 The T2 tile project is building an indefinitely scalable computational stack. Follow our progress here on T Tuesday Updates. And we're back. It's probably going to go on a little bit long this week, but hey, it's been two weeks. So the COVID-19 disease got a little bit closer to a lot of nerds this week as it took the life of John Horton Conway, a renowned British mathematician best known at least outside of mathematical circles for inventing the game of life, a zero player game, right? And it's a two dimensional grid with squares that go on and off and so forth. And that's obviously relevant to the movable feast because it looks like it. It's a 2D grid of things happening. And there's very frequently running into, you know, people making the link between the movable feast and Conway's game of life. This was from the YouTube comments. There was a comment on Reddit some time ago. I didn't go dig it up. Probably better than that. I didn't say, you know, somebody should tell this guy about Conway's game of life and so forth. And, you know, so it's very important to me that the movable feast machine isn't Conway's game of life just because that's what people tend to think of when they think of anything because Conway's game of life is so famous. But the interesting thing about it, at least in this particular context, is that John Horton Conway himself did not actually respect the game of life the way his legions of fans, the way the game's legions of fans respected it. And this was an interview that Brady Herron did with him several years ago where he talks about it and how successful it was and how Martin Gardner's column in Scientific American Mathematical Games was originally described. People got also excited about it. It was really nice. It's kind of funny because, you know, Conway is fundamentally sort of a nerd, wild man. And you know, he doesn't do political niceties well or all that often. And so here, you can hear him sort of trying to be magnanimous to fans of the game of life when, in fact, he thinks it's really sort of dull. The only mathematical interest to it is showing that it could be done, coming up with the rules, and then doing and proving that they were universal, that anything you could do with any computer, you could do one way or another with the game of life. And then after that, all the rest is uninteresting from his point of view. He doesn't hit on it too hard here, but the story, my own story about it connects to it from that and the Artificial Life Conference six years ago in New York City at the Javits Center, which has now been turned into a hospital for COVID-19 patients, that John Conway gave a keynote at and that from some points of view, depending on how you look at it, I was a backup keynote speaker for John Horton Conway. And so, you know, this was the poster that was in and, you know, there it is. He was notably the inventionist at the time of Game of Life and so forth. Now, how do you become a backup keynote speaker? Well, you know, six years ago, I was, you know, young and naive and I was trying to bring the robust first thinking to the world as best I could. And I wanted to speak to the artificial life community because they were, you know, the closest to the area that I was in that seemed relevant to. And so I sent an email to the conference organizers in 2015, offering myself as a keynote speaker, you know, not really understanding that that's not how keynote speakers are actually selected. And, you know, so I didn't get invited and but I went to the conference nonetheless and I had my work to show and so forth. And then during the conference, some of the conference organizers came by and said that there had been some concern that Conway might not be able to make it. He might be late. He might be later. He might be ill. I don't remember exactly what it was. And, you know, would I be ready to step in as a backup keynote of John Horton Conway didn't do it. And, you know, really, I think, you know, mostly the conference organizers were kind of mocking me a little bit. But if, in fact, he didn't show up, then they would have a problem. So it was sort of a win-win solution from some points of view. In fact, he did show up. He arrived and gave the talk and it was quite a talk because he really laid into people that were doing cellular automata, not just Game of Life, but pretty much everything about cellular automata was, you know, from his point of view, sterile and uninteresting. And he spent most of the lecture, at least it seemed like a long time back half of the lecture, talking about his free will theorem, which certainly baffled me. And as far as I could tell, it was baffling an awful lot of people in the room. But if you drill down through it and I read the paper afterwards and made a little bit of sense out of it, you know, I'm not really much of a mathematician, but, you know, what I got out of it was that, you know, basically it was saying that if you want these particular properties that seem like it's something that a scientific theory of reality ought to have, then on the only definition of free will that made any sense, that there ought to be things that couldn't be predicted by the path, the historical path of the system, then if free will possibly exists, then elementary particles like protons and stuff would have to have free will. And that under in this guy, Cantron Lansman, a number of people have criticized the work as being a tautological or sort of too uninterestingly tautological or something like that. But it twists on the idea of determinism. And the quote from David Hodgson, science does not support determinism. And hey, I'm alright with that. And I think really, that John Horton Conway's criticism of his own game, the game of life was fundamentally because of its determinism, which meant it was detached from reality and purely defined by mathematics, by the by the simple rules, and at which point from a mathematical point of view, the rules, you know, there it is, the consequences unfolding of it, that's engineering. But the criticism he laid on the artificial life community that day in 2014, in the Javits Center, didn't actually land on our work, didn't actually land on the movable feast machine, because we had jettison determinism. And the course of the computation that happens in the movable feast machine, inherently, necessarily depends on the underlying physics, the actual manufacturing of the chips and the temperatures and everything that happens, the races that the tiles happen against each other. And that, I believe, is escaping the free will theorem that John Horton Conway, in fact, was talking about when he was being the keynote speaker that I was allegedly the backup for. So, thank you, John Horton Conway. I was a fan of the game of life. It was part of what drew me in to distributed computing, cellular automata, living computation, and so forth, got me where I was today. Even though I think I agree with what I take to be John Horton Conway's criticism of his own child, the game of life, there is another step to the game. And we want to try to help people to take that step as best we can to move beyond determinism, and to focus on the reality of systems that are embodied in reality, the real systems out there. And so my feeling is, you know, I criticize the game of life sometimes. I get cranky about it because so many people mistake the move of feast for it. But really, you know, it's a wonderful step. And, you know, it's a little bit too easy. That's what determinism makes it a little bit too easy. But, you know, an easy step will always lead you wrong eventually. But if you keep on walking, all roads bring you home. Rest in peace, John Horton Conway. And, you know, in a sense, he's sort of a spiritual figure for the living computation foundation as well, game of life, computers, and so forth. The other news for the living computation foundation from the the memoriam to the most practical, we're getting very close now to actually having a bank account for the living computation foundation. In two weeks, when we talk again, I am determined that there is a PayPal button on living computation.org. We shall see. So what I want to talk about for the rest of the time today, maybe another 10 minutes, do a demo is, you know, intertile events. And, you know, this has been Dave's white whale for a long time. You know, going back to episode 244, last August, and certainly longer than that, depending on how you score it, have been saying, you know, intertile events right around the corner here need to do this, whatever it happens to be. Part of the problem, and we got to this talk about it in the last couple of weeks, was that I had the again, naive, I'm going to be a naive assumption that we were going to be able to take the simulator code base and kind of just plop it on the tile and shim around the edges and get the whole thing to work. And so in this particular notes that I'm talking to myself, you know, can we feed packets down to the cash processors and then, you know, make a bunch of delegates to sort of do the shimming. And it just wouldn't go and it wouldn't go and it wouldn't go and months went by. And so finally, the right, the report taking the different perspective from the point of view. And finally, just a couple of weeks ago, we got to, well, screw it. Let's rebuild it. And, you know, the way I'm looking at it now, it's not even a rebuilding, it's a new start. And so this is what we've got today. What we're building is a native engine for the T2 tile. And you know, I started building it up. And at this point, it's gotten, I don't know, probably several thousand lines of code I haven't actually counted yet. And I've been bad about checking it into the repo. But the idea is start fresh, blank screen, new main program, start working way down, rebuild in and just gradually pull in the sort of low level bits that are sort of, you know, least committed, like the display related stuff and so forth. While in the process of building up this new approach to doing events where we're going to have lots of event windows and flights simultaneously and so forth. So I wanted to show you the progress we've had over the last couple of weeks on the T2 native engine. So I've got a demo here. And there's another issue coming up, which is, you know, if we're going to need to be displaying this to people via video, you know, we've got to get some better way. This this camera that I've got this, you know, it's one of these sort of knock off go pro type cameras has trouble getting good images of LCD screens. And I've tried to work on it. I've got it down here today. And I've got it sort of hidden under like a photographer's cloth. Because it seems like basically you want to exclude the rest of the light and then try to underexpose the camera as far as possible. This thing doesn't let you underexpose only a few stops and so forth. So it really doesn't look very good. And it's a problem. I'm not exactly sure yet how to solve. But all right, let's look at it. So step number one. You know, what do you see here? Well, you see nothing here. But the key is, is that that is actually the grid where the atoms are going to be. Why are there no atoms there? They're not implemented yet. But we have the user button, which has been sitting here really not doing much of anything for anybody for a long time, is now there to bring us up to our global menu. And we can look, we have a whole thing now for showing the internal log of the engine code, I was going to say the simulator code, but it's not a simulator anymore. It's an engine. And stuff goes by and oh, and yeah, here's another thing we show you forget my phone here. There, see active light. Yeah, there we go. So the thing is not only detecting light changes, it's injecting events into the master event queue. At the moment, all they're doing is getting picked off the queue and reported to the log file, but they're available there and will expose them one day to alarm code to figure out what they want to do. I don't this probably doesn't look that great. You know, there's a trade off, you know, any small fonts just doesn't video very well. I mean, there is hope because you know, my camera, my phone camera works better than this little thing. So we got to figure out a better way to do it. The sites we already saw, that's the grid view. And then there's the the tile view, which this is trying to be what T two vis used to be that and you can see here, we've got a bunch of the grid voltage, the core temperature or the edge temperature and so forth various number status reports being put up here. The background is changing color. Because I was just doing some performance checks and the way the display works, if you only change a little bit of the display, it can actually go faster. So I wanted to be doing full screen updates because that's more like what'll happen when the grid is full of atoms that are all shifting around. But a lot of this is really just stub stuff, you know, the x ray icon thing would just show myself that I could do the icons and so forth. But it all is there and it all actually is running and it's all in a single image. T two vis the stuff that we had before was a completely separate program from MFM T two, which was the code base of the simulator on the tiles. This is now all one, which is why we can get back and forth between them seed drag. Yeah, that would be great. But we don't have seeding drag yet either. So so that's what it is. Now, you know, a lot of this code is brand new and rickety. I mean, so I've got a new little language sort of like an MFS file, but it's not really because that would be pulling in too much of the stuff from the original simulator that allows us to configure these windows, you know, and do it really quick. Yeah, so here, we have a little yet another I'm sorry, it's not Jason, it's not anything familiar. It's just a bunch of little, you know, declare a description of a window and then give it a bunch of properties. So where's T two info or here it is. No, it's the status panel. That's the one. This is terrible lighting. So suppose let's make the foreground color just wait. Alright, so hopefully that'll be more visible. And we can rebuild this little configuration file. It's a pearl script. Always is. And restart the thing. And we'll see what it looks like here. So this is the reasonably quick edit loop coming up. So there so now it's white. Hopefully that's a little bit more visible. We're gonna have to start making an awful lot of user interface decisions for visibility for camera recording ability going forward. So that's the status of the native engine. It's not nearly as much progress as I wanted, of course, but it's what we've got. And so the next update will be in two weeks. April 28, which will be right before the a life paper is due for real May one. And so for the next couple of weeks, I'm going to push on the native engine as best I can, but I have to switch back to the wall of science stuff for the paper and get that ready and get that out. So hopefully, folks are managing to stay sane managing to stay safe. Hopefully all this terrible disease hasn't come too close to you. We're still doing all right. Thanks for dropping in. We'll see you in two weeks.