 Hey folks, welcome back to the Artificial Life Advanced class. We are at T-minus two and counting. Check it out. So that's it. It's been six weeks. I had some big fun. That's a, we'll see why I picked that particular shot of molecular configuration in a couple of minutes. I have a tremendous amount of stuff. The goals for this time was proto-ancestor runs. I'm calling that what we just saw, that the proto-ancestor did not successfully have an offspring, but it, man, it did run. Publish a book. I'll talk about that just a bit at the end. Have big fun. Okay. Yes. I had some. Will we see grandchildren for the ancestor? Maybe. Maybe. Maybe. It's close. Okay. Lost to talk about starting with timeline revisions. Okay. Last time, we had a lot of stuff still missing, diamond cutting missing, redeploy centers missing, and so forth. As of now, redeploy centers, I'm saying we have that. In fact, there was some stuff going on that was a little bit tricky in that video. I'll talk about it. Well, it's not during this time, then during the live stream afterwards. And now we had the whole month of December as a simulated evolution stretch goal. We are now taking that over. Real evolution will not happen. Real evolution might happen if there's bugs, but we still do not have diamond cutting. That is the big, gotta get diamond cutting. And then finishing the ancestor code, tuning it up, making it all work. That's the last step. Then we ought to be able to be doing runs on the grid, atomic live chain. So you know, back in T minus nine, the question was if we're using a pocket chain to represent our digital DNA information, we have to be able to grow a loop. But the loop is closed at both ends. We can't just grow around like we could with a swap worm. So and I made up new grow, which eventually got renamed to shape shifter and new grow, puts out a new corner, which puts out a new side, which eventually hooks it all up. And that was I convinced myself was absolutely as the best you could do in principle, because there was no way for it to happen in one event. Because if you have an event window on one guy, it cannot see all the other things that would need to change. So specifically here, this, if we wanted to do something with this EC, it can see its whole box, it can see its neighbor to the east, but it cannot see all of its neighbor to the north. It cannot see northeast. It cannot see all of its neighbor to the northwest and so on. Similarly, if we look at this empty one down here and say, could it do something? It cannot see. You cannot see all four pockets. What are you going to do eventually in November? The pin dropped. Yes. Well, but what if we let the event happen on the grid, like instead of having a particle to grow the loop? Suppose we have a field of loop growiness that can just happen like this. And from an event that's happening at this hard cell G, it can see all of the four pockets around it, which is what we need to make an atomic revision that could affect four codons at once. And this is exactly what the diamond sequencer has been doing all along. But I never thought that that could apply sort of everywhere. And in fact, it works great. Here's an example, event is here and it pops out a whole new pair of empty codons, ECs. Here's another one. I call it re-corner, which flips one in and out. Same thing. You know, once I got that, so I had to reimplement a ton of stuff to do this. It took a long time. But once I got all of that stuff going, I could really just write rules fairly quickly. And then here's the shrink rule. And with a shapeshifter, we never even got as far as shrinking. It's just too complicated and horrible. I couldn't stand to do it now, boom, like that. So that's what we just saw in the opening video. Those loops that we saw, they were not growing out like fingerprints like we saw earlier in the countdown. They were live. They were bouncing all over the place. Okay. Getting a grip on, quote, an e-vow. The whole idea that Von Neumann's original universal constructor was demonstrating was that, you know, in order to make a copy of something, you don't have to have a smaller thing inside it. Instead, you could have a description that gets used, a double use. It gets interpreted as code. And then it also gets interpreted as data and gets copied. So at one point, we need to execute the code. In the other case, we need to quote it so that we can talk about it as if it was data. That's what, when I was originally building the diamond sequencer, the whole idea of having these four different functions attached to the pockets around the diamond sequencer, having those be configurable was key to this idea. Oh, and by the way, yeah, we had originally a diamond sequencer was right above the root in the center of the diamond. In this, what we just saw there, the diamond sequencer associated with the mother started to drift. It started to head east relative to the root. That's how we're getting at deploy centers and then the functions to change. Yeah. And it also turns out that aside from the instruction chain, which is really important, the other three, they're, you know, at this point, I'm kind of calling them red, green and blue, using them for more or less whatever I want. They don't have that much special functions. They have a little bit. So here we are at the diamond sequencer. We've got the white patches, the instruction stream. We've got a do in there. And when that executes what happens that it copies itself to a different pocket and it re grips the processor. So now it's executing over here and that allows the do. And then what the do is it waits it. Oh, man, I didn't say it that there's a little gray rim around that red flag there. And that gray rim means that we've been requested to roll forward, move on to the next non empty codon here. And that eventually happens in the gray border goes away and it's a copy Adam that comes up. And when do sees the second Adam that came right behind it, what it does is it replaces itself with the do and then says once again, roll on to the next one. And it's now the copy that starts to actually take the loop. And now we are quoted. Now the contents of the program loop are no longer being executed. They're being read by the copy process and fed into this transfer cable that's leading to the daughter. And so there it is the daughter loader. That's the first instruction that gets passed through the line to set up the daughter to receive the rest of the program. So and off it goes. And finally, when the when the copy reaches the end of its designated copy, it erases itself and re grips back to the way it was. And now the loop is being treated once again as instructions being created as code. And this is all working. It's really cool. All right. So parallel distributed processing. Yeah. So again, this is something I didn't realize immediately at first. But the way to make this all go is that we have the diamond sequencer that's running the sequencing that's evaluating the loop. We deploy a second diamond sequencer inside the same diamond grid to represent what's going to be the daughter. And so it gets deployed and it's set up so that its execution stream is pointing back at the thing it came from. So that whenever whatever instruction manages to get to that pocket, first the daughter diamond sequencer will execute it and then off we go. So here comes the daughter loader Adam. I was hoping to make that originally just be a do but it's customized enough that that was one of the cases where I cheated a little bit and made a special one just to get the daughter going. And here it is. It's now reached into the the instruction pocket and it executes. And so it again, it re grips the processor makes a copy of itself someplace else. And it pops out one of these little looper things that pops out into a loop like that and then here comes the copy and it all just starts getting fed through. The field effects start making more room in the daughter. Off we go. Another crashed. Check this out. So yeah, there it was. That was the same thing that was from the Beethoven sonata last time. The power supply on Lotus one had gone bad and the voltage that was getting fed to that the Lotus one tiles were seeing was like nine volts and it was supposed to be 12 volts. And they, you know, I had to actually get rid of the debesler so I could see the power lights around the thing. The whole thing was flickering. It was a terrible mess. I replaced the power supply with something beefier and booted the whole thing back up. And what I realized was a whole bunch of the tiles, four of the tiles in Lotus one had gotten their discs corrupted. What happened? The common data manager just discovered all the checksum errors and fed the data into them. And in fact, the Lotus one completely healed itself. As far as I can tell, and it's running fine. I didn't reimage anything. I didn't do anything. They took care of themselves. Nice. And finally, Publications Division report, you know, the reason I officially take November off is because it's Nano Remo National Nominal Writing Month. I've been doing it for several years now. I haven't published anything. I still haven't published anything this year. I keep my street going. But what I was working on was a book by my dad who died in 2008. He was a philosopher. He was an ethical ethicist. And he wrote a book about a new way of doing, of thinking about ethics to make an ethics of a serve as a philosophy of life. And so I use the format that I got from RIN, one of the folks that has followed us on the T2 Tile Project. Thank you very much. It's got, you know, the drop cab and nice fonts and all that stuff. And building it all out. My best friend from seventh grade has been helping me with proofing, doing error correction and getting the citations cleaned up and so forth. And I'm writing a forward for the thing. I think I'm going to write an afterword as well. It's funny because the contents of this book, which I never actually really read until just the last couple of months, really gets at a bunch of the bottom up robust first complex systems, emerging properties, all that stuff. My dad and I used to have discussions about it. And I think this book in some small measure reflects the arc of his thinking from the traditional view of ethics, where it's about what you're supposed to do and what society tells you to do and have to keep order and all that stuff to a much more relationship networking based person to person based on caring. And wow, that's really good. So we're titling the book companionate caring and ethical philosophy of life by Sheldon Seagley. This is a early possible cover. None of it, none of this is finalized yet, but very exciting. I would love to get this out in 2022, but since we've got the a life creation countdown coming in blast off in the same time, we'll see how it actually works. And that is it in two weeks. It will be T minus one as close as you can get without blasting off. And the essential number one goal is programmed diamond cutting. I've been avoiding that. And I'm worried about the loops being all sprawled all over the place. And we cut it, they'll tear the loops. Time to bite the bullet and face it. I would like to also finish the forward to companionate caring. We shall see and have some more fun. Folks, thanks so much for coming by. I hope to see you next time.