"No mystery," Don had said. The mystery was even deeper, stranger. But Don's keen mind had helped put several things in perspective and uncovered new clues. As Alan Ross drove the familiar route home from Don's codo, he became intensely awarethat his conscious mind was not participating at all in the process. Not only were the little adjustments to the steering wheel and gas pedal coming from somewhere well below his consciousness, he was not even aware of the decisions to take a particular off-ramp or turn -- except perhaps as a disconnected observer of a pre-programmed machine. Alan's most significant conscious contribution to the entire routing and driving process was waiting for the end of the concherto's second movement before turning off the ignition key. And he didn't even remember turning it on or finding the station! Alan continued to think about this, while he automatically walked up the stairs, took out the key, opened the door, turned on the light, relocked the door, did a dozen things to get ready for bed, and wasn't aware he was doing any of them. On one hand, the notion of being on autopilot at 85 miles per hour was a wee bit scary. On the other hand, the driving program was originally written by myself. Wasn't it? If it wasn't, how would I know?? I walk around all day, unthinkingly executing subconscious programs: Programs for saying "Hello. How are you?" Programs for putting coins in the soda machine, making a selection with no lite on, popping the top, and retrieving the change. Programs for logging into the X-terminal and entering the proper password. It's OK. These are programs I wrote for myself. I took a sequence of actions, encoded them, and saving them ready to go when we needed them again. I did it myself, somehow, to save time. Right? Right! Yeah, sure. Of course it was I who write these programs, even if I don't remember debugging them. It would be scary if somebody else wrote the programs that people every day trusted their lives to. But they're just learned sequences, that's all. Nobody else could possibly write a program that fit the structure and networks within my own brain, even if he did have some way of getting the codes in there! I wrote them. I know I did. Right? Whoa!! Let's not get silly or paranoid, here. Of course I wrote all of the programs in my head. Nobody else could put stuff in there, even if they wanted to. I know that mental programs and patterns are just selected weightings of my own neural networks. And even if somebody - just for argument's sake - wanted to "program" me to do something, they would have to know the layout of all of my neural connections, which are different from anyone else's. Just a set of weightings wouldn't be enough! Even paranoids can have enemies, but nobody could possibly have that information. And that means nobody could possibly send me a program or a neural pattern that would work in my brain. Period. I feel better now. I think I do. Forget it. For better or worse, I wrote all my mental programs. Somehow, I generated the program that obsessed me to build the device and to scan all the genomes looking for that silly pattern. Even my compulsion to to go to Antarctica was the result of some program that my own mind wrote. Nobody else could possibly have created a program that would work in my brain. And why would they want to. Only a paranoid would think anything so weird as having another person or creature actually insert a program into their brain and then somehow activate it and have it take control of their actions. The idea of somebody else or something else slipping their own programs into my head is scary, to say the least. It's downright creepy, and why am I programming myself right now to become obsessed with the idea?! Why would anybody want to do that? Never mind why. HOW would they do it? I can't imagine how. But it's an interesting problem. How? Well there was that business, years ago, about cut-up worms and other worms who acquired memory from the pieces. Scientists, who liked to call themselves "wormrunners", trained one set of worms to get out of a maze. Then they cut up the brains of the first set and fed the pieces to the second set of worms who had never seen the maze but knew how to get out of it quickly. What the first set learned was transferred to the second set which ate them. Yecch! Well, I'm glad I didn't eat any worms. Anyhow, that's something very different, isn't it? The worms just transferred data, information, memory. I can tranasfer memory from one computer to another. With flash PROM or old-fashioned cores, I could just unplug a module from one computer and plug it into another, and all the data would be there. OK, it's a little harder with ordinary DRAM, but all I'd need to do it would be to apply a little bit of power periodically just to keep it "alive". Just keep it alive or refreshed, unplug it from the first machine, plug it in and, voila!, the second computer now knows what the first one learned. No big deal. Now, that's just information. Not the same thing at all as programs, decisions, or especially neuroses and compulsions. And if you did store a program in the memory you could transfer it, but it's just a copy of a program in some memory. That doesn't mean it could take control. I've got lots of programs on my disk, and I can even copy them into memory, but nothing takes control unless I activate it by typing a command. So, even if you did transfer a program, by surgery or by worm pieces, there's no way it could get "executed" or take control. Right? Well, ... Naaah! I've got to find out more about how the brain differentiates between data that just lies there to be read and used by a program, and real programs that take over and sterr a person's course thru life. In a computer, they are totally different, even tho both can be represented with the same bits. I guess I don't really know what differentiates programs from data in a brain. Whatshisname at Poly, the guy who was always adjusting his belt bigger and smaller, and babbled incessantly for a while about that book. Something about Bach and Escher. Not an art book. "Golden Braid?" Was that the title? Funny guy. Seemed weird and not too bright, until you got into a conversation about one of the few things he liked. Like electronic circuits or compression algorithms. He was always outgrowing his pants. I remember him always twiddling with his belt. A few people thought he was perverted, but he was just trying to fit his pants to his growing belly. Like a girdle? That's it. Godel, pronounced almost like girdle, mit eine umlaut ober der "o". Not the professor, but the book. "Go\"del, Escher, Bach - the Eternal Golden Braid." Funny, isn't it how the thought of the belt triggered the memory of the title. "GEB" was an abbreviation they sometimes used for the book. And it had a picture of some wooden blocks, carved so that they cast shadows of the three different letters, depending on how you rotated them. I think the idea came from Crystallography and groups of symmetry operations. Anyhow, what was the guy's name? Not the author - that I can look up. What was the name of the guy with the belt? The professor who always talked about how it really explained how the brain worked, without getting into all that biology that the neural net people try to snow you with. I think he said there's no such thing as data, just programs. Or was it vise versa? I've got to look him up and talk to him. - = o = - "I'm really sorry to hear that. He was one of a kind." "I appreciate that. We all were stunned, Dr. Ross. Marty was a fascinating person and we really miss him around here." "He was an oddball, all right, and just meeting him people often underestimated him." "Right. You didn't realize how brilliant he was until you got to talking about something he liked." "Like Fourier transforms or analog circuits?" "Yeah, exactly." "Well, professor, I really appreciate your taking the time to update me, even tho I don't like the news. I wanted to talk to him about something he said many years ago. I didn't pay much attention to it then, but it might have helped with something that's been bugging me recently. Gee, I wish I'd come back and visited him sooner." "Maybe I can help you. Marty and I worked pretty closely just before he died. "No, it wasn't about his research. It was just about a book he once read and got excited about. He was obsessed about this book for a few weeks, and bent everybody's ear about it. You know how he gets. Got." "Got." "I didn't pay much attention to it at the time. Now, I want to remember some of the things he said about it, about the new ideas and new ways of looking at things he said the book stirred up in him. The book had nothing to do with software or computer circuits. He said it really explained how the brain works. Strange loops. Neural networks that program themselves. Something like that." "Godel, Escher, Bach?" "Yes, that's the book. I picked up a copy, but I'm afraid the stuff I'm after was what Marty said about the book, and the paths it launched him on." "Yes, he alsways said that that book started him thinking about so many things. I've read it. It's a classic, all right. But it's been a while and I don't remember any specific comments he made about it." "I'm reading it now. It's a fascinating book. By no means a dull one, but it is very slow going. I'm a pretty fast reader, but every few paragraphs I want to put the book aside and think for a while or pick up a pencil and paper and work something out." "Savor it. That's the right way to read it. That's the kind of book it is" "Sometimes, hours or days later, I'll recall something from the book, maybe put it together with other things I run into, and go off on a tangent working out some hidden thread." "Me too. Even years later, I recall things from the book I had never realized were there. Sometimes I even go back to it, but it's hard to fing things. I should reread it. Yes, Hofstetter's very good at planting seeds, and some are delayed sprouters. "Planting seeds. Delayed sprouters. Hmmm," Alan mused. "In fact, you could almost say that's one of the main themes: seeds that sprout. A little bit of information or logic, often in a very small package, gets planted and waits for the right environment to come along. Then, suddenly, the seed wakes up and grows. And often the triggering environment contains exactly the right things for the seed to grow. The seed only contains a bare minimum of information. And maybe even less, so that some of the information must come from the environment, and all the seed has to do is POINT to the pieces of the environment which it needs to make a tree or whatever." "And then there are seeds within seeds, and seeds for growing seed-making machines. I think Marty said something like that. I didn't know what he was talking about, then. That's one of the reasons I came to look for him, now." The professor just nodded, sadly. Alan continued, "what did you mean by 'when the right environment comes along.' How would a tiny seed recognize an environment?" "Oh, that's easy: patterns." "Huh?" "Patterns. Just patterns. The whole brain is based on pattern recognition. That, plus a way to build programs and store. The programs are then triggered by matching a pattern. Sort of like awk." "Awk?" "Never mind. It's just a unix program that associates actions or mini-programs with pattterns that trigger tham. I was just making an analogy. It's not important." "Awk. How do you spell that?" "A W K. It's just the initials of the three authors. But about the brain and the way it works and what GEB says and what Marty thought about all that... I could tell you a bit more, but I really think there's somebody else you need to talk to. - = o = - " Joe.