"Hello World."

I should have posted the above earlier, it being a rather classic "first output" expression.

In either case, the world is still spinning as far as I know. That being said, it's spinning a little more than usual thanks to wonderful, wonderful e28. I shall need sleep very, very soon. Barring Warcraft.

I really ought to go to bed, though, games or no. I have been dragooned into fasting tommorrow, so I'll need to be up early to eat (and then I'll promptly go back to sleep - first class is at 11).

Interesting things happening in my life: (this is why you read this, no?)

I got a crash course in Heidergain (how do you spell that?) philosophy in Matrix class today. Prof Dreyfus came down to discuss some aspects of Revolutions and we got bogged down in the nature of Matrix reality. I see his point, and I can even see how the situation he sees is probably the most logical explanation for the Matrix. An in depth look at this follows at the end of this post.

Also, E's okay with me crashing her whirlwind tour of the bay area. This is way cool because Berkelian that I am, I still haven't seen much of the Bay, and I really ought to. Would like to get a cool trenchcoat too because it's cold and I need another coat. The drizzabone is nice, but I get rather odd looks wearing it.

And, best of all, I'm the strangest person D knows. Also the sweetest, but that's hardly a compliment... Strange is cool, though. Nark! All hail the cheerfully insane!



Okay! On to philosophy:
It's good to have someone who knows this explain this stuff to you. There's a tendency to fall back to your own personal philosophy rather than fully embracing a philosopher's standpoint. I know some feel that we shouldn't accept someone else's ideas on faith, but the point is that accepting and understanding another philosophy is not the same as believing it. Rather, it behooves one to comprehend what's offered so we can properly contrast it to our own views.

And this utter rejection of differing points of view is what plagued the class today. In hindsight, it could be said that Prof Dreyfus didn't fully explain the foundation of his reasoning and philosophy before hitting us with it's implications, which meant that almost everyone disagreed, and once those antagonistic viewpoints surfaced, it was obviously difficult for people to change their minds.

What does Heideger say? The basis of our argument today was the principle that while there is an objective reality, humans never really perceive it. It's not just a matter of sense leading to intepretation, where our brains pluck meaning (and different meanings, depending on upbringing, nature, etc) out of a stream of purely physical information. Rather, our individual state during the act of perception affects what we perceive. Like a selective filter that filters information prior to intepretation... Examples we covered today included the different ways people viewed other people: in Greek times, a person was a Hero or a Slave, under the Christian worldview, you're a Saint or a Sinner. Both these dichotomies are incompatible, and we see a time where we have a paradigm shift in perception. Under the Heidegerian view, how we categorise people would affect how we perceive them - what chracteristics we choose to notice. And this filtering is intrinsic to the very act of perception. You cannot perceive something without loading it with your worldview. It's just impossible.

Our matrix example is birds. Birds in the matrix are governed by a program, the Oracle says (and how are we to disbelieve someone who can accurately fortell the future?) Now, the bird "frame" (AI reference. A "frame" is all that an AI system needs to know about a bird to be able to recognise and simulate it) is limited to, as Smith says in the original movie, to what existed in the early 21st century. (I think I'll swap over to regular programming jargon. Let's call the "frame" an object - which is what it most closely resembles in the matrix world anyway. An object is an instance of a class. A class is a framework or template the defines the properties of objects. Classes can be inherited, so for example, a "bird" class could describe birds in general and pass on properties to a "pigeon" class which would then proceed to define more specifics about the type of bird in question. In AI speak, a pigeon is a bird.... Objects can contain information, and they also store functions or methods - means of manipulating themselves or other objects. Basically the upshot of this is that the bird program can very well peck at you because it probably has a "peck somebody" function stored. Birds also weigh something because there's a "how much do I weigh" data entry somewhere in each bird. ) So, what is the problem here? Well, the viewpoint the birds were programmed in is probably rather limited. In class Prof D. mentioned that people once viewed birds as natural wonders, then later as creations of God. Now we see them as biological creatures, endangered species, whatever. In any case, because worldview affects what we see, the matrix has an extremely hard time reacting to people with really odd points of view. Of course the birds will have information to cover most points of view: a happy person would see the birds in a cheery light, or an angry one in some other way. A religious person might perceive that the birds are as aspect of the perfection of God's creation, etc... But the number of such views the matrix can deal with is limited. The matrix knows what each of these people wants to see, and will send the appropriate data. A hungry pigeon eater will see the right sort of data he might need to catch and eat a pigeon, or to think that a pigeon is tasty (possibly he's the only one to catch the "good" scent off the bird - because his mind wants that smell to be there and the matrix is programmed to respond to that desire). A pigeon loving bread tosser wouldn't receive the same signals from the matrix. What's bad about this is when someone comes in with a completely novel POV. The matrix can't deal with it. This person has "trouble accepting the simulation." If this happens enough and people don't "accept the program", then the matrix will crash, which is a bad thing.

Now, that is a key point about the matrix. It, by it's very nature, limits creativity. Or more specifically, you can only see things though matrix vision or not at all. This divides the world into people who are content with this lack of freedom and those who aren't. And therein lies the reason why a person might want to be out of the matrix - sure you have all the steak you want, but them you can't experience life in new ways (one of the key "joys" of human existence under the Heidegerian model is that of creativity). Those who stay could be labeled as "unauthentic" and those who leave "authentic" (why this term exactly, I don't know, but I guess it's because they don't fully experience the range available to them as human beings)

Now this all only works when you assume that the matrix program is limited. That only a limited amount of bird data is stored. What would happen if every atom in the bird is simulated? Would that not lead us to a "real" reality? To dispute that the experiences in such a "completely simulated" world are identical to the "real" world would be to dispute the very basis of the empirical method and science. Basically, there were a few adamant souls who felt that every physical law would have a model in the matrix, and that it could all be simulated. Hence, the matrix would allow creativity, insofar as the real world does - there is no limiting of perception because you get all the data you would have gotten if you were in the real world.

Prof D. is a rhethoritician and philosopher, and didn't dicuss the technical requirements for this. The people proposing these arguments were also not physical science students. I can say, with some justification, that the "model everything" approach doesn't make technical sense. Lets look the reasons why:
1)Why am I wasting CPU cycles? Obviously the machines wouldn't waste precious CPU cycles simulating and excessively complex simulation for the entertainment of mere power sources. It's like building the internet of cattle. A waste and they won't appreciate it anyway...
2)Computation, computation, computation. To simulate a world as complex as reality, it requires one heck of a simulation. That means a heck of a lot of computing power. Now, this may seem like a similar argument to point 1, but what I am going to say here is that you couldn't possibly come up with enough CPU power to do the whole physical simulation thing. Why? Take Stephen Wolfram's research into computational life and complexity - even simple systems can be arbitarily complex, and their patterns of behavior are impossible to project/preduct without actually performing an observation. This means that there is no "short cut" to simulating a system. In most truly complex systems (ie, most systems in the world around us) you might as well calculate the position of each atom every picosecond to get a good simulation. And true enough, this simulation would be essentially reality. The problem is, by the time you get enough computing power to calculate the position of every molecule in a city, you might as well build the damn city and sit back and watch. In fact, the "simulated" system takes more computing power than you can contain in the city itself (obvious principle: the best/most efficient simulation system for something is that thing itself) So, to simulate 1 planet full of humans perfectly to the level required, we need at least enough space for 1 planet full of humans. Obviously the machines don't have this much real estate on hand, so this you need some sort of simplification of the system to simulate it at all. And any simplification sets limits on what is achievable in the simulation vs. reality itself.

So essentially I'm saying that perfect simulation is technically impossible. Without perfect simulation, all simulation is limited, and by that very limitation, limits the scope of the people living within that simulation. The matrix is thus a not nice place to live, more or less. This argument on technical grounds was what was missing in today's discussion, so I hope I can count on this as a rebuttal to all those who persisted in the argument for perfect simulation and nothing being missing.