 can really complement your reign so well. This is a recording of the finish of the link for April 5th, 2023. Yes, and we're talking about the Jerry's Brain Plus Generative AI or ChatGPT project. And it's what fits where, but we also just finished talking and sorry we didn't turn the recording on earlier. The book project that Pete Kaminski put in front of us that's happening on, where the conversations were happening on Monday mornings. Sorry, go ahead, Flancia. No, no, that's perfect, thank you. I guess recordings don't get the notes, but yes, fill it with the link in doc an hour or so. Right, no, about like the complement there, I guess where like your reign is a very link oriented, very much the topic of this course or the core motif. And I think it blends itself well, this experiment maybe to the crowd source approach, which is also I think key as a, or making it very interesting as an experiment precisely in the United Model's area, because crowd source and how we crowd source in a responsible way that respects the sources that actually deals explainable models or attribution and so on is very much key. And I feel like an Aura like approach, where like you will have some people contributing links and some people contributing notes, right? You can imagine like people going through your reign to put it some way out actually, explicitly and fill it in the notes in the notes. Of course, we will need conventions and so on to do so, but like at a very small scale, we have seen this in the Aura where like, the fact that someone wrote something, even a stuff somewhere, makes it more likely other people will go and write their own thing. So it's sort of like calling attention to something. And then it's essentially you will continue with approach that fits your reign and your role flows and other people who do the same and the complement of the combination of all these gives this corpus, right? Yeah, I'd love to talk about that sort of in more length and slower if we choose to go back into the topic because it's really interesting. And I wish the brain had an API because right now there's no way to go create something and then add that back into the brain. That would be really pretty phenomenal. I have to do it manually, right? Which actually sort of sucks because if there was a way of enriching the brain as I use it, like in the brain data file, then I could make it so that the data actually gets more and more useful. Oh, and I forgot to say, so the place we ended up on for the free Jerry's brain calls is an experiment where I'm gonna take one or several brain casts that I recorded and posted on YouTube, which have transcripts. We're gonna take those transcripts, map them to the thoughts that I point to as I'm busy giving a tour of some sections of the brain. So I did one called SNP, which is all about the 2008, 2009 global financial crisis, right? So I have 20 minutes, 25 minutes of recorded video and I can easily even by recreating the journey and through all the different thoughts I can point to everywhere it points to in the brain and we can feed that into ChatGPT as a small corpus and see what that does. So we can kind of start there. Right, interesting. Yes, so how about what I have tried so far? ChatGPT has this particular interface, right? Where you have to provide the content and so on and you can imagine, you know, there's three directions. One is a more ChatGPT direct approach where you will say this is like, you know, these are example links of a part of the brain. Can you complete more links or, you know, like, or write about these relationships, all these things, you know, or, you know, continue this tool, the braincast, you said, you know. And the other will be, you know, I guess go for a more like fine-tuning approach where you, I guess, living behind ChatGPT for a while, you know, we focus on, you know, like an open-store language model and fine-tuning that which has the different degree of like complexity but also like pay off. Yeah, on the API side, I heard that several times, Jerry, right, like if he's like, and there's no way the brain will do what you need. So we can scrape the brain. We've done a couple other things and programmatically I'm not nearly as smart as Pete and Bentley and Mark Antoine who have done some things to basically get data out of the brain through a couple of projects we did. One was called MemeBrain and another one was called Brainy McBrainFace, just because we needed a fun name. And so that we've got, you know, we've got that code. Yeah, and then I wanted to add a third project which is more me but fits everything we're talking about here, which is a couple weeks ago I realized that I'm the most cyborg person I know that I have externalized more of what I think than what I believe than anybody I know. And gosh, I think the future of work has a lot to do with becoming a good cyborg, whatever that might be because we are going to increasingly need to be symbiotic with our software if we're gonna be powerful. And I'm watching what people who have mastered ChatTPT are achieving with a couple of prompts and internalizing how prompt engineering works well. They're generating things that would take a human several months worth of labor to create as a great starting point and it's not a finished artifact but oh my God, you know, it's just even being able to start from that place is fantastic. So I created a presentation that I want to sell and make a living from called and originally I was calling it my life as a cyborg and I had a conversation with a guy who said great topic, boring title, how about something like Confessions of a Cyborg? So that's what I'm calling it and I own cyborgconfessions.com but I haven't put much on that. But I test drove this talk on Sunday morning because one of my buddies heard me mention this and said, hey, would you do it for our group? So Sunday morning early, I did the talk for like 40, 45 minutes and it went really well. Like it worked just great as a first pass including just as a first pass. And then I'm really interested in the way that this symbiosis between us and software fits into all these other things that we're talking about and how it all works. And then last thing for my update, Paul, Rony and I keep having a little trouble synchronizing to get the podcast restarted but I'm still hot and interested in like getting hyper talk moving again as a podcast and also because the artifacts I want that podcast to leave behind online are exactly the kinds of artifacts that we might use and reuse for a Neo book that we can involve in all these kinds of projects. Like that is just a podcast that happens to be a series of interesting conversations that would feed the whole medium that we're trying to evolve here. So that's it. But those are the top of mind things for me right now that relate. I mean, I'm personally interested in all of these. I need to read more about like these people that you mentioned, I guess. Yeah, on the podcast in particular, I feel like, I mean, I'm pretty interested in that with the link to the book, in particular, because the conversation we have here seem like they could yield at least some of the primary matter for a podcast, right? After it and so on. So I guess I'm interested in sitting out on these pipelines. On the podcast, I also have this like projects like drafted but shelled essentially, which will be like, can we get a draft, a first draft of a podcast, like more easily with a group of people by just going from a chat room with voice messages to like, our composition of those. So essentially like maybe develop like a simple tool so that we can like send back and forth ideas asynchronously. And then, you'll get the first draft of the podcast this way. So we have, I mean, we've got chat rooms effectively on MatterMost right now for all the different projects I named. So we could easily do that. And you mean just as a way to organize, as a way to sort of structure things? Yeah, I mean, in general, like what I want to do is do this for a telegram, which is a mobile news and like what I use, but you know, I will look into MatterMost, but essentially it's like a side project of a side project as usual, but it will be like, you know, making it easier for people because a lot of people like, I know people who just have like rules but they send back voice messages, right? With friends. I know having like a vote that says like, I just, you know, I'll just like generate like a weekly podcast episode out of what are you sent. You tell me that, right? That sounds great. And just using chatGPT to sort of structure some of that or summarize or assess it would be great. That'd be awesome. There is some work on generated a podcast, which is interesting. Exactly. It was a pro concept or like commentary, but I think there's going to be like interesting, you know, the interesting ones are the ones where like you won't tell. Yeah, exactly. And also the way that Paul and I are trying to frame hyper talk as a podcast is actually as a collection of different podcasts that are hosted by different people, some of which might be four episodes long. Like, so what we're looking for is people who would like to pick up and say, hey, I have an idea for a topic that I think will last six episodes. And then they go record those episodes. We wrap it in a way that it looks like part of hyper talk. And then we, you know, Paul and I sort of act as both producers and also hosts of episodes and guests on other people's episodes or whatever. But in that sense, the calls we're doing here, if we made them a tiny bit more formal, could easily be podcast episodes ongoing. I mean, depending on what, you know, if we plan these calls a tiny bit more, I really like how unstructured and loose they are. Right. But there's no reason this couldn't simply be the kinds of things we've done and talked about in Freedom of Fellowship of the Link are really good fodder for hyper talk. Yeah. I honestly think that if we went back and like, you know, we took like recording projects we mentioned and recording threads and said, you know, we're going to discuss these three in order every call. I think we will probably be most of the way there. Yeah. Just so, yeah, yeah. So hold that thought for a moment and... Yeah. And I'm really into the making a podcast by the way. Oh, good. Okay. I love that. And I kind of want to put in the back of both your minds, like if you were made King tomorrow and could and had plenty of time and resources, which we don't, none of us have, but you know, pretending that, what would your podcast be about? What would like, how would you sequence like a half dozen or 10 episodes? Who would be your guests? All that kind of thing. King of podcast then. Yeah. I thought you went just King. Sorry. It's a very limited kingdom. The domain is very tiny. Okay. Now it's fine. You're already the King. Yeah. Yeah. Exactly. What are you using your powers for? It's like the King of Sea Land. You own a little oil ring that's kind of old and decrepit off the coast of England. It's not a lot of territory. And by the way, I read it, I was reading an article about Gerard Piquet this morning who was in reinventing soccer and something he calls the Kingsley, which is changing all the rules to make soccer, football, fit better the TikTok era. And I'm like, oh, that sounds like a really bad idea. But anyway, it's intriguing. Yeah. Reform is always an interesting thing. Yeah. In baseball, they change a bunch of important things like the size of the bag. And it's like, wait, what? And they put a clock on the pitcher. Whoa! They can't stand on the mound and scratch their crotch for a couple of hours. And football now has bar. Yeah. And I hold these conversations, you know, like, whoa. I would put it like, I don't like bar. I'm like, what do you don't like? The fact that it's rational and that you have the goals, it's like, well, it takes away something. It's like, really? What does it? I don't like it. It's not too fun. I don't like when a really, really awesome goal is disqualified because somebody's shoulder was an inch too far over the line. I'm like, you know, can you give it to them? Can there be like a little grace there? Right, right. They should have one, they should have one goal a year. But it's like, you know, the goal was beautiful right now. Exactly. But like, yeah. It should be like a beauty rating that influences the VAR decision would be. Right, right. People can vote. That would totally work for me. Yeah. Yeah, so I do have like a plan for a podcast, essentially. Yes. So yeah, we'll have to share. Have you put this in a GORO? Does it have a thought in a GORO that you can share? Yeah, yeah. Well, so what I've been doing, I mean, just I don't want to take too much time, but essentially what I've been doing is trying to develop fluency into a pattern language. So yeah, so it's number, yeah. Oh my God. So I have something else to check in on about that. Keep going. So essentially I have like about like 51. It's like Chris's video is going a little arrived, it's online. Yeah, yeah, it's like everything's like. So I think, you know, going back to the Neobook and so on, I think one of the things I thought is like, you know, pattern language as an example for like a possibility of chapters and hypertext at the chapter level, right? Which is why pattern language, the original one was a bit like. So yeah, so I have like about 50 patterns. Depends on how you count. And my default is I'm gonna make them into like books and like, you know, like essentially, you know, slowly and essentially anything that can map to a number sequence. For example, like podcast. That will be my default source of inspiration, essentially. And there's a way of easily imagining creating a podcast sequence where each podcast episode, in fact explores and helps you record any of the patterns in your pattern language. Exactly. Oh my God, oh my God. Okay, we need to talk. Yeah, yeah, because I found it very like very fruitful and like, you know, of course it's just inspired, right? Right, right. The people proceeded to make it beautiful. And like, yeah, but then I was there. So yeah, and it just seems like a generic. It's sort of like going back to generative, I guess. It's a bridge for me in the sense that I mean, I would be happy to talk about this a little bit, you know I'm also like a bit, it's interesting, but like essentially these numbers now means things to me. So, you know, when I, you know, it's just the fact of being able to like really commit to memory especially within a number and like a topic or a pattern. It's sort of like it feels like a mind palace. And like I've been using in this quite a bit in all the last year or so. Very interesting. I'm attracted usually more to the names of patterns than to the numbers of the patterns. But same thing, it's a mnemonic. Chris, we keep talking and not letting you in. Yeah, that's okay. It's all good stuff. I don't know what my podcast would be. You know, it changes every two weeks. You know, what's the new topic of the week or the day? Right. Although if I could and maybe it's related to a different related thing in terms of playlists of things, particularly because you were talking about a book or books, if I could wave my magic wand and say here is a website with a bunch of interesting stuff on four topics, let's say of 5,000 topics it could have crawl this site and spit out, you know, a book version of the five or six topics. And then allow me, because it's pretty easy within either caliber or tools like sigil will let you kind of rearrange chapters within an ebook to make ebooks. And then, you know, you can hit a button it's on your Kindle and you're ready to go. But that could be a super cool way of here or here's a list of five websites, pull out all the data. Now, and it may be one thing if it's like the New York times and you try and pull something from a really massive store but let's say my own website, you know, and even that has over 60,000 posts on it. But if I subsection it and say, I want these four topics, give me that as a book and then I can reorder it and then spit it out and, you know, create a pattern language to do that or to say, go to Wikipedia or go to the Agora and give me this stuff and then have a base of material you can play with, that would be awesome. Right, and it ties in really well to the generative aspect if you want to generate a glue or maybe in this first pass when you are like just like a tiny thing to generate connectivity essentially and say this is where, that sounds amazing. I don't hear you, Jerry. Oh, no. Sorry, I'm sorry, I muted myself. A brief check in on pattern languages, two different things. Yesterday I had a repeat call with a couple friends, one of whom is Daniel Lindberger who was part of the Liberating Structures pattern language collective and the other one is Marie Bierida who long ago worked at Qualcomm but she's really interested in unmanagement. So unmanagement, writing a pattern language for unmanagement is kind of the umbrella topic for our calls and that's moving slowly but it's interesting and Daniel is, I think, skilled in writing pattern languages if you want to talk to somebody who does that. But then two weekends ago I spent the weekend with four other people, one of whom was the word Cunningham, the inventor of the wiki, all of the other people were deeply, deeply enmeshed in Christopher Alexandrian thinking and pattern languages. One of them was Michael Mahaffey who wrote a book about Alexander and does consulting around pattern languages and all that kind of stuff. It was a very interesting weekend and I'm happy to sort of report more about it but one of the angles we were taking which is really strong for me is how do we make pattern languages more accessible, more better known and more useful. And I think you might have heard me say something like this before where the example I use is the one, two, four, all pattern from Liberating Structures. Have you heard me talk about that before? Okay, good. So that's what I mean by instrumenting or making patterns more useful. It's like, oh my gosh, here's a pattern that's instantiated as code and is in fact helpful to do a higher level of choreography than a junior facilitator might think of or be able to do that. But how do we do that for all these useful patterns? And then how do we take the process of thinking about some topic that has structure, distilling it toward patterns and maybe here chat DPD is super useful as well because I easily imagine giving chat DPD prompts that say, hey, the framework for my pattern language is situation, a complication solution, examples and take these free texts that Chris just put on this webpage and create 10 patterns that follow this pattern frame. That's probably pretty doable, right? Or even one better is go into my notes and find things that are linked together on a certain topic and generate something out of that that I can then edit, which is what everybody's gonna want. And that's lately within kind of that broader note taking and wiki space. I think that's the UI hurdle everyone has is I have hundreds, thousands, millions of notes, but the hell do I do with them? How do I move them around? And it's, you know, I think one of the benefits of the physical old school index cards is you can move them around and play with them that way. But when you have it all as digital material, I don't think any of the note taking apps do even a half a job of allowing you to kind of go in and move things around and play. And in general, it's hard to do special reasoning or apply or pattern matching and so on. And it's interesting because that's, I think to me, that goes back to the embedding problem, which is so key to like, well, I mean, I don't want to just imagine at the world level, but, you know, embeddings are like crucial at, you know, like machine learning. And like, essentially what we do when we visualize, you know, organizing special in 2D or 3D and written about that is like producing and embedding to some extent, you know, and I could imagine, you know, like you mentioned the links and this goes back to something I've been trying to do for long, but, you know, on and off as a side fun project, which is like- Is this a side project of the side project? Yes. Yeah, I should have, like, yeah, I actually have like a lot of projects and it's all like, yeah, so. Sorry, keep going. And sometimes, yeah, most of them won't happen, of course, but yeah, nice and fine. And I can visualize in the hour, right? You know, a way that makes sense. It is a hairball currently in any, and you know, that happens with many large graphs. And like, I'm sure, you know, like, well, I'm pretty sure that the crisis, you know, 60,000 post-site probably has categories and, you know, probably it will look better, but I will still imagine that when you say things are linked together, there's going to be complicity in extracting the old clusters to put it some way. Things that can like, you know, be taken apart and be understood and also be put plan or to put it some way, you know? So I wonder if there is potential, therefore like this sense making toolkit to put it some way where you could say, these are all the sources are linked and we're going to share to some extent like the procedures to mix, you know. That sounds awesome. I subscribe to a few too many sub-stack pubs, for example now, and the good ones, I read pretty much daily or weekly or however often they come out and then I weave them into my brain, which is something the authors are not doing. So one of the things I can't stand about sub-stack pubs is that it's just a stack of newsletters you sent out. And sub-stack does nada, and I mean nada to offer context or power tools or anything. It is extremely primitive technology as far as I can tell. And so some of these authors work, I have like Heather Cox-Williams, so Heather Cox-Richardson. I read her daily and we've, you know, every day how it all fits and a few others. But that woven context is part of what we want to be able to share to tell our story, right? So could that be automated? Can we get the tools to do the weaving for us or even on an existing corpus? That shouldn't be that hard. Hey, here's all of Heather Cox-Richardson's newsletters. Go find some links and make a mind map of how these things, what the dependencies are. That is pretty cool. Or even I want the edited version of what Jerry's read this week and the links he's made, and maybe just the top 10% of those links and how they interact and have that. And that's the sub-stack I want. You know, great as Heather Cox-Richardson's stuff is. I wish I had time to read more of it. Right. I will say that it's, I think the same problem affects social media, right? In the sense, you know, like another timeline. I mean, of course, like MicroPulse and like so on, like, I think of Twitter and master mostly, but like, essentially, you know, very time-based, sequential, not linked, very hard to add context. And I think the same opportunity for like building tools like Complement, it's sort of like this huge space that is completely left open. Maybe because corporations haven't found a way to monetize it, or maybe it was against monetization in some cases. So, you know, we keep having these, you know, like very linear experiences, you know? Scroll through a screen, right? Read this post one after the other. I find most people are having a hard time imagining something which I call leveling up media, like what's beyond what we've got now. And I'm not seeing a ton of creative experiments. There's some old tools that seem more sophisticated than new tools, you know, like Tinderbox or DevonThink, are, you know, visual mapping of articles and Stephen Berlin Johnson's a big fan of DevonThink because it suggests connections across documents and he loves that, right? He uses it kind of as a brainstorming bucket. But that's just one feature that you want from these kinds of tools. It's like a one feature, it's a one-trick pony that software. So, here I wonder, you know, you're going back to Matthew and Peter's? Yeah. I think many work on like the for thinking map whether we got to an conclusion we could get, you know, next time on which are the best in each category. Because I think one was like visual thinking, I think, or, yeah. So I wonder if we could like get this variation from, you know, or there's tools that we can somehow reuse, I don't know. And I think the mapping project that they're doing fits very nicely into the infrastructure we're trying to generate. So, how does that work together? Sorry, just go ahead. The part of what's missing I think too is we have all these linear inputs that we read and you may absorb bits and pieces of things as you go by but generally you're going to see the same things repeated over and over again for, you know, if you're on let's say old Twitter with their algorithm it's going to rise up to things that think she want to see or that are being repeated multiple times. But what's missing and maybe it's at AI can fix it or help it. But we are missing the ratchet for how do we take the pieces that we know and have context for? And instead of rereading them over and over and over again because I guarantee over the next year we're going to see rehashes of the same five Trump things that we've heard in the last day and a half thousands of times. And I don't need to see it thousands of times. I already have a base context of what this thing is. And there are people who are ignoring it actively and won't have that context. So maybe when they decide they want to dip in six months from now they'll need that context. So how do you give them that so that they can then ratchet up to the next level and build on it? So you have the brain or I have a set of notes that is a preexisting mass of data. And I want to be able to plug that in and say, here's you have my context now system. And then I can go to sub-stack and say, I want to see interesting things in these areas that are things I haven't seen before that will allow me to kind of get to that next level up. These are cool things. Sorry, go ahead. Briefly, and I think this builds sort of goes in parallel to what you were just saying, Chris. The idea of generating pattern languages by looking at your work or something like that. One of the things that was coming into my head from that was, yeah, and how do we sort of reduce duplication of everybody's patterns, everybody's genius patterns? And there could easily be an application of chat, GPT or something like it to say, hey, who else's patterns does this look like? And can we consolidate, date them or rationalize them or blend them? Such that a piece of my pattern language winds up being patterns that other people have written, but I love them because they encapsulate better patterns that I came up with on my own and then discovered we're already somewhere instantiated in some other place. And then we create kind of like meta or polyglot pattern agglomerations, pattern pigeons maybe, or pattern trade languages. Pigeon language, yeah. There's dialectic. Pigeons, yeah, yeah, yeah. P-I-D. But that sounds really interesting to me because one of the things I don't wanna see is each of us inventing all of this all by ourselves and creating these very elaborate, but distinct and separate worlds of thinking. Makes complete sense. Thank you. So, okay, that looks very interesting life, Rajit. I'm like, I added to the notes. Thank you, Grace. On the direction of, bring your own context, which is what, you know, how I could summarize one aspect of what you were saying, Grace, I guess in my mind, in the long game to put it some way is in the user agent. I don't know if you've gotten to the same conclusion, but like thinking, in particular thinking, maybe a bit preventively or, you know, cautiously, defensively about, you know, how some entities, you know, are different entities that are like sometimes targeting for control of the internment, to some extent and so on. I sort of feel that, you know, in the worst case, we'll keep receding into like, you know, what there is no control. And in the end, it may be about the user agent, meaning, you know, do we own the browsers we use? Do we own the actual tools we use to interact with our sites? And once we get there, you know, that's essentially where your context will be. So to some extent, why doesn't the internet browser do this for you is a question. Before that, of course, an extension, to some extent, could actually implement this as a, you know, V1, but potentially in the future, because extensions unfortunately are sort of like, depending on browser APIs, which can be being changed for a variety of reasons, it seems like owning the browser may be the only long-term insurance policy, right? And in that case, then you could say, well, any site you interact with, and you can imagine this going back to the cyber direction, you may have like a generating model or agent assistant engaging in most of your browsing sessions. You go to a site, you want to find something. Now you have control if, that's what you have, right? Yeah. Find something in the settings in Amazon, right? I think this is something that is coming in the next few years, probably, where like we will actually have like, you know, these cyborg, more like cyborg browsing sessions. And here is like where, you know, that actually leads in the direction of having that, I think Chris, you know, having the browser also react, negotiate with the sites, you know, what you actually want, your algorithms, based on what you already know, I have said it interests you. In some sense, I think Chris alluded to this a little moment ago, but in some sense what we're describing here reinvents the algorithms that are currently part of the platforms we're on, that are being done without our knowledge and behind our backs and often to manipulate or addict us. We're sort of describing some parts of crowdsourced algorithms and means of de-duping our feeds so that we see what's new and what's different and not the 15 repetitions of the same set of six points and also de-duping our insights so that we can crystallize knowledge instead of each sort of running off in our own directions or something like that. But that feels like the part of the feed. Oh, and then there's context collapse. Chris, do you want to talk about that? Well, I want the opposite of that. So you can go to social media and jump into a conversation and have no idea where the two sides are coming from. In fact, maybe they don't even know themselves where they're coming from or each other is coming from. Usually most people at least have a good idea where they're coming from. But then they have a conversation and a third person may come in with no idea of any of their history or connections or any of that stuff. And the whole conversation collapses down because they're all missing that context. So how do you come in and be given even some quick prompts of the context you're missing, where people are coming from so that you can participate both meaningfully and in a kind fashion. And then allow you to actually have a context ratchet so that once you've had that conversation or that thing, it's now part of your context and you can move on to the next level up. You know, simple things like poverty in America, we can't fix and it's worse and becoming even worse because the two political sides want to blame different things for it. And neither one of them has any kind of historical context really of where we've even been to have the context to fix the problems. And so more often than not, we're treating symptoms rather than treating the disease. And if you could, you know, up that level of the context ratchet, which is, you know, the only way, the only words I've got to describe what this thing is or at least what seems smart, there's probably a better phrase for it, but how do we do that? So that you can take the two or three things, compare them and find the new idea and actually move forward. So this brings me back to something we discussed sometime before, which is like the idea of like translating between mental models, right? To like facility communication. And you know, like this speaking language or like, you know, like common, like this sort of like intermediate pattern language could be like in this direction as well. So I wonder, you know, in the case of that, of, you know, like political discourse and debate and so on. Yes, can you find words, maybe new words or like less politically charged words, which haven't been used before, but two sides could agree on, right? To some extent, or new framing and so on, I don't know. Too often I think we find whatever the new words are, become the new weapon, at least in the political sphere. So, you know, I was looking at some myths of American history and one of the ones, it's in Kevin Cruz's book, new book, Myth America, MYTH and they talk about the idea of American exceptionalism. And in the 90s, the Republicans beat the Democrats over the head to the point that they quit saying it and they capitulated and said, yes, America is exceptional without any history of what's going on, just because it was politically expedient to do so. And I would posit, I don't think it's mentioned in the book at all, unless it's much later than I've read to this point. But the phrase woke is a new word that essentially is that same idea of America, the opposite of America exceptionalism, we're not exceptional. And these are the areas we're specifically horrible at and we could become better at making things more equal, more just, more, nobody's gonna pick themselves up by the bootstraps, it's an impossibility. But the word woke is now the same, it's the same meaning as saying America is not exceptional two decades ago. We've just invented a new word and guess who has picked it up, they're using it now as a bludgeon in every marketing, everything to raise money or to say anything you don't like if you're a Republican, just call it woke and suddenly it's a bad thing. And so that prevents the two sides from ever coming together because that whatever that new word is probably in the case of woke, it comes out of African American culture which makes it four times worse than American exceptionalism and it'd be willing to bet your dollars to donuts. No one who uses American exceptionalism unless you're a historian realizes it, it comes out of early communism and apparently it was a phrase Lenin used early on. Oh, seriously? Yeah, and I was like, when I read that I was like, that's super interesting. So the history of it goes back to the 1920s in early socialism and then Lenin picks it up. Although I have found it doesn't use the phrase American exceptionalism specifically but Henry Adams has a book on democracy from the 1880s where and it's a fiction piece where he's got, I think it's a baron from the UK talking about America and how they think their, you know, the idea of American exceptionalism is there and I have a strong suspicion that when he won, he died in 1918, I think, and his education book became a Nobel Prize winner the year after he died. And so I think a lot of people were going back and reading his stuff in the 1920s and that's where the phrase got picked up and turned into this thing. So, you know, without that history piece of it, everyone's lost and then you're, you know... Important also is that the words don't just sort of idly show up and either pick up steam or not steam, they are the subject of intense pressure to use or not use them by political parties who are trying to make a point around the words or who are trying to diffuse a very effective word or who are trying to reappropriate a word or whatever. Like this is half, this is part of my junior history thesis, well, theory is that human history is a fight in the cockpit over the joystick of who controls our society during the next stretch of time. And both of the parties, usually there's like two, but sometimes there's a few more, think they're about to lose all the time. They get really paranoid and really angry and they take pretty extreme measures sometimes, which sucks. That's why human history is so lumpy. But these words are crucial because the narratives in our heads are the tools for winning control over the policy over everything. I'd nine times out of 10 too, our general policies are not so horrifically far apart. I think where some of the social media stuff drives us away, we need a little more manufactured consent or at least manufactured consent towards the middle consensus. Or maybe that's what the agrar should be doing. No hope. I don't know how you pull it off, other than getting people back together again and disintermediating the algorithm. Yeah, my hope is that, okay, so the short of the working plan will be try to organize around like a fuzzy clusters, which could be like discoveries in pattern languages or some like other thinking tools, 325 projects where there's like high leverage opportunities. I think it's like, you know, what we discussed about like writing context or like there's a lot of these projects we discussed which seem to hit that open niche issues. And like then essentially go after some of these projects and try to set up a reinforcement loop or self-improvement loop, right? So that essentially people are more willing to maybe compromise the open-minded in some context because that actually has some concrete like a result. Like, for example, like access to our corpus or access to our commas to put it some way. And, oops, I just did that wrong. And that's what the fight over books in our schools is right now, right? That's very minute. It's like what is allowed to be in the corpus because it infects the body politic, right? Right, right. And, you know, the problem too is, you know, most kids, kids are reading, but I don't think kids are reading nearly as much as they did even a generation ago. So, are you counting all of their texts and instas and TikToks? Yeah, but that's a different read. That's a different. Oh, that's it. But it's words, right? It's words on, it's text. Oh, well, the, I would say the level of the conversation is probably a whole lot lower. Whereas, you know, reading books is a much better thing. And you're gonna discourage some books over others or get back into the, you know, somehow we have forgotten that book burning and book banning is a bad thing. And now it's suddenly, it's a doable thing in the United States again. Very remarkably short memories. Like stupidly short memories. Yeah, yeah. You know, and that's one of the things that if you were a Republican, who's, you know, banning all these books, the one book you probably, the one thing you want in your history books is probably, hey, burning books is bad. And you know, that's something that's in the history books. Although even in your American history book, it's at the dead end and you probably don't ever get that far in sixth or seventh grade. Right. You don't read far enough in to know that Nazis are bad. And in some school systems, I understand like Germany lives and relives and thinks about the mess they made in World War II all the time on purpose because they think that that's the way to prevent it and they still have neo-Nazis. And Japan doesn't teach World War II pretty much in schools. It's my understanding. They've just avoided. So there was a moment where something happened. They were shooting a documentary or a movie or something like that. And the Japanese actors who showed up had no idea what the movie was about. Because the movies was set in World War II and they were like, these are new events. What is this? Wow. Kind of crazy. Yeah. I'm gonna have to drop off the movie today. It was also during, no, I think that was also during a time to when the emperor was generally infallible. Right. And the Congress was making decisions there on Japan's behalf that the emperor then rubber skanked. And nobody was willing to say the emperor's not got any clothes on while they're doing it. So it's also in a shame-based culture. We don't want to admit you were wrong. And particularly when, okay, we, as a culture, we seem to have moved past that. So we won't acknowledge it. Versus, and I saw an article a couple months back too that said it's much easier for Germany to admit wrong with respect to the Jews because there are relatively few now left in Germany. So it's not like you're living with, the people still as your neighbors that you have wrong versus in America, it's much harder for us to say, yeah, I think all generally agree slavery was bad but we won't let it go because they're all still here with us. Right, right, exactly. Thank you very much. Yeah, I'm not sure how to stop the recording.