Loved this podcast, guys. Truly outstanding.

About halfway through, Trey mentioned neural networks, and it reminded me of something I read once. I can't remember whether this was in a novel or what — so it might be nonsense somebody made up — but it piqued my imagination anyway.

Story goes that somebody somewhere (I'm doing this from memory, so it's gonna be really vague) was working with genetic programming and FPGAs. FPGAs are these little general-purpose microchips that can be configured to do different things. This experimenter set up an automated test rig. It would start with a randomly configured FPGA, send it an input signal, then measure the output. Then it would tweak the FPGA in some slight but random way and measure the output again, with the goal being to get the chip to give a predetermined output from a given input.

Then he let it run over a weekend or something, and came back Monday morning to check out the results.

The rig had come up with a handful of FPGA configurations that would give the correct output. But when the experimenter looked closely at them, he saw some truly bizarre things. Like one of the configurations included a current path that wasn't actually connected in any way to the main path through the circuit. Thinking it was just redundant, he removed it and tested the circuit … and it didn't work any more. Turns out there was some hideously subtle thing going on with induction between adjacent current paths or something — I'm not an electrical engineer so I didn't understand the details, but the point is the circuit design made basically no sense from an engineer's point of view, but it worked. It took advantage (I guess you could say) of these weird secondary effects in ways that nobody would ever have designed it to do, but because the test rig wasn't set up in a way that excluded those kinds of designs, that circuit passed. In other words, nothing explicitly prevented it from "surviving," so it survived.

From what I understand, neural networks and such are built around similar ideas. If you start with the premise that you don't care HOW it gets done, and build in some mechanism to introduce variations and then another to cull variations based on some criteria, then you can get things you never expected.

That's why I don't have a problem accepting things like HAL in fiction. I don't even consider HAL to be a magic bean; certainly no computers currently in existence even vaguely resemble HAL, but we don't understand enough about how computers could be built to declare that HAL is impossible. If the movie asserted that HAL ran on the power of love and could make cupcakes appear out of thin air, that'd be magic. But the idea that a computer might be designed fundamentally differently and thus do fundamentally different things … that doesn't bother me.

Anyway, I'm rambling. But I just felt like spouting off a bit, since this week's podcast left me so full of ideas and thoughts and stuff. Which I guess is my way of saying thanks for being interesting.

You know, while we're on the subject …

I've loved "2001" ever since I was a kid. I have no memory of seeing it for the first time — I was born in '72, so obviously it had to have been on TV or something. But at first, it was just the "wow, cool" factor. As I've gotten older and more overthinky, I've come to really admire how much depth the movie has. It's open to a variety of interpretations, which makes it fun to watch and reflect on and think about. There's just so much THERE, in an intellectual, pondery, puzzley sort of sense.

Take this for example: HAL is really the only character in the movie who's motivated. I mean in the film-school sense. He's the only character in the film who actively wants something — to carry out his programmed mission — and acts accordingly. Every other character is purely reactive. The apes at the beginning of the film are starving; they'd have died out had the monolith not showed up. Floyd doesn't really do much of ANYthing in the movie. He just responds to the discovery of the monolith on the moon by lining up for a photo-op. Even Bowman, who's arguably the film's protagonist opposite HAL as the antagonist, acts only in response to other events. He only proposes disconnecting HAL after HAL's fault prediction turns out to be wrong. He only goes through with it after HAL freakin' MURDERS his shipmates and threatens his own survival. Really, the only initiative he takes in the movie is choosing to board the pod and visit the Jupiter monolith … until you remember that he's basically following the orders given by Floyd in the briefing video.

So consider this: We have the apes, sitting around literally waiting to die until the monolith shows up. Then we have man at the end of the 20th century, master of the globe and near-Earth space, complacent and idle until the second monolith is discovered. Then we have Dave Bowman — who is the personification of humanity — content to eat his earth-toned paste and watch the news through the whole movie. Nobody actually DOES anything on his own in the whole film … except HAL, who again, is driven only by his programmed-in desire to follow orders.

Viewed through that lens, it's really kind of a misanthropic movie. It says that humanity just sits on their asses until some external event forces them to respond. That in the big picture, we're all just apes cowering under rocks and gazing up at the moon in a depressing parody of comprehension.

But we can also think of it in exactly the opposite way. The ape-like pre-humans in the first act have gone as far as they can go under their own power. They're gatherers and scavengers, and they're surviving, but only just. The monolith comes along and gives them a little nudge, and on the strength of JUST that little nudge — hey guys, antelope thighbones are heavy — humanity explodes over the whole Earth and out into space. The match cut from the bone to the satellite is probably the most effective flashforward in all of cinema, emphasizing the straight-line connection between the apes and modern man by turning four million years of history into an "and then."

Remember that the Tycho monolith didn't actually do anything, unlike the original Africa monolith. It didn't change anybody who came into contact with it. All it did was send a radio signal out into space. It was humanity that chose to follow that signal. That was an even subtler nudge than the first one, but it was enough. We built a big-ass spaceship and fired it off into the dark just to see what was there. And what we found was a door, and we didn't just go "Huh, door," and go home. We opened it and barged through, because that's what humanity does.

Viewed through THAT lens, it's a powerfully humanist film. Look how awesome we are.

And thing is, I don't think either one of these interpretations is bogus. I think they're both totally valid. We can get into a debate about the artists' (Kubrick and Clarke) intent, but if we let the work stand on its own, it can be appreciated in many different ways, ways that totally contradict each other while remaining valid and internally consistent.

That's why I love "2001." That's why it's more than just a movie, to me. It's not merely a story, not merely a sequence of and-then events told in order using pictures and sound. It's a sort of open-ended meditation on what it means to be human. It's a springboard for thinking big ideas about existence and the cosmos, and that's fun. I don't necessarily mean to say I'd like to do nothing but that for the rest of my life — sometimes you'd really rather watch a pretty girl walk in slow motion away from an exploding gas station. But "2001" provides an opportunity to think about big, philosophical things, and sometimes that makes for a really fun time, and that's why I like it.

You're welcome Dave.

Just FYI, we're also working on a glossary of terms we've been using but not necessarily explaining in each episode, but at the moment it doesn't exist.  Don't panic, we promise, we're working on it.

Last edited by Matt Vayda (2010-02-16 19:22:14)

Totally agreed that the one-computer-to-rule-them-all plot point was just a clear case of science-marches-on. I'm willing to let that slide, personally, for two reasons. First, I've accepted way worse strains on plausibility with the excuse that the work was created at another time. And second, well, there wouldn't be much of a story if Dave and Frank had just done a three-finger reboot of HAL and gone on with their day.

But as for the other point, about computers needing to know what toes are before they can be like people … yes. You're right, at least as far as I know. The human sensorium consists of an unthinkably vast quantity of data pouring into our minds every waking moment, starting from birth, and that's just not something we can emulate in a computer. How the brain manages to store a lifetime's worth of memories in such a small space while being powered only by candy bars is a total mystery, at least to me.

But on the other hand, I think one of the fundamental themes of the movie was that HAL wasn't human, or even a particularly good simulation of one. He spoke, yes, and was polite enough to say nice things to Dave about his sketches. But his actions and decisions during the film were totally INhuman. He showed no sense of morality, for the obvious reason that he had none. He had programmed instructions, and those were as the Word of Almighty God to him, and everything else was secondary. When it reached the point where he determined he had a better shot at achieving his programmed objective by killing off five human beings, he killed off five human beings. Or tried to, anyway. No "they had to die" speech, no horns-of-the-dilemma. Just a mathematical calculation. For all his fantastical attributes, HAL was still a machine, and the fact that he was a machine was an important story point.

Think about it this way: Imagine that instead of being a computer, HAL was another astronaut. The mission commander, or whatever. Give all of his decisions to a human character. Seems to me the second act would completely fall apart. No human being would do what HAL did without AT LEAST struggling with it a little bit. Well, no non-psychopath, anyway. Which I guess explains why HAL seems to be universally described as a "psychotic computer," despite the fact that that's pretty much a contradiction in terms.

One of the things I disliked about "2010" (and I say this as somebody who owns a copy of that movie, watches it regularly and enjoys it very much) is that they tried to redeem HAL. Oh, it wasn't his fault; he was just sick. I (a doctor, by the way) have made him better. The others are distrustful of him, but I trust him, and in the end I'll put our lives in his hands so we can have a dramatic pause after which HAL nobly sacrifices himself so that others may live.

Not a BAD story arc. Just not one that's really true to "2001." HAL was never Darth Vader. He was just a machine. He was the thighbone of an antelope. And like we humans tend to do, we built him without truly understanding him. It's (arguably; go with me here) a straight like from the antelope's thighbone to the atomic bomb; human history is wall-to-wall with cases where our ingenuity outraced our wisdom. HAL was just the latest in a long line of unforeseen consequences.

What was it, about a year ago … like an Onion article or something? About the government contractor that was trying to figure out how to make battlefield robots run on blood. We joke about it, but the truth is, that's not that far from the kind of things people have done all throughout history. We're curious and we're lazy and we're just clever enough to do things without thinking them all the way though; this is both our core weakness, and our unique strength.

Anyway. I've completely lost sight of my point here. Thanks for the opportunity to ramble pointlessly for a couple minutes. ;-)

Some great points being made here. Jeffery's bit about our imagination:

"We're curious and we're lazy and we're just clever enough to do things without thinking them all the way though; this is both our core weakness, and our unique strength."

reminded me of a similar point brought up Michael Crichton's Sphere.

I hope I am not spoiling the story for anyone here, but basically what happens is that the characters gain the ability to make what they imagine real.  There is a quote that goes something like:

"...if you imagine good things, you get delicious shrimp for dinner; if you imagine terrible things, you get monsters trying to kill you."

In the end, as the main protagonist, Norman, is inside the Sphere, he explains it (to himself) thusly:

"The ability to imagine, is the largest part of what you call intelligence.  You think the ability to imagine is merely a useful step on the way to solving a problem, or making something happen, but <i>imagining it</i>, is what makes it happen.  This is the gift of your species, and this is the danger, because you do not choose to control your imaginings; you imagine wonderful things, and you imagine terrible things, and you take no responsibility for the choice.  You say you have inside you both the power of good, and the power of evil, the angel and the devil, but in truth you have just one thing inside you, the ability to imagine."

I would go so far as to say that the ability to imagine is what the first monolith imparted to the apes on the plain, if indeed it imparted anything at all, but for the sake of discussion let us say it did.  One might wonder why it did not also impart a sense of morality e.g., use the antelope's thighbone to hunt for food, not to kill each other, but to be fair morality, particularly absolute morality, is a human construct, and indeed, another tool we imagined.

I am aware of the whole religion aspect with regard to morality, but I'm not going there except to point out that religion is also merely a tool we imagined.

Now, let us imagine the time between the first two monoliths as being a measure of our ability to imagine.  We find the second monolith, and it sends out a signal.  Nothing is imparted this time, but our imagination has developed to the point where, while we can imagine what is out there, we need to know for certain.  So we imagine ourselves a vessel, and we imagine ourselves a computer to run it.

At this point the trouble we face is, ironically, a lack of imagination.  In the beginning, we knew nothing, and then we gained the ability to imagine.  So we began to imagine things we did not know for certain; in essence this was the beginning of science.  This would start small, "I can imagine what is on the other side of that hill, but I need to know for certain."  So you set out with all these images zipping around your head, and then you reach the crest of that hill, and whatever you were imagining is replaced with what is actually there.  Add several hundred thousand years of evolution, and you get a story where we have developed our imagination to the point where what we are imagining can be realized; we can imagine a vessel that can take us to Jupiter, and a computer to run it, great!  And why? Because while we can imagine what is out there, we are not content with that; we need to know what is out there. 

The problem we have created for ourselves in HAL, is that our imagination has grown beyond our control.  In effect what happened with HAL, is the same thing that happened with Apollo 1.  In both cases we used our imagination and built something truly awesome, and in both cases we were let down by circumstances we had not imagined.

I have to wonder then, what the intelligence that created the monoliths would have made of us.  Would they see us as we see HAL?  Generally speaking HAL is regarded as evil (he is #13 villain on AFI's 100 years, 100 Heroes and Villains), and certainly no one will argue that killing his entire crew was a good thing, but to those of us who have studied the film, we generally understand that he is simply trying to complete the mission, and for HAL, the mission is the sole purpose for his existence.  While we may disagree with his methods, we also have to understand that HAL is a product of our imagination.  We could have imagined HAL with a sense of morality but why?  Who would have imagined a set of circumstances under which HAL would calculate that the best chance for the mission to succeed, and for him to eliminate the crew? 

We have reached a point in our imaginative evolution where our tools have become so complex that we no longer understand them fully; I doubt that any one person in that universe could accurately predict how HAL would react to any given set of circumstances.  We have effectively lost the ability to imagine situations and things that our imagination cannot grasp.  I really wish the film had touched on religion at some point, because in addition to moral value, another key purpose of religion is to explain the unexplainable, or at least help us come to terms with things we cannot understand.

Based solely on the film it is difficult to posit where the third monolith fits into this model, but I wonder if an argument cannot be made for Dave’s return to Earth as the Starchild, as messianic.  I do not mean to suggest that the creator’s of the monoliths are deities, but as has been pointed out in numerous commentaries, any sufficiently advanced technology can be regarded as magic, or if you prefer, the work of God.  We spend the entire film with characters that do not regard the world they live in as in any way extraordinary.  Even today that is true, and my argument is that while we have films like 2001 to remind us of the awesomeness of our world, the people that inhabit the world of 2001 need someone to come along and remind them.  Obviously there is no one inhabiting that universe that can do that, hence the need for someone to come along and point it out.  Dave experiences things he could not have imagined, he himself is re-imagined in a way, and sent back as a messenger of sorts.

Remember in school when the teacher would give you some paper, crayons, scissors, glue, etc., then leave you basically on your own for a while?  Most of the time these efforts were rewarded, and given a place of honor on the refrigerator, and every once in a while someone drew something that is got them a visit to the principal’s office.  Now imagine it a few years later, Mom has pulled out a bunch of them and you’re looking at them together, and suddenly you remember how much fun it was.  You remember how powerful your imagination was back then.

Now imagine that all of us are those children, and the Starchild is the mother. 
Now imagine that you get an idea.
Imagine that idea being Photoshop.

That’s 2001: A Space Odyssey.

To go back to HAL and his "evilness," one thing we didn't mention in the commentary was the concept of the Turing Test. We can debate back and forth whether or not HAL constitutes real sentience or just an approximation of it, but based on what we see in the film, HAL would pass the Turing Test and therefore, on some level, qualify as genuine intelligence.

Wasn't he a good simulation of a human though? As in the District 9 commentary, I wish I could say I could find such behavior in humans impossible to believe. Sadly, such behavior is all too human. That's the precise definition of a psychopath or sociopath, after all; one who mimics social etiquette without the genuine emotion. And all children are born sociopaths, caring for nothing except themselves (and excellent manipulators to boot) until they are forced (hopefully) to learn genuine empathy.

In those ways at least, HAL IS the most human character in the whole thing.

Last edited by Brian (2010-03-03 19:09:20)

(Something else I wish I'd remembered to talk about in the commentary:)

It's entirely possible that it's flat out impossible to get human being level higher reasoning WITHOUT a limbic system. Because of Descartes, we tend to think of logical reasoning and emotion as two separate and distinct parts of our brain, or more colloquially, we picture ourselves as "thinking" with our brain and "feeling" with our heart.

Now, obviously we know that's literally not true and all thinking and feeling is done in the brain, but we still consider "thinking" and "feeling" as distinct modes from one another. That's certainly how science fiction has always treated the subject when it came to artificial intelligence.

But I read a book a few years back called "The First Idea," which proposes a model of human thought where higher reasoning develops out of simpler emotions. The authors suggest that as children we start off with very simplified emotional states - we're either happy or absolutely inconsolable, with no in between. They also suggest that we also have a limited ability to have a sustained interaction with another human being.  You get so many moments of give and take before the baby's brain resets like a goldfish. But as we get older, the chain of moments we're capable of doing gets longer - you can carry on a "conversation" for longer and as a result, develop more nuanced emotional states.

Instead of only being able to be happy or inconsolable, you're now capable of being inconsolable, angry, frustrated, and annoyed. Eventually, some threshold of self awareness gets crossed and you're capable of saying, "I know I'm annoyed." And eventually, "Why am I annoyed?" which leads inevitably to, "How do I fix my being annoyed?" And we've arrived at higher reasoning.

It seems to make sense to me, though I'm obviously no clinical psychologist and it does destroy a hundred years of science fiction robot writing, including my beloved Data. And there's some interesting crossover between this idea and the fundamental principles behind acting - or at least a certain part of it - that I can't quite articulate just yet.

Nice work on this, guys.

Gold medal Super-win! for Trey for mentioning Space Food Sticks, the greatest chocolate rubber compound invented by mankind.

I find it interesting—and a problem with the film—that the discussion here is all about HAL.

HAL’s dilemma is that the humans are planning to turn him off, which would jeopardize the mission, so HAL decides to get rid of the humans and carry out the mission himself. The problem is that that is actually entirely incidental to the story Kubrick is telling (the ascent of mankind) and merely serves as an entertaining complication. After all we are not introduced to computers or tools earlier in the film as a kind of villain-of-our-own-making. But it ends up being so entertaining, that it takes over the movie.

I remember reading in the TV guide when i was a kid "man travels to jupiter and meets himself." Talk about people still not getting it 10yrs later.

I listened to this again last night and just wanted to belatedly say how good it is. Trey's enthusiasm, the guys pointing out a lot of details, etc. I love it when Teague, who straight up never liked the movie, asks "why". It's one thing for people to say "it's boring, i hate it" but when the discussion is "there's a reason people love this thing, i want to know explain", it's interesting. Then he has an epiphany!

And the great comments above. The only thing i can add is when Brian talks at one point about how the spinning gravity thing might not work if the cylinder's too small because G would be lower at your head... That doesn't sound right. Seems to me the direction of force goes through your body like an arrow from the top, and pushes the whole body down. Who knows...

And great show notes. The Kubrick and slit scan links alone are interesting reads. The cute ending left a smile on my face. A great commentary.

It's been a while since I listened to this one (need to put it back in the queue), but assuming that's what Brian really said, he's not technically wrong, but he's partly wrong. The force exerted against the inside of a spinning drum is, in fact, a function of the radius of the drum. But from what I understand, that's not really why a small-radius centrifuge wouldn't work. I've read that it'd fail because of a combination of balance issues and Coriolis effects. If the radius of the drum is small in proportion to the size of your body, then the force pulling on your balance organs in your inner ears would be sufficiently non-parallel for it to really foul up your sense of balance. And any freely falling object inside the drum would appear to fall in a huge curve, which would also work toward upsetting your self of balance.

All in all, it'd be upchuck paradise, is how I've heard it described. All things considered, better to just stick with zero G for a trip of less than a couple years.

There's a book I haven't read in years, like since-developing-taste-ago, but considering it's written by Eric Idle it's probably still excellent. The Road to Mars. As I remember it, some comedians travel a solar-system wide vaudevillian circuit while their android tries to write his thesis and figure out the nature of comedy. Deals with as much of the "human/robo wuaaa?" as you can read into it while remaining really enjoyable and funny.

Oh, and this down here is the reason for the post, but didn't want to just drop a bullshit image link into an interesting conversation.

http://twitpic.com/2wuzsv

I made it halfway into the second part before turning it off. I don't disagree with what seems to be his only observation, but he kind of annoyed the shit out of me and made my ears hurt.

If at some point he makes it clear that he knows the whole subtext bit about the monoliths, then, yeah, I'd be able to have a non-contentious conversation with him about the movie. I just don't like his style. (And by the time I turned it off, he hadn't seemed to indicate that he knew what was going on with the monoliths and the general metaphor of it.)