 This is the Neo Books Call for Monday, February 5th, 2024. Turning on captions. And we're all set. How is everybody? Californians, are you like wearing fins outside? Are you good? You all right? You're pretty light over here. I'm in Palm Springs area and people are getting some rain, but the real downpours on the coastline. Yeah, April was in Santa Barbara for a talk last week or week before last, and they're like major photos of flooding everywhere and, you know, crazy. I was expecting it to rain in San Diego. I think it's sprinkled a little bit, but it hasn't really started yet. I think that the wave made a U-turn and sort of turned back up and really slammed into the coast above, just above you. Something like that. It looked like a starry night painting. San Diego is up this evening. You are, it's coming. Oh, great. It keeps shifting back. First I think it was Sunday and then Monday and then now Tuesday. Dave Whitzel, what is the NOAA network? I don't think I've seen it either. This is something Brad DeGraph is working on with Lake Day Waterbury and I don't know, a few other folks in there. I'm still groping it, but trying to map the people who are thinking about regeneration and then map them by bioregist. And then they're using, I thought, some interesting tools. And I think they've been scraping LinkedIn so they have kind of a, you know, it's like a rich enough database so you can kind of play with it. What's it called, Dave? It's in the chat there, who knows? Or let me put it, I can put it easy. Who knows, collaborative mapping of global regional expertise. Yeah, yeah. Global regeneration expertise. Sorry, I misread that. Yeah. We'll have to get these folks talking to Vincent. Yeah, I think maybe they are, I don't know. I'm always a little confused over who's talking to who, but, and then it's just, you know, because I've always wondered, it's like, you know, if you could just get into the back end of LinkedIn, right? It seems like you could do some of any amazing things with it and, you know, they just close you out of, but this seems like- I've seen people doing that. I think it might have been this call anyway. There was a few people doing a semi-back end of LinkedIn, that's kind of interesting. I would love a social graph from my LinkedIn network. Right, well, and wouldn't you want to see it in the, like I would love to see the social graph of the regeneration network, right? Yeah, well. And they can probably draw a bunch of it. It'd be great to have sub-logic, you know, ideological operators on the whole thing. Show me the people who are in OGM and the regeneration network and Bing, there you go. Well, they don't want to let us do that. Well, so LinkedIn zealously protects its social graph, doesn't let people crawl it, does a whole bunch of stuff, but then they don't offer power tools themselves, which is like, hey, stupid, like you should do this. Why don't we offer them some power tools? That'd be good. And then that funds the OGM for the rest of forever. The rest of forever is nice. Yeah. Cool. We had said last week that we would, we started a conversation on collective authoring and we said, hey, let's pick that up and keep going next week, which is now. I think the topic is awesome. And juicy and fruitful for our various collective efforts, I think, because we're trying to figure out how to create shared meaning, how to propagate ideas, how to involve people in different ways from peripheral likes and thumbs ups to deep collaborative editing to shared manuscripts to whatever else. And I would love, Pete, I don't know if you have a starting point for the conversation that you wanted to sort of posit as a place to go from, but I was thinking maybe we could do a go around and see what questions everybody's got about collective authoring that we should address, for example, but Pete, if you wanna, and Stacy, go ahead. I was just gonna say that I left the call last week thinking that I wish there was a word other than collective authoring. That's a great start. Because it's sort of like combination of collective editing and synthesizing, and I think having the right word would be more inviting to participants. That is a terrific way to start the call. Pete has his hand up, please jump in. I was gonna say something similar, a bit different. Maybe different, but similar. I don't know. I wish we had a taxonomy of collective authoring or whatever, partly because I found out there were many more kinds, in discussion last week, I found out that there were many more kinds than I had previously thought, ranging all the way from we're typing on the same page at the same time or typing on the same page at different times or out to, yeah, I authored a book collectively. I have a chapter, she has a chapter, he has a chapter, and all of those things, if we call those all collective authoring, then, well, we need to, we need to, I would dig in and define all those and then talk about the individual ones. And having said that, the thing that most interests me is Wiki-style collective authoring, which is asynchronous co-editing. And that's where a lot of my energy is and where I would like to see more. And then, that reminds me, we left, I left with the assertion that, oh, it's easy and fun and productive to do that. And it was counterintuitive how that even works, like, what do you mean, everybody edited the page and yet nobody claimed credit for the page, doesn't make any sense. The realization I had right after the call was, it's easy and fun when people have shared goals. And in the old Wiki days, usually the shared goal was, let's make the information space as good as possible and let's make each page as good as possible. And when everybody shares that goal, it's easy to go in and edit a page with some more heuristics and discussion around that, but not much, it's really easy to go into a page and say, oh, I'm gonna make some changes in the spellings or, oh, there's another paragraph or two I could add or, oh, I need to have a whole section where this is actually a subject that has some debate and a diagesis and let me organize it so that there's the pro side and the con side and the neutral side and I like that. So then, of course, all of that is easy when you have shared goals and then getting to share goals is never that easy. So then the flip side, the reason why we publish books edited, I forget what the term was, but edited volumes. The reason why we do edit volumes is because the goal that we could agree to is let's publish a book together and then we didn't go into any more depth than that. I wanna say it my way, I don't want them involved in the way I say it and I certainly don't want to mess their stuff up. So I'm gonna stay in my chapter, they'll stay in their chapter. So that to me is a tragedy. So there you go. Awesome Pete, thank you. Great start, Klaus and Gil. Yeah, so I'm taking a turn, not from my thinking or so, but maybe the way I've been looking at this. There is a shift in the way that AI is being deployed because particularly with chat GPT enterprise coming online and the way this works is that companies are now developing a proprietary form of AI and chat GPT enterprise allows that by protecting the intellectual property that companies are developing. And so the learning is not shared with the platform but it is staying within that frame. So we just made contact with, we developed this networking meeting and there are several medium-sized companies engaged here. There's, I mean, there are in the millions of revenue but like eight digit maybe revenues and so these companies just don't have a lot of the resources they would need to really broaden out and connect across platforms and exchange information efficiently. And AI is filling an enormously important gap here. And so I tested, so I mean the process of setting up an AI for one of these companies and they are subcontracted with Siemens so they are pretty loaded, they have resources. And when I started a training process for that particular AI for this particular company, you realize that this AI really doesn't know a thing when you first start. You're really dealing with a blank sheet of paper. And then you have to fill in what you want the AI to know in order to deal with questions for that particular company within their frame of reference. You know, the things they're working on, the tools they need, the relationships they have and need to build and so on and so on. That is incredibly powerful. My son worked for CEMSERA, he's the head of talent branding for CEMSERA. That's what they do, they are now getting into chat GBT enterprise. And this is a significant past a billion dollars in revenue. So that's a significant company. So that's how AI is being deployed. I look at Neobooks as a training to four specialized AI. So the information in volume one, for example, now coming up in volume two, really is an iteration of things you need to know to really engage in a meta-level perspective within the food and agriculture markets in the sector of the economy. So that's where I'm at. It's really, what does AI need to know in order to address complex issues? Where you are asking for advice and for help to reach across vast amounts of data and bring it together. And so each time I'm asking a question in the AI that I have set up, the original AI that I used to develop the Neobooks, I ask a question, I expect maybe 20% of what comes back. The rest, 80% is added perspectives that I wouldn't have thought of on my own, simply by the way the AI contextualizes the information. So that's sort of more the direction I've been taking. I think there is an amazing business opportunity in developing an AI and then putting a user interface in front of it. So you can have the AI in your fold, you put an interface into the front of it that gets you through a query and through a customization process which then allows the use for your clients or for that you can develop. That's sort of my thought process at this time. Thanks, Klaus. Just to connect that back to collective authoring, I'm sort of reading part of what you said as how do we collaborate with the AI as part of the collective authoring participants in that sense, right? Right. Yeah, I mean then the question is ownership, right? I mean, who really developed this thing and who answered and because you are, I mean, this has to be a commercial enterprise, right? I mean, it has to be a revenue generating enterprise because you will need resources, you will need to pay people and so on, but the potential is massive. I mean, in the food business, for example, there is this thing right now where they're talking about low carbon intensities course. So the biofuel industry is getting incentivized to have farmers crawl feedstock for biofuel in ways that reduces their carbon intensity and they develop the scoring system to do that. And based on the lowest possible score, I mean, they're starting maybe at 35, whatever that score means. And then this one guy is saying, well, I was able to bring it down to minus five and so he gets a premium for his crop. Well, that's a fantastic idea. And the government put money behind it. There's a bill, E41 whatever it's called, where the government is actually funding this, right? So that's already in place. It got snuck in without anybody knowing it. None of the NGOs even saw this coming. Well, why don't we take this one step further and develop a nutrient intensity score because the same core data can allow you to extrapolate nutrient density. Then you can communicate with consumers of saying, this has a nutrient level index, A, B, C, whatever. So by doing that, it's a complete game changer because now the public is becoming aware that there is a link between soil health and nutrient density and nutrient health which means cut health. But you would need an AI supported tool because in order to make this affordable and cost effective for smaller farmers, that's just one example in my space, right? In the sector of the economy, there is so much opportunity in deploying AI in ways that it's just completely going to be a game changer. Thanks, Klaus. As we go into this thing, maybe I could ask you to focus on is, in what different ways would other participants in your ecosystem, in the conversations you're having and the things that you wanna do, how would they want to interact with the information you're providing? Just as readers, as audience, or as co-authors, as commenters, as critics, as whatever else, like what are those, what are those types or qualities of those interactions because that will dictate a piece of what we're looking at on how do we, how do we author, because to me a comment is, that's also authoring, commenting on a work, even though it seems like it's outside the work, is also interesting. And I think we're trying to facilitate your publishing a lot of things so that they look like a sub, so that they're subsec pubs, you're doing podcasts sort of things, all those things are elements of collectively offering in some way. Thank you. Yeah. Yes, to the enormous potential, I wonder if anybody's asked the AI's about the ways we can use the AI's. Last to your point on soils, I was stunned to hear the Secretary of State talking about the importance of soils in global sustainability regeneration strategies, kind of remarkable that it's penetrated to there. I just wanted to comment real quickly on the thing Pete said a while ago, I think it was Pete who said about a taxonomy of neo-boki approaches. I know in the stack of projects that I'm looking at, there are very different levels of collaboration. I'm looking for different projects. Some are like everybody on the keyboard together, writing in a document, some are anthology collecting pieces from different writers, some are having open public comments and editing revision of books or chapters. And so I think that taxonomy would be really helpful both for our workings, but also for the invitations we're making out to other people. That's all. Thanks, Bill. In the early days of group wear, there was a little two by two matrix that was synchronous, asynchronous, in-person and virtual. I think that those are the two matrices. Pete, does that ring a bell from back in the day? And people used to sort of say, oh, okay, so forums and discussions are asynchronous over here, but not in person. And then like real meetings are in person, et cetera. I think we need a more elaborate, more interesting, more useful version of something like that probably. Other thoughts on the puzzle? More refined and fit for purpose because our purpose here is different than what that was back in the day. Yeah. That was just a really simple attempt to differentiate between various kinds of group wear where nobody understood the term. It was just a first cut at how do we do this? Dave says, perfect modularization. Yes, totally agree. We've talked about nuggetization. Then we had a side discussion about, hey, is the word nugget the right word? We have not replaced it quite yet, but the whole idea of if we want people to participate a lot, how do we create modular reusable media that enhances that kind of participation? Because if you publish a book, the way people participate in books is they write reviews and post those separately or they send the author a comment or a critique or a correction or whatever else. That's a really blunt, not very collaborative way of creating ideas and sharing ideas. Other thoughts? So yeah, I'm still getting my head around where you guys are going with this. And so my question really I guess is around why? Like why are we doing this? And so for me to answer that question, the idea of a new way of creating content that isn't for the purpose of creating content, but for the purpose of knowledge, which are two different things. Nowadays, most of the content we create has nothing to do with knowledge. And how do we create knowledge in a way that is open? Meaning it's not my knowledge or my view of things, but a bit of knowledge that I have that I think has some legs and I'll put it out there. And I don't feel ownership of it. I just want it to be out there and for others to validate or improve or do whatever it is that needs to happen. And how do we build a foundational piece of information that people can add to? Because every book we write, every book we read, every book is just 90% fluff for lack of a better way to put it. We're filling in the 300 pages we need to fill in and then there's maybe four or five good ideas in it. The ideas often get lost in the book and get characterized as somebody's idea when in reality, whoever wrote the book probably talked to a thousand different people to come up with those ideas in the first place. And so for me, I'm wondering how we have a different way of looking at knowledge information and how we take it to a new way of publishing. And I don't even think that that's the right word. A new way of disseminating this information that is truly open. So that to me is, that's what I would define as my why. Does that kind of relate to what you guys are talking about? It's a very different space. What were we? I'm totally on and I'll take a swing at answering it. I just described the traditional process of how books get in the world and how people reply to books by publishing a book or a book or a book. Publishing a review or commentary or whatever or maybe there's a place for a comment thread on a bulletin board somewhere where the author is or something like that. We were using Massive Wiki to create the Neo Books in part because the Massive Wiki is built on out of markdown files on GitHub and GitHub buys us version control and community editing, collaborative editing and a bunch of other stuff. But two of the big wins with GitHub are that every version and every change that gets submitted gets kept and there are several different formats for collaboration inside of GitHub. Fork and Pull is the most famous one which means you can go look at a repo, fork it to your own workspace, make a change and then suggest a change which the original author can pull into the main line or you can fork and wander off and just do something entirely different with it, et cetera, et cetera. And those are, they're baked into the platform that Pete has chosen as a starting point and they're pretty geeky. So they're not like what Muggles use every day. Muggles are used to going on AOL and typing into a forum or a comment thread or something like that. And that's a piece of how they interact which is evanescent, it's fleeting, it goes away. You could easily imagine a whole bunch of different ways probably way too many ways to add interaction, interactivity, reusability to a nugget like a markdown file on a GitHub repo. You could connect that nugget to Hypothesis which is an open source comments engine that preserves URLs for the comments and allows you to be part of a community that's basically building ideas across works inside of Hypothesis. We're fans of Hypothesis but don't really use it. Massive Wiki is wired to actually prompt Hypothesis to show up so that you could go use it and make comments on a page, that would work fine. We could attach and implement a comment system like Discuss, D-I-S-Q-U-S, or Discourse which is a threaded forum discussion board. We had a Discourse server early in OGM. We could create some other form of more organized comments that go. Or we could rely on Wiki Sociality which is where Pete was pointing early when he said it's really important to agree on purpose. Why are we doing this? And what are our intentions here? And what makes a good Wiki good is that social agreement because otherwise everybody just does whatever the heck they want and the Wiki kind of falls apart. But a good Wiki winds up setting up some norms, practices, and then templates or sort of rhythms that make the Wiki as it grows and as more people contribute still really vital and interesting and big and good. And so a piece of what NeoBooks is counting on is some piece of Wiki Sociality and Wiki Dynamics which are its own little thing. Wiki Dynamics are really different from using hypothesis, are really different from blogging and using TikTok and doing duos on TikTok, for example, which is another form of commenting and collaborating. So I think what Gil is like, well, how do these all fit in a taxonomy? I'm interested in that as well. I don't think we're frozen in place without a taxonomy but I think the number of options that we have is too many and we need to figure out are there different contexts in which certain options are better than others? Which ones do we wanna start with? We're sort of starting with a few by default because we're doing markdown files on GitHub. So by default, there's a couple of things right there that we fall into. But I think that's the, we're doing this because we wanna create the thing you were saying on our last call which is, what's interesting here really is not the book, it's the lively discussions in the community that keeps improving the resources that it has. Where we end up learning how to share ideas and improve those ideas. Absolutely dead on, that's the thing. Now the question is, you could imagine having six people collaborate with you on co-authoring the text of a nugget, it's really hard to imagine 6,000 people or 600,000 people doing that. So where do you put the gates, the filters? Whom do you let where, these are all just variables and it's all just software. There was a company called Mixed Ink I'll put a link in the chat in a moment. There was a company called Mixed Ink that went away but they had a piece of software that would actually let a bunch of people collaboratively edit a document. And at the end, when you saw the finished document you could trace which phrase came from which original author. So as everybody copy pasted, if whatever they kept preserved the original owner of the original writer of that little stretch of the document. Very interesting, it was like an idea posing as a company. They died pretty quickly I think but the idea was kind of cool and I have not seen anybody emulate them although you could back into that information from the GitHub version control, et cetera, et cetera. You could reconstitute that feature if you thought that feature was really important to have. There's also things like Eric Eugene Kim created purple numbers. I think it was him on Wikis so that you could actually have sort of paragraph or sentence level addressability inside a Wiki instead of saying this page you could say this paragraph on this page. Even that's a really interesting affordance. So sorry, riffing on a whole bunch of different things but the plate is set richly for modes. It hasn't proven to be easy or contagious to do this well yet online even though we're kind of all online doing something all the time. Instead, we're busy writing in on email threads where every post just disappears into the bit bucket basically is my thought. Sorry, Pete, I went on longer than I expected to but I hope Jose that helps flesh out why. And every time you say something, it's very funny to me you're like, gosh, I hope that sounds like it's orthogonal to what you guys are doing but I hope it's sort of in the picture and I'm like, no, no, no, you just described exactly our mission. So back to you in the booth, Pete, and then Rick. You're still muted. Thanks. Before I go, I wonder Jose, does that trigger anything what Jerry went through or have any questions or? It triggers that we're playing with the technology and I'm wondering if we need to play more with the needs. That's what triggers for me. I understand all that at the technological level. I understand most of it, not necessarily all of it, but I think those are just ways of doing things and I don't think that that's really the issue. There are lots of ways of doing these things but I think it's what's the model of the thing we wanna do that is gonna answer the why question that to me is, and then the technology will come about but until I understand the model, I still just see a whole bunch of hammers all over the place and us trying to nail a bunch of stuff and I don't know exactly what it is that we're nailing. Yeah, it makes a ton of sense. Isn't it interesting how often we all take turns saying something instead of taking turns and asking a question. I was actually gonna say something kind of similar, Jose, for better or for worse, one of the things that we do, especially Jared and I, is talk a lot about tools and you're right, it's actually, it's a chicken and egg thing because there are certain tools that we know about that enable something that's actually pretty easy to do that is impossible to think about unless you started doing it. So wiki is like that, right? And I don't wanna say that Wikipedia is the only wiki in the world or the only kind of wiki, it's really, really, really not. It's not a great example of a wiki, it's an okay example of a wiki, it's not a great one. There's lots of different kinds of wikis but a wiki is a way of working together that we haven't done a very good job of over the last 500 years with maybe exception of the Torah, which is kind of like a wiki. So kind of back to your question, I can actually kind of start with Wikipedia. So the idea of Wikipedia, kind of like you said, what are we trying to do? What should we try to do? What am I going to try to do? Or what is this team going to try to do? The people that started Wikipedia said, hey, let's build an encyclopedia, but let's build it so anybody can click a button on the page and make changes and click save and we'll see what happens. So I think it's kind of as simple as that. You decide what you want to build. I want to build a collaborative knowledge base about soil health or about 1960s muscle cars or whatever you want to build a knowledge base around and then you figure out what technology to use for it and more importantly than the technology, what does the technology enable that we can take advantage of in social ways? So Jay kind of hit massive wiki has a get back end and that enables certain kinds of, it's actually, that enables it, but that's not what we use. What we use is social conventions around the ways that we edit pages or who edits pages when or how we edit pages together. And the technical layer underneath it is really important but it's just an enabler for social activities. And then the social activity is coming to agreement on those coming to agreement on what we're building and then coming to agreement on, hey, when I swing the shovel this way, I think you need to make sure that your shovel is swung that way before I swing my shovel. You end up being a fair bit of that stuff early on, but at some point it becomes all of that fades away and it becomes the building of the thing. You asked it in an interesting way. Why are we talking about this in new books? So I think to answer that question kind of specifically, a core principle of new books is that they're written collectively or written collaboratively maybe. And we've said that, we haven't really understood what we mean by that. And so we found ourselves in a discussion. Oh, well, let's write a collective new book and I'll write one chapter and you write one chapter and then I'll be a collective book, collectively written book. And others of us say, hey, that's not collective authorship. That's serially or simultaneously editing a document or a book, but you didn't do anything collectively. So we're trying to figure out, I think back to taxonomy, I think we're trying to picture lots of different kinds of working together and I think it's fuzzy in everybody's brain because we don't have a picture or a list somewhere of, this is the wiki way to do it. This is the everything to way to do it. This is the Reddit way to do it. This is the TikTok Duo way to do it. So having said that, I've got some energy to work on a taxonomy of collective authorship. And that might be a fun thing to do in one of these calls. I think maybe we have too many people and maybe that's just a two or three person thing to at least set something up. So it would be fun to do that altogether. It would be fun. I would volunteer to do a little bit of that for homework for next week with one or two people. So that's my offer. Thanks, Pete. Rick, off to you. Yeah, I'd just like to rip off of what Pete just said. I was gonna say something else, but what I like about your idea is that our conversations are more theoretical. There are different buckets. There's the tech bucket. There's the Neo bucket and then there's the actual whatever people are gonna do. And having a clearer purpose about why are you doing call or three? What degree of call or three? And what's the outcome? And one of the outcomes, I mean, what little I know of Klaus's book and what little pieces I've seen of it is we'll have a nonfiction thing. And I don't know enough about your work class to say what it is in terms of how it's used and what the output's out is one of the most important outputs is how can we use the technology more effectively to create ongoing learning communities? And going off and reading a book by yourself is one thing, but if you're actually using it for creating learning communities, then I think you would design them differently. And I'll just throw a few ideas out. One is where people are sharing their own personal stories about the work they're doing and that the purpose of the storytelling is to engage other people and getting their reactions to the story of what's told, what they got out of it and does it inspire them to tell their story? So storytelling, I think is incredibly powerful. Another category would be more of a Socratic one where you have something relatively brief and it's actually designed to set up synchronous and asynchronous learning. You know, one of the things about the email threads is that I can't keep up with them. You know, I've got so many other things and I pop in now and again. There's nothing aggregating in a way that it's easy to come back to. And how can we use social media to actually use some of that huge wasted energy that people put into email threads and seeing whether you can have a better purpose? So to me, it's a question of how do you create more of an ecological framework of learning, of transformational and collaborative learning? And, you know, I think the technology is fantastic. Well, we can deal with it, but what I'd like to push the needle on is we're doing a lot of talking about it rather than doing it. So I really appreciate Pete's offer of doing that. I'm quite happy to add my two cents to it and whatever way would be helpful for what you're thinking of doing. But I would, you know, if what I'd like to see happen is doing. So the doing would be whatever draft, you know, Pete is able to come up with, without whatever players, you know, that we, if we have time, we should look really read in advance and give some thought to it beforehand so that we can then enrich whatever, what Pete is doing. And to me, that's sort of modeling different levels of co-authoring from a core team to an outer team. And then people who are just readers who see something before the thing is published who can provide a reader's perspective of what's going on. So anyway, food for thought. Rick, thank you. I just want to reply to you for a sec before going into class. The reason, well, I think the reason we're having this slightly abstract conversation is that Pete and I, in one of our many conversations about how to do all this kind of stuff, we're like, okay, okay, so we've got nuggets and I'll just paste the nugget that exists in the world on the OGM Wiki. There's a page called Nuggets Are Really Powerful. It exists in the world. And Pete and I were like, okay, how do we want, which of these many ways do we want anybody to interact with this nugget? Because the sooner we solve that problem, the more consistently we'll be able to do it and use it going forward. So we kind of need to, that's why today's topic is collective authoring is that we actually hit the pragmatic question of, here's a nugget, how do we want people, what permissions do we want, what tools do we want, whatever. We ran into that immediately. So I'm sorry if this sounds abstract, but we bumped into it like a ship running on a ground on a coral reef. It's like, oops, if we don't solve this, we don't get to sail the ship. Yeah, if I can just briefly, I really enjoy the abstract and the theoretical, the meta level thinking that goes into things. So I can get off on that. All I'm saying is that, I mean, it's critically important. So when you operationalize it, it's having the desired outcomes, whatever you think they should be. So I'm, but I'm just nudging, just nudging us to move in the direction of being pragmatic. But I agree with you, the abstract is incredibly important and it's often shortchanged. I get into these arguments with people about, the doers and the thinkers of the world and the doers tell the thinkers, they don't know what to, they can't do anything. And the thinkers tell the doers, you don't know what the hell you're doing. And to me, that dichotomous thinking is totally an utterly futile. And we get into these town gown arguments that are futile. So this is not a criticism, this is just an observation about, let's move in this direction and do something pragmatic. And then we can come back to the abstract and learn from it. Thanks, Rick. Klaus? Yeah. So I sent around this frontline report on AI yesterday because I thought it was very rich in summarizing, you know, where AI is, what it does and where it's heading. So let's take, for example, this goal, right? The Google team that won against the Grand Master of Go, which is like an incredible feat. And the audience, millions of people, tens of millions of people watching this, never thought it was possible that the machine could outdo the Grand Master of Go. Well, it did. So how did this really happen? You had a team that programmed the AI to know as much as it needs to know to accomplish this. Now, if you go to the same AI and ask it later on now, how do you restore soil health? It has no idea, right? I mean, this thing is highly specialized to beat a Go master. It knows everything about Go, it knows nothing about anything else. So AI is very narrow, right? It's very specialized in what it knows. So when you are inserting teams to edit, to write, to contextualize, where do you insert it? After AI has produced text or before, guiding it to produce the best possible text. So my understanding is that the initiative and the energy needs to be in the coding, in the programming, in the teaching, in the guiding process of AI. So what else do you need to know in order to come up with a better answer? So anytime I'm interacting with AI and it produces something that is not satisfactory or it just doesn't make sense or it just seems to miss something, my impetus then is to go back and say, what else can I profite with? What other information can I give you to come up with a richer context and a richer answer? So then when you do that, you end up with a chatbot that is very specialized. So in my case, my chatbot is highly specialized in understanding food and agriculture in the context of evolution, in the context of climate change, in the context of food security, in the context of culture, cultural impact and all of those things. So it can produce intelligent answers to questions. And what companies are doing now, where chatGPT is going with this chatGPT enterprise, is it's giving a tool to companies to insert their knowledge base into a tool and then add on to this collectively as a team, add on to write the best possible source feedstock for this AI to assist them then in executing or developing strategy and outcomes and so on. So it's just, I think that's, it's just, AI forces us to think differently because at the end of the day, we can't compete with AI. And this is really also something that really hit me when I was watching this again, this frontline report. The Asians know is that our belief system have no problem of giving up and handing over saying, yeah, we can't be as smart as AI. It's just no way you can outthink AI given, you know, at the same level of source information, it will outperform you. So they are turning, so the Chinese and the Asians in general have fully embraced AI as a tool. They understand how it works because the example that they got out of this goal exercise was pretty much, I need to teach this thing, you know, to think in very broad contexts and then it will give me outputs that I couldn't do on my own, different way of thinking. A couple of thoughts which just said Klaus, the documentary you shared with us is from 2019 and AlphaZero and AlphaGo were 2017, 2016. And what's interesting is that you are completely accurate in saying that these were very narrow domains, they were playing the game of go. And some of the big advances that have happened since then, the reason 2022 was such a big year is that they're ganging together broader thinking context, a collection of experts, kept capacities, they're actually able to run, you know, we don't have AGI yet, we don't have generalized intelligence, but boy, we've made a big leap from the very, very narrowly capable tools from before. Also a second interesting thing, maybe AlphaGo was taught with thousands of go games because go has been played for many, many centuries and they had the best games of go to teach it. AlphaZero was just given the rules of go and because go is 19 by 19 lines with black and white stones and here's how you capture territory, it played itself until it got way better than AlphaGo, the version that only played human games. That's really interesting. That simulation is not, I don't think possible in the real world, unless we were to take some new gen AI and just drop it into the world and say here, go read as much as you want to and get smart by interacting or by interacting with, I don't know what that would look like, but that little experiment is not doable in practical broad real world examples, but it means to me that these things are really capable of creativity in the sense of they don't have our own biases built in and what we're doing with all these gen AI large language models right now, is we're feeding them all of human culture, which is full of biases and then we're worried about, oh my gosh, the thing has biases in it. It's like, well, yeah, we gave it like what humans do and we seem to be pretty biased. Anyway, all those things, yeah. My point was, where do you insert a team to engage with AI before or after, right? So my argument here is you engage the team and this collective writing in the process of coding or training the AI, what comes out of it, right? To rewrite this, I wouldn't rewrite it, I would rewrite what I have thought the AI if it is unsatisfactory. So that's the point I was trying to make here. Okay, thank you. And I think you've raised a really interesting question, which is what is the role of AI at any stage in this process and how do we engineer what we're doing kind of to fit that, to take advantage of that? Gil Klosserich. Yeah, this is maybe off topic, but I can't resist responding to Klaus. I think Klaus, what you're speaking about highlights some of the real challenges in how we're thinking about AI. And Jerry, you clarified some of that with the AlphaGo and AlphaZero and whatever we call now, square root of zero or something. The tendency is to think about programming and treating AI's as like a massively, like a huge beyond imagination computation engine with instructions that it proceeds through. And it's very different than what we're seeing and it's certainly not the way human beings work. And I've been finding that in the guys I've been working with on building some AI's on my side where I see that their tendency is to try to distill a set of rules of human interaction. Like what does Gil characteristically say to a client in such and such a kind of circumstances which can get at something but doesn't get at the heart of it and certainly not at what intelligence is. The Go Master isn't just a master of the rules in 10,000 games of Go, but she also had breakfast that morning and she heard a poem the night before and she went to a concert the week before and that affects her play in a way that no kind of code can get at. And even the AlphaZero was simulating the whatever the number of permutations that's possible there, does that exhaust all possibilities for surprise and innovation and the something else that humans do so far seems to be distinct and different than what anything we construct can do. Even the AI's of now are, they're still as far as I can tell they're still stochastic parrots. They're very big and very sophisticated and they produce what would be around this house called simulated intelligence. We wouldn't call it artificial intelligence but it produces something that feels like intelligence and is good enough for some purposes but is different than what we would characterize as intelligence. So that said, I'm not sure what all that, my comments to your comments and Klaus's provocation that started it off, I'm not quite clear what that has to do with the Neobux project. Thanks, Bill. Can anyone tell me? I don't know. Stuart by the way is in the Chinese embassy in San Francisco awaiting a visa. That's why he's being quiet. He can chat in the chat but it's just listening in. So Super Magistrate's response, the entire Neobux project is based on GPT interaction. In the title it says written with the prompted and edited by chat GPT 4.0. That's the whole baseline here. I think that's your book, right Klaus? Is the way to say that. So by the way, a thing that we've hit a couple of times and I wanna say it real quick, I apologize Jose and Rick. Neobux and how you write Neobux, I think the Neobux definition doesn't strongly include it must be written collectively or it must be written by one person or it must be written by somebody talking with chat GPT. I think that the Neobux idea is kind of a general thing where there's nuggets and repurposing and things like that. And then there are ways to write Neobux. And then even in ways to write Neobux, maybe you're using chat GPT, maybe you're using, I don't know, a jazz soloist. The collective authoring. Collective authoring to me means they're using a Wiki, they're using a Google Doc. They figured out how to do collective authorship like Graeber and Wengro and they're sending emails and probably drafts back and forth through each other and Microsoft Word or God knows what. So there's lots of different ways to do Neobux, I think, is one of my top line kind of things. And Jerry kind of got us started on collective authoring because it seemed like an interesting topic from last time. Maybe it's interesting, maybe it's not. But even collective authoring isn't the only way to write or chat GPT and you aren't only the only ways that you can write a Neobux. Thanks, Pete. Jose then Rick. Yeah, Pete, I agree with your plural. Rick, can you wait, hold up a second, Jose was ahead again. No worries, Rick. So at first when I was listening to Klaus over the last couple of meetings, I felt like, or we have in the same meeting because Klaus is focused on AI. And I must admit that today I kind of flipped to Klaus' kit. A bit, I'm wondering that if what Klaus is bringing to the table is actually quite interesting, are we looking at how we've been thinking of knowledge, the creation of it in writing and documents in books or whatever physical or virtual medium. Is that in itself something we're stuck on because that's the way we thought of content of knowledge and that maybe Klaus is trying to point us to something which is there's another way to think about what this information is and how it works and where it fits. I find that a rather intriguing question because maybe it's time for us to stop thinking about information as what I write or you write or somebody else writes or even how we write it together but more about how do we put it into a new format that gives us a lot more than what we can been able to do ourselves. So that to me is an interesting question. As to the comments about Neobook is collaborative or not collaborative or doing it this way or that way and this tool or that tool. I think again we're talking about the how and I wonder if for example what comes to mind for me is is there a way to look at a model of what we wanna write and or create and is that model for example starting with first principles that what we do is we change the way we think about what we're creating at the informational level not at the doing level and then work from there at the informational level and then think okay well now how do we with this informational level how would we use the right tools to build it and why is this informational level the one structure that the one that we wanna use so still playing with those things because I think there's a bunch of pieces here that we're playing with that are rather interesting. No answers, just questions. Thanks, the same, all right, go ahead, Pete. Did Rick get to go? Oh yes, actually Rick was next in queue and Gil is about to leave the call because he's gotta go places. Thanks, Rick, go ahead if you want to. Yeah, that's fine. That's one thing about hiding your cameras so you don't know what turn you are so that's the reason why I jumped in. The downside of using the hide feature, I actually am interested in Gil's before he leaves and actually in Klaus actually, I had a dental appointment. This is sort of tangential but relevant. I was at the dentist this morning while I was there I listened to a podcast that was called The Next Economy Now and learned about a woman by the name of Kate Taylor. She and her husband have a foundation about how to redesign banking for regenerative agriculture. And I learned from a colleague friend of mine who says the USA is in a bit of a pickle because the number of farmers over 60 is quite significant who are retiring and young farmers can afford to buy the farm. So it's prime picking for the agro-industrial complex to sweep up these retiring farmers who can't get buyers for it. So I mean, it comes back to food policy but I found this podcast absolutely fascinating. I just wanted to put that on there. And if it's old hat for you guys, that's fine but it was certainly new for me. So, and just a final comment, I agree with you. We need a plurality. The question with a plurality though is you may have to have different first principles depending upon different domains that you're working in. So. Thanks, Rick. I think several people recognize the person you were talking about as Kate Taylor of Tomcat Ranch and a bunch of other really interesting things around regenerative ag and such. Yeah, it was just, the podcast just came out and she was talking about the recent developments that and again, if it's old hat, that's fine but for those of you who are behind this curve which I am, it was highly informative. If you have the title of the podcast or a link to the podcast, you can drop that in the chat. That would be awesome. Thank you. I already put it in actually. Oh, I'll scroll back. Pete, floor is yours. Thank you. I wonder, one of the, I have a hypothesis. I think one of the problems that we get that we suffer from with Neobux is that it is by nature kind of a matter project. And so once we have one level of meta it's easy to keep going kind of at other levels of meta. If I may ground this a little bit, I think the Neobux project is the kind of Jerry's idea that let's figure out, I'm gonna say new but new is not really the right thing. Let's figure out some different kinds of ways of acknowledging in the world. So for centuries, we've been using this thing called books and maybe more recently magazines and novels and fiction, non-fiction. So books have become both kind of a centralized and well-known way to do knowledge, well-respected in lots of ways, reviled in other ways. And let's kind of try to break that mold a little bit. They're useful and they're counter useful in some ways at this point. I remember I remember being in my headspace about Wikis or something like that and being on a trip and walking into a bookstore, an airport bookstore and thinking, I get it, books are where knowledge goes to die because it's like everybody writes stuff down and we print it all up and then we store it in these warehouses and then at this point, it's not a thing. They don't live anymore and I'm overexaggerating a little but not too much. So anyway, the idea of Neobux, Jose to pick up kind of where you were, the idea of Neobux has always been not to focus on the book part but to focus maybe on the Neo part and use book as a way to get some attention and to talk a little bit about the gravitas around knowledge and information and things like that but then free it from all of that with the Neo part. And so sometimes a book is, I don't know that we've talked about in this space but maybe we have, maybe sometimes the knowledge comes out in a song. Maybe sometimes the knowledge comes out and I've certainly talked about this, maybe it comes out as a GPT, a chat bot that actually knows everything that you wanna convey and the main interface to your knowledge is some people walk up to it and they either type or they say, tell me about soil health or tell me about muscle cars. Why should I care about muscle cars? So the Neobux idea has always been multi-polar representation of information and then how do we form teams that work on quote-unquote Neobux and the idea is the Neobux project can help people who are streaming in going, I've got stuff I wanna convey or communicate or learn. Help those people stream in and help them connect with ways to do that. One of the least interesting ones being publishing a book. One of the more interesting ones, maybe publishing a wiki, maybe another more interesting one is setting up a forum where you larder the forum with a lot of information and you pick a few of your closest friends who know about the topic and maybe there should be a soil health forum that is a quote-unquote Neobuck along with a GPT, along with a Kindle volume co-written by Klaus and Chad GPT to let people know, hey, there's this thing called soil health and even if you found it on Amazon, you should still learn a lot about it and maybe we'll also kind of point you to more interactive ways to do it or more rich ways of interacting with a knowledge that Klaus and Chad GPT kind of put together. So it's kind of like all of those things. And if you now go back to that, what I said was Neobucks is kind of a colony of folks that are helping subject matter experts or subject matter inquirists, people who want to know about subject matter. Neobucks is a project to help those people connect with modes of expression. So Neobucks already is a meta-level thing where the idea of Neobucks was never to be about one particular thing. It was about the way of using the techniques that we have social techniques and informational techniques to help people publish, if publishes even the right word because it's not really interact and express and learn is probably, those are probably better verbs than publish even though, because we say book, we say publish, that's not the intent. Pete, everything you just said plus a hundred from my perspective, I just like really appreciate the way you said that. I'll add one thing, which is, I don't think Neobucks is meant to be about just learning and building knowledge together cooperatively and collaboratively. And we're talking here about how do we author collaboratively. This is also about debate and conflict. And it's like, hey, someone else used this space to go tell us what their ideas are and we disagree vehemently. How do we see that? How do we do that? How do we enter that space so that we can maybe figure out some middle ground or convince somebody or whatever else it might be. But it's not just about how do we build the encyclopedia Galactica together. It's more, how do we model the real frustrating interactions that are happening online in a healthier, better way that slowly builds this knowledge thing over time. Whatever that sort of means. Dave, then Klaus. Yeah, thanks, Pete. You probably ought to like snip that and, you know, pasted it to the website or something like that. It's a helpful description. And I guess I was trying to, like I just, you know, kind of the horse I'm beating is I think, pretty related in this. It's like, okay, I want to do, I want there to be a specific thing in the world. I want landscape regeneration. So Klaus wants, you know, soil help. And I think we're going to need a lot of learning around landscape regeneration. So I want this open source stack of knowledge about landscape regeneration, which at first I thought of it's open. We just want open knowledge. But it turns out that's not really enough is it needs to be dynamic, right? It needs to capture the learning part. So the body of knowledge, the technology base or whatever we want to call it has to keep improving. And ideally it's organized enough so that as it improves, you know, it's able to help the next group do their landscape regeneration, right? So there's an acceleration function as part of this. So, you know, to me, what we're looking for is collaboration that enables the technology to dynamically improve. And we want to do it open so that it's not captured by capital interest. Thank God there's another underlying being there. And which has kind of brought me around in the last couple of months to say, oh, well that means there needs to be, I'm calling it a business model, right? So the part of I would argue the neo books problems, like how do people work with each other in the incentive problem? It's not a technology platform problem. It's an incentive problem. Like what, well, the question is why do they work with each other? You know, and if you look at why you write a book, you know, there's a set of reasons that an individual author was writing a book, you know, none of which would probably make money unless you're really a rare author. There are a bunch of other reasons. Why would I write a bunch of, why would I write a book with a bunch of other people, you know, needs I think a different mix of incentives? And I think that personally, I think that's kind of the piece that we really haven't done a good job of cracking is trying to understand what these incentives look like that the shared contribution makes sense. And so I was, I don't know, I stuck in a link to complex reciprocity in the chat because I was Googling it as we were talking. And it's like, oh, there's a whole field here that I know nothing about. But, you know, in some sense, I think that's the problem we're trying to deal with is like, why, you know, what is it that's gonna get us to kind of participate together? Then I do think there's design things like modularity that matter. And I also think that there's probably differences. Like you basically need different, it's a multi-sided platform and you need different incentive structures for different people. You know, some people are gonna wanna make something pretty and so you have to give them a way to make things pretty. Some people like to make sure the words are right. You let them edit and some people are gonna, you know, wanna add their own chapter, I don't know. So you're gonna, you have to kind of structure the system that enables the people who have that incentive to participate in the way they wanna participate. And I think that becomes a design problem. Business models. Dave, thank you. It's interesting, the whole, how do we reward people for participating or how do we motivate people for participating? They could eat a tremendous amount of our time. I will point out that there is no business model for Wikipedia or maybe the Wikipedia is a public commons good where it's all donations fund the actual hardware and some people's actually staff time and all that kind of thing. But the vast, the majority of the people who did those edits or Oxford commas or whatever were not getting any money and didn't expect or want any money. No, but they were incentive. There was an incentive. Correct, but it wasn't modest. I'm using business model as a shortcut for resources are coming in and productively creating the system, right? And yeah, they, for some reason, people are incentive to participate. And I get the impression the whole, there's like a whole cohort of people are incentive because they like to create structure and being other people and shit, you know what I mean? It's Wikipedia, right? Totally get that. So Dave, way back when Clay Shurkey talked about the plausible promise and he said that when Linus Torval started Linux his plausible promise was, hey, I really want Unix to run on my PC which is kind of a wimpy machine. And when I'm done, it'll be under the GPL which already exists. Therefore anybody who helps will be able to use it. That was his plausible promise. And as soon as you say business model my little radar goes up and I'm like, eh, but you mean I think roughly what Clay meant back when about a plausible promise. And sometimes the plausible promise is there will be financial award for you if you participate and that's awesome. And I totally agree with that. Maybe, but I think maybe business model is still not a bad term because if Torval had finished writing Linux and thrown it over the wall and said, look, open Linux and nobody had done anything else with it it would have been open, but not matter. What made it matter is it's continuing dynamism, right? Which is because of the business model, right? When IBM decided they could put a billion dollars into Linux and save money that was a business model decision. So the plausible promise I think is where we've been stuck for the last 15 years, but it isn't enough. So we need something more than that to generate the ongoing dynamism. And the Wikipedia I'd say has captured that. I mean, I don't know what it's growth curve looks like and stuff, but it has, it's sustainable and it keeps improving I'd say. Linux keeps improving. The foundation if you look at it has added hundreds of other packages not the operating system, right? And I don't know, some of those must be surviving and some must not, but I don't even know what that looks like. But anyway, I really do, you know they're probably a better from the business model I use it to be a little bit to be provocative but I really think it's kind of true because it is that that creates the dynamism. Thank you. Klaus, when Pete? Yeah. One thing we don't want to lose track of is that this is a fast moving field, right? While we started this new book project how long ago? Less than a year ago, right? The ground has shifted. I mean, GPT 4.0 got released. Then OpenAI released GPTs. Now all of a sudden you have thousands of GPTs hopping up with specialized content. And now comes GPT Enterprise. And you wonder where is this thing really going? And so what OpenAI is providing here is a tool set that you can use to customize for your very specific type of business and to assist you in solving very specific issues you're dealing with in your business. So one of the most important aspects of this really is that the skill sets to program a GPT is the hottest thing emerging on the labor market. If you can go to an assist a company to set up their GPT in ways that broadens their knowledge base and that activates an AI capacity within their specific field, that is an amazing asset. And so, I mean, I'm seeing it right now because I'm working with some companies and I can within an hour, it's like David, within an hour I could lay out a ground plan for you on how to capitalize on your project and give you a business plan. And there is also a commercialization evolving where you take a GPT, you develop it and you put a user interface at the front of it. Now you can take subscription services for your interface. You can charge based on time, based on membership, whatever. And you can capitalize on this. And these are things that are happening as we speak. So then the question comes, so how do you swim with this? Because this is out and running. And I always like to compare it to when Excel spreadsheet came out, right? When Excel spreadsheet came out, at first the accounting profession was, oh my God, I'm losing my job. This is gonna get us unemployed. And indeed, the typical accounting work shifted dramatically, but then somebody discovered, hey, I can play a what if game with this thing. And it changed the nature of accounting. So now you have AI is the same thing. People are starting to figure out what you can do with this thing. And the first reaction was parallel paralysis. Oh my God, our jobs are going away. And then all of a sudden you realize you can do things with this that are requiring an entirely different skill set. And if you embrace that skill set, you can go out and just be on top of dive again. So just a different way of thinking. And David, I think there's a snowball chance in hell you could compete with someone with your project who's using AI, you couldn't. It would outperform you, it would outrace you in no time. Yeah, I mean, you may be right. I guess I'm assuming that AI is also a spreadsheet, right? I mean, so if you were to say to what you're saying to me is you can go to AI and tell me how to do landscape regeneration anywhere in the world and the answer will pop out. I think that's not what I'm saying. That's what the technology needs to do. So that's what we need to have happen. So we'll use AI to get to that point, right? But we don't have the data that we need. We don't have the algorithm. They don't exist, right? But they will exist over the next 30, 40, 50 years, right? And AI will be embedded in their help with search and things like that. What we want is a collaborative process of creating that, right? Which I think humanity is still gonna participate in. So I mean, I assume you're right in that, you know, we'll increasingly see AI as the front end to the body of knowledge that will be developed either by people or machines and that data will be used then on the ground to implement change that, you know, restores landscapes. But it's not a magical process that you can do now with chess with the GPC-4, right? That fades away. Anything you need to know about landscape and landscape rejuvenation is known. It's in the library. It's under science. It's right there. I mean, what is it that you would wanna know that is not known? I mean, give me an example. Hydrological models of water supply in Playa Viva. And they are not known. And how they can be influenced. Yeah, nobody's written those. That's totally wrong. They don't exist. But you know the science. You know the physics. No, that's being argued about too. Okay, I offer you a one-hour work session. I offer you a one-hour work session. I don't want it. But you can talk about it. Yeah, you don't want it. Hold on one second. There's an interesting philosophical question that LLMs raise, which is, do LLMs obsolete note-taking, writing, collaboration among humans and all that because the future holds, we're just, this thing is going to be eating information faster than any human possibly could. It's going to know the answers better than any human possibly could. And we're just going to be interacting with this thing, this conversational beast over time. And that's it, right? And I don't buy that future. And I think that pieces of that will show up, but I don't buy that future at this moment at all. And I'm trying to figure out how do we use AI extremely sagely in the process of as humans curating what we know? And what is the meeting ground of human curated knowledge with AI generated or curated knowledge? And how does that even work? But I don't buy that writing has been obsolete. I saw a couple of posts early on, hey, forget about note-taking. This whole tools for thinking sector, it's screwed. LLMs have just destroyed human note-taking because there's just no reason to continue taking notes. These AIs are going to know better everything. And I'm like, no, I'm not drinking that Kool-Aid. And I think that having a debate about that might be a thing we could do on a call here except I just made a mistake picking collaborative authoring on this call as a topic. So I don't wanna jump on that real quick, but it's a really legit, interesting question. And I'm very interested in one of the wisest ways of using these new AIs, which I think are for not, Klaus, the reason I said that was 2019, this is now is that they've really leaped forward. They've been a whole bunch of very interesting advances and there will be more because now we can even use the AI to improve the AI. It's crazy. Singularity, anyone? We haven't even said that word here. Anyway, I wanna just bubble up a little bit and say there's a philosophical question here that's important to our quest. I'm very much on the part of how do humans author stuff together occasionally using AI really well? But I'm not interested in having AI author my particular text. I've got a bunch of ideas I wanna line up in a row and put into the world and see what happens. Pete and Jose and then we're about at time. I have a lot to say that AI throw and con in other venues. And for me, the other venues are places to talk about the pros and cons. I will say that we're not to the point where AIs have meaning or knowledge. AIs, the way I like to say it and then I always kind of contradict myself when I give somebody some advice. But the way I like to say it is AIs are really good at slicing and dicing language. Lots of knowledge is expressed in language. So AIs by proxy are good at slicing and dicing language-based knowledge. Still in all, when you get something back from a chat bot unless you read through it and ensure that it makes sense and ensure that it's meaningful to your audience, you just got a steaming pile of crap. Bill Anderson has a, I forget the person's name but he's got a philosopher that he likes what she says about AI. And it's something like AIs don't make meaning, people make meaning. The output of an AI doesn't mean anything. It's until somebody's gone through it and said, I agree with everything or I agree with 80% of this, I'm gonna change 20% and add some stuff. So not so interested about AI discussion here. Much more interested like Jerry, kind of I think about how do we collaborate and how do we up level our sharing of information and knowledge and thought and I really love Jerry, he said something like sometimes it's one person sharing stuff, doesn't have to be everything, a collaboration. I wanted to get back to Dave, David where he said business model and then a couple of words after that was incentive. And so Dave, I kind of disagree with your need for going back to business model. I really liked when we ended up with an incentive model. So this to come back to where I started actually this call incentive models is how you drive shared goals. So one incentive model is I work on this because I get paid by my boss. Another incentive model is I want everybody in the world to know about 1963 muscle cars and there's three other people in the world who will help me. So we have a shared incentive. Even out to Wikipedia, Jerry you're right, Wikipedia doesn't have a particular, it kind of does have the Wikimedia Foundation has a business model. Wikipedia doesn't really have a business model but it certainly has an incentive model. There's lots of incentives to create with Wikipedia and that's the thing to look at. So when, I think when David says something like or any of us say something like business model if I'm gonna keep translating that to incentive model on my brain because then it takes out the weird money thing and lets you look at other stuff like David said. Some people like to beat other people up on Wikipedia and that's their incentive. Hearts, clubs, diamonds, spades, the essay by Bartle. So thanks David for leveling me up from oh all we have to do is come to a shared set of goals and really what we need to do is understand incentive models and work with different people who are incentivized by different things sometimes with money but often not. And that's where we get that the ability to do shared goals. And the only thing I gotta say there is I think they're often incentive by money. So I mean, I feel like we can't write that part off. Yes, yes man. And the success of something like Linux is a financial store, right? And there's a financial argument for why we like it. I think yes and no, your points are well taken and the fact that IBM or something it's funny when I never want to disagree with you David by the way but it's great that big companies poured billions of dollars into Linux but there's a lot of open source that's not incentivized by money and I think Linux didn't succeed because it had money pouring in. It helped a lot and it got in many more places. There's an opposite incentive which a lot of open source gets taken advantage of because it saves somebody money. So there's some like horrific stories where there's some critical piece of internet infrastructure that's maintained by one guy striving to death in Eastern Europe someplace. And all he wants is to be able to eat and sleep in a warm place and then he would be covered but everybody else is like, well, I guess I will just use this for free. So there's the opposite side of that too but there's a lot of open sources just, hey, if we share, share alike, we get more out of it. So there's as much of that energy and dynamism in open source as there is money from corporate sponsors. I'll put another very interesting discussion and I think incentive models are really a key to the NIO books conversation just because we need that shared goals thing. Totally agree. The internet incentive model was how do we avoid our communication network being taken down by Russian bombs? There was and there were a whole bunch of contractors who wanted to make a lot of money from it but when the internet was born, it was competing against the advanced intelligent network which is what all the for-profit telcos were trying to build so that they could make a lot of money shipping all this cool new video and other kinds of stuff around that advanced network got eaten by this weirdo protocol called the internet. It just basically like ate its brain like cordyceps which I find really interesting. The DARPA money and incentive transferred into academic people and civic-minded people. So that was the next evolution of the internet. People like John Pastel and a bunch of people, professors and students and things just trying to get stuff done together and for the better of the world. Thanks Pete. Jose? I just wanted to throw my head in the ring there with Pete on working on something for next week. For the taxonomy? We'll figure out what it is. Sounds great. Can we have that conversation on the Mattermost channel for this call? And Pete, you're muted. I created a taxonomy wiki page already. Yeah, the document is in the chat. It is to collaborate asynchronously through the Mattermost chat. And if you're not doing Mattermost chat or you can't figure out the wiki, just send me an email or something. If you think I can add value, Pete, I would also join in. Thanks, thanks. Remind me too, obviously. Cool. So we're gonna do that in between should we report in on that next week? Will that be the start of our call? See what progress we make on that and then see where that takes us? That should be one of the check-in things, yeah. Cool. I don't know when that call is. If you could set me a note. Oh, next Monday, just the next call in this sequence. Just the near-bix call. The next near-bix call. So there's no extra meeting between now and then? I think we should do it async. Yeah. At least try. Yeah, so the next one is Monday the 12th. So the first step is look at this webpage and tell me how it sucks. Cool. And if anybody wants to edit the wiki with Obsidian, I'd be happy to show you. And if you go down to the bottom of the page, you'll find a link. So if you wanted to use a line-oriented editor on GitHub and you have a GitHub account, you could edit the page directly with the link that's currently by default at the bottom of that page. But that's kind of the hard way around. But it does work, right? Yeah. Yeah. Cool. Which is one of the reasons Pete and I decided this was an interesting topic. Or ever software challenged. Good. All right. That seems like a good place to wrap this call. Thank you. Thanks all. Thanks. Fun navigating these wrappers with you all. Yeah.