also add that the game is a Windows game but works extraordinarily well to play on Wine.


Nice cake

My wife baked me a really nice cake for my birthday.


How Spain won the European Championship in Football

Once you see the true reason it's not very surprising that Spain won the European Championship in football some weeks ago. The truth is: Cesc Fabregas is Sylar! With a player like that you don't even need him on the field, he should be equally useful on the bench.

  • Sylar
  • Cesc Fabregas. Or is it the other way around?
    Cesc Fabregas


Word of the Day: Hagiography

hagiography, writing devoted to recording and glorifying the lives of saints and martyrs. This form of Christian propaganda was much practised in the Middle Ages but has few equivalents in modern literary equivalents apart from G. B. Shaw's play Saint Joan (1923). By extension, the term is now often applied to modern biographies that treat their subjects reverentially as if they were saints. A writer of such works is a hagiographer.



The worst thing about new books is that they keep us from reading the old ones.
Joseph Joubert, 1754-1824


HV71 Swedish Hockey Champions!

My favorite hockey team became the Swedish champions yesterday! And what a victory it was.

They've been the favorites from the the very start of the series in the fall. And they lived up to the expectations by dominating the series, being the number one throughout the whole season. They've left the other teams to battle for second position.

The playoffs started well and until the final there wasn't really any problems. In the final they met Linköping HC which ended up second in the series, a very good team. It was good to see these two teams meet in the final as they had been the best two teams during the season.

The final is played best of seven matches and started out with LHC winning the first two games. Things were looking a bit slim. The third game when to sudden death and luckily HV71 managed to win which was just what they needed. In the two next games HV71 dominated the game and there were times when LHC must have felt completely humiliated, especially when HV made three goal being one man short during a single period. Those kinds of things just don't happen, in particular with teams as good as these. The match had now turned, HV71 leading three games to two.

The final match became a very even game with HV suffering from a lot of penalties. LHC also had their share of problems with referee decisions not going their way. After full time they had two goals each and HV71 was one man short for the first two minutes of sudden death. LHC had been playing well at the end of the third period and the match seemed to be going their way. But HV being experts in killing penalties didn't give LHC much of a chance. After five minutes of sudden death they managed to finish off the match. It wasn't the most beautiful goal in the history of hockey but it made them the Swedish champions.

Even though I'm heavily biased I must say that it was some time since there was such a worthy winner in the Swedish hockey league.


Fun city name

Department of fun city names is proud to present: Batman.

View Larger Map


Paul Graham released his much haussed programming language Arc some time ago. The general reaction was very negative, people had had very high expectations and were very disappointed. Defending himself, Paul Graham writes that this reaction is to be expected, the stuff he makes is often so ground breaking that people don't realize there is something important there. And he goes on to explain his six principles for making new things:
I like to find (a) simple solutions (b) to overlooked problems (c) that actually need to be solved, and (d) deliver them as informally as possible, (e) starting with a very crude version 1, then (f) iterating rapidly.

This got me thinking about how I tend to make new things. I agree completely with points a and b. Point c doesn't always apply for me. Sometimes I do things just because they're fun, intellectually stimulating.

As for points d, e and f I usually do the opposite. I try to find the perfect solution from the get go and I usually don't release things until I'm completely satisfied with them. This is not how I would like to work though. I think they way Paul Graham does things is much better. I guess I'm afraid of the critique I will get from delivering imperfect solutions, especially to problems that don't seem important. I have to work a bit on my trust issues.

In the end, Paul Graham's article doesn't explain what important and overlooked problem Arc solves. I'd really like to know because I don't think it looks that impressive either. So please, mr. Graham, instead of just calling us dumb, please explain why your solution is such a stroke of genius.


Kim Hartman

It's the second week in a row with a major snooker tournament. A feast for all snooker fans. I thought it appropriate to write about something snooker related. So I'm going to say a few words about the man who singlehandedly has popularized snooker in Sweden.

Kim Hartman is the Swedish snooker commentator for Eurosport. He has a very unique way of commentating sport. Most of the time he is simply silent and when he talks it's in a very gentle paced baritone voice. It's simply music to the ears to hear him talk. On top of that he come up with the most colorful and sometimes whimsical expressions to describe the play. This has even rendered him an own fan site containing quotes from him. Here's a translation of one of my favourites:

To play snooker with a pool cue is like playing golf with a shovel.

Kim Hartman's style has become very popular; papers start recommending watching snooker to listen to Hartman, I've heard of seniors who have no interest what so ever in sport watching snooker all day long and he has been chosen the most popular commentator on Eurosport. So, without any real effort, just commentating the sport he has done an amazing job of popularizing snooker in Sweden. Quite a remarkable feat. I wished you all could have a chance to hear him.

An extra bonus is that Hartman has started writing comments on Eurosports Swedish homepage eurosport.se so we all get even more colorful quotes from him that way.

Brilliant short animation


A Pointless Comparison Between Three Random Shows

There are three American shows that I watch at random intervals: Letterman, Leno and Conan O'Brien. Which one is the best? I don't have an opinion on that specific subjection but I have a lot of opinions on what they're good and bad at. Here's my list:

  • Best Interviewer. It's definitely Conan. He can be both serious and hilarious during the same interview and he usually has good questions. Letterman is second with Leno a boring third.
  • Best Band. I think it's impossible to grade musicians, all the bands are good. However, there is one band that plays the best music by far and that is The Max Weinberg 7, Conan's band. I'd say Leno's band comes second and Letterman's a distant third.
  • Best Writers. All shows have a crew of writers to write the funny parts in between the interviews. Leno has by far the best writers. They're doing some real comedy on that show. It actually makes you laugh. Letterman comes second here. It's not that funny but not as poor as for Conan O'Brien. His writers staff should be fired immediately. It's mostly just absurd, gross or stupid, only once in a while funny.
  • Best Sidekick. This is rather a matter of who is the least bad side kick. Kevin Eubanks is quite OK as a side kick for Leno. Max Weinberg mostly does weird stuff but he is not as annoying as Paul Shaffer.
  • Funniest host. Conan O'Brien win this one hands down. The way he can make jokes ad lib is second to none. He is totally hilarious on a regular basis. Letterman comes second here I think. I just don't think Leno is that funny without a manuscript.

I don't have a winner although Letterman has never been a favorite of mine. I think it will be interesting to see how it goes when Conan takes over Leno's show next year. Hopefully they can take the good thing from both shows to create something thoroughly entertaining.

How come I watch these shows? We don't have anything like these shows in Sweden and some channels show the American shows at times which suit me quite well. But I don't watch the regularly.


The Future of Science

From Kevin Kelly, former editor of Wired via Phil Wadler's blog.

"Technology is, in its essence, new ways of thinking. The most powerful type of technology, sometimes called enabling technology, is a thought incarnate which enables new knowledge to find and develop news ways to know. This kind of recursive bootstrapping is how science evolves. As in every type of knowledge, it accrues layers of self-reference to its former state."


What I got for Christmas

This Christmas I got a lot of toys or toy like stuff. I thought I'd give a sample of what I got:

It feels like a bit of a jackpot. Lot's of fun stuff to play with.