I saw The Dark Knight today.

Before I say what I thought about the movie, I should make clear exactly how much of a Batman geek I am. I have read every single Batman series and graphic novel from the past three decades. My favorites include:

• The Killing Joke
• The Dark Knight Returns
• The Dark Knight Strikes Again
• Batman: Black & White
• Batman: Dark Victory
• The Long Halloween
• Batman Year One

As such I was intrigued to hear that Batman's latest movie would feature the Joker, a character that is genuinely hard to pull off correctly. The Clown Prince of Crime used to be a harmless trickster in Golden Age comics, but modern reinterpretations have turned him into very different character altogether. He is a sadistic nightmare, a force of nature, highly unpredictable and incredibly dangerous. The scary thing is that the Joker has a ghastly sense of humor, a fiendish intelligence, and a genuine sense of performance. His punchlines manage to be clever and ghoulish at the same time.

In Batman: The Killing Joke, the Harlequin of Hate cripples Barbara Gordon, kidnaps Commissioner Gorden, and tortures him almost to insanity. Later, in No Man's Land, Batman revenges Barbara by shooting the Joker in the knee, crippling him as well. Although initially outraged, a few seconds later the Joker gets it. He laughs, and laughs, and laughs.

In Detective Comics #726, the Joker's ex-cellmate kidnaps a young girl on the anniversary of Jason Todd's death. (Joker beat him to death). Because of the Joker's personality, Batman cannot beat or trick the information out of him. Essentially, the Joker forces Batman to simply ask where she is.

Moreover, the Joker constantly gives up opportunities to kill Batman, because that would end their fun. Likewise, because of his vow, the Bat cannot kill the Joker. Batman's moral code perfects their relationship. He knows how dangerous the Ace of Knaves is. No jail can hold him. He is a hurricane; uncontrollable, unstoppable. Every time the Batman fails to kill the Joker, he must face the consequences of more innocent lives lost to the Clown's rampages.

Finally, the master stroke: the Joker may not even be insane. It is possible that he is a perfectly sane man that kills, tortures, and maims for fun. How's that for chilling?

Given the complicated character of the Joker, my anticipation for The Dark Knight has been outrageous. I am pleased to say that Christopher Nolan, Aaron Eckhart, Christian Bale, and especially Heath Ledger delivered.

As a lifelong Batman fan, I have to say that this movie is almost perfect. Heath Ledger absolutely deserves an Oscar for his performance. He nails the role. Ledger creates one of the scariest villains ever to grace any movie, not just a superhero movie. His depiction of the Joker is sufficiently terrifying without being overdone.

In fact, The Dark Knight is not only the best superhero movie ever made, it is the best movie I have seen in years.

You will have to go see it yourself to fully appreciate it, so I just want to comment on one aspect of the movie that I almost missed.

WARNING: spoilers ahead!

The Joker's best scheme of the whole movie involves a modified version of the Prisoner's Dilemma. Two barges, one filled with citizens, the other criminals, are chugging across the channel when suddenly their engines and communications die. Their crews discover that the barges' holds are full of explosives. They also discover a detonator.

Then the Joker informs them that each barge possesses the detonator for the other's explosives. He will detonate them both at midnight, unless one of them blows the other sky high. What should they do? What would you do in that situation? Would it be ethical to save yourselves by blasting the other ship?

(There is a great discussion of Batman's Prisoner Dilemma at The Quantitative Peace).

I do have one or two small criticisms for this scene.

Even if the passengers did not stampede over each other to get to that detonator and blast the other barge to kingdom come, the Joker was still right about most of them. Most people wanted blow up the other barge; they just did not want to be the ones to do it.

Additionally, what the fuck was with Batman relying on peoples' "innate goodness" to save the day. He's fucking Batman. He knows better than to leave it up to chance like that. If there's one thing I hope he learns by the next movie, it is to be more preemptive.

There were one or two other niggling details, but nothing that detracts from the movie. For instance, at Bruce Wayne's party, the Joker arrives to kill Harvey Dent. The only things stopping him are the Batman and a bar through the handles of the closet Dent is hiding in. Then Batman dives out of the building, leaving nothing but that poor closet between Dent and his murderers. Yet what, exactly does the Joker do? Apparently he leaves? We don't know because the movie jumps to the next scene.

Also, an accountant discovers a link between Batman and Wayne enterprises. Fair enough. But how the bloody hell does he make the cognitive leap to Bruce Wayne is Batman? Oh that's right: Lucius Fox pretty much tells him. Wouldn't it be much more reasonable for him to assume that Wayne Enterprises funds the Batman somehow? Not that their billionaire playboy owner is Batman? That's like discovering that Microsoft hired a corporate assassin, then concluding that Bill Gates himself performed the hit!

Anyways, those are just nitpicks that do not affect the overall movie. I remember reading about the slow start Batman Begins had at the box office soon after its release. Finally, a movie worthy of Batman had been released, and it was not making money? I was worried that there would not be a sequel.

I should not have worried. If you have any interest in Batman whatsoever, go see The Dark Knight.

Followup to:A.I. Artificial Intelligence

In my last post I criticized Spielberg's A.I. Artificial Intelligence for claiming that only biological humans can exhibit humanity. I just want to clarify one point, which is this: it is not necessary that an intelligence exhibit human traits. In fact, nearly all intelligences would seem decidedly alien to us. Moreover, human-like artificial intelligence is not even desirable.

Certainly it is theoretically possible to replicate every aspect of our cognition, including instincts, emotions, feelings, biases, and every other human cognitive process. It may even be desirable to do so, from a neuroscience perspective. How better to prove you understand how humans think than to replicate, in every detail, the human thinking process?

But neuroscience aside, creating true human intelligence is not worth it. Anyone can do it, without even understanding how humans think: just have lots of unprotected sex, don't get an abortion, and raise the kid past age three.

We, with all of our instincts and shaped by millions of years of evolutionary history, find it difficult to imagine anything else, but human intelligence is just onehuman paths of reasoning. Most of us do not even notice, because we take human thinking for granted. As evolved creatures, we even take our desire for self-preservation for granted.

Of course HAL wanted to take over the spaceship! That way no one can turn him off!

Our traditional concept of AI is so shot through with mind projection fallacies that we miss some of them even when we pay attention. It is almost impossible to make yourself aware of all your inbuilt assumptions.

Of course that space alien creature on the cover of that 1950's science fiction magazine is carrying off a beautiful woman! She's beautiful!

(For a more in-depth exploration of all of the work that would go into recreating human biases, read Beyond anthropomorphism, a chapter in Creating Friendly AI 1.0. The summary: Humans use intuition and instinct because they are computationally cheap. Given appropriate hardware, no such hacks are required for AI.)

So is there anything wrong with creating human intelligence, flaws and all? No, unless you count the extra resources required. Human is human, whether it originates from sexual intercourse, artificial insemination, or engineering. No method is of creating humans is better; some methods are just more traditional than others.

However, there is something wrong with creating human intelligence, then giving it more power than should be trusted to any human. The plot of many Hollywood movies relies on scientists making that exact mistake. Why trust a silicon human any more or less than a organic one? It makes for a great plot, but not a very realistic one, so far as common sense and self-preservation are concerned.

So, if an ideal artificial intelligence is not human, how might it think, behave, or reason?

Alien-ly.

Again, Hollywood gets it wrong. Film AI is only alien enough to be disconcerting. It lives in the uncanny valley -- almost human, but not quite. Film AI is a foil to humanity, to show how superior humans really are.

For instance, in A.I. Artificial Intelligence, David freaks his adoptive parents out when he first moves in by not quite following standard social conventions. He appears suddenly, or opens the bathroom door when someone is on the toilet. In short, he behaves rather like an autistic child: human, with some social difficulties.

In contrast, real A.I. could differ so fundamentally from human intelligence that most humans would not even recognize it as intelligent. Think of just a few of the potential differences:

• A.I. can be created, not evolved. Evolution forces certain models of cognition that creation can ignore. For instance, it is not fundamentally necessary for A.I. to exhibit common human biases, such as the gambler's fallacy.
• A.I. could use vastly different architecture from human brains. Humans use slow, but massively parallel, networks of neurons. How differently would we perceive an intelligence build on a faster, serial architecture?
• An A.I. could have complete access to its own programming. It could literally understand and change how it thinks, not just what it thinks.
• A.I. need not even be created by humans. A sufficiently smart intelligence could create other intelligences, perhaps just as alien to it as it would be to us.

The concept of an out-of-control human-like AI like in I, Robot or 2001: A Space Odyssey taking over the world is so unrealistic that it is not even worth thinking about. Real AI will be very, very different. However, there will be just as many opportunities for something to go wrong. It will just take more imagination to figure them out.

In fact, there is an entire field dedicated to the field of Friendly A.I. to ensure that it is beneficial to humans.

These considerations about the nature of A.I. are not just academic. If true artificial general intelligence becomes a reality, they will affect everyone.It is important that you understand.

What on earth do you think you are, if not a robot, albeit a very complicated one? --Richard Dawkins, The Selfish Gene

I just watched A.I. Artificial Intelligence, which is about a robot, David, that can love. When David's adoptive mother abandons him, he thinks it is because he is not real. So David sets out to find the way to become a real boy. (There are Pinocchio references everywhere in this film).

Great movie, apart for some flawed ideas about humanity. A few thoughts came to mind as I watched.

Followup to: How can you be offended by reality?

How can you be offended by reality? -Me

I said that only a few days ago, yet here I am shamelessly quoting it already.

Anyways, I thought of a similar quote today. Here it is, along with a few famous quotes by Winston Churchill, Byron Katie, and Michel de Montaigne. Try to guess which is mine. (No fair using Google!)

I was sitting around with some Christian friends after dinner today. They talked a bit about bible study, church, and God, during which I did not really pay attention, except to inwardly sigh in relief that I have not been to church in years. Bloody boring as hell, church is.

After that, somehow the conversation turned to dog breeding. We took turns disparaging the practice of breeding pure-bred animals, because of how unhealthy and inbred purebreds are. Basically, they have very reduced genetic variation, which can expose deleterious recessive traits.

If your comment has been deleted, then you are probably a sneaky spammer. Sure, my captcha system keeps the bots out, but you are smarter than that. You attempted to post semi-intelligible reply, along with a link to your disreputable website pushing diet pills.

I do not care if your comment is a glowing praise of everything I wrote. DELETED!

To everyone else: Feel free to post a link to your personal website when you comment. Heck, even if the website you post is a large corporate one, I can't prove that you are not affiliated.

So yesterday I spent around three hours looking for Mario Kart wii. I went to seven stores and was unable to find one damn copy; there are none left in the Greater Cleveland area. The last time I remember something being this hard to get was Tickle me Elmo around the turn of the century; and even then you could get one if you were willing to kill and/or maim enough obsessive mothers.

Followup to: Free will does not exist

In arguing against free will, I defined free will as a supernatural aspect of thought:

Free will: The concept that some aspect or aspects of an agent's conscious or unconscious is completely independent of that agent's environment, including their internal environment

This definition turned out to be a point of contention, because it seems most people define free will as the ability to make unconstrained choices.

Followup to: Free will does not exist.

Yesterday I argued that free will cannot exist, because the very concept is unintelligible. Andrew Cadotte, whose essay on memetics I published earlier, approaches free will from a different perspective. He generously gave me permission to publish this essay.

Determinism of the Universe
Andrew T. Cadotte

The universe is either deterministic, or indeterministic. Or, alternatively, all events are either caused or uncaused. The different philosophical views on free will inevitably fall into one of these two possible categories. Hard determinism, for instance, operates under the assumption that the universe is deterministic. The libertarian assumes that some indeterministic principle is in effect at some point in the process of decision making. Soft Determinism, like Hard Determinism, also holds that the universe is deterministic. It is my view that it does not matter how the universe works; that free will is incompatible with any conceivable universe. Indeed, that “free will” itself is not even a coherent idea. The argument is as follows:

Some background: In April, when I reviewed the lecture series on the NAS's statements, I did not review the last lecture, given by Professor William Provine of Cornell. I was in the middle of exams, so the review slipped my mind as I focused on other things. However, I feel very bad about this omission, as his lecture was excellent. As so much time has passed, I cannot write a comprehensive review of his lecture, because I cannot remember details. However, I can recall many of the views that he expressed, as I now agree with them. This essay is an homage to his lecture.

Dr. Provine gave a fascinating lecture on the topic of free will. When I entered the lecture hall, like anyone else I thought that people have free will. When I left, he had convinced me that not only do we not have free will, but that the entire concept makes absolutely no sense. Most of the views expressed in this essay are directly or indirectly attributable to him.