Open Book, Open Process

We’ve decided to be open about our process, sharing both our successes and challenges. By documenting our experience, we aim to provide insights for others interested in AI-assisted content creation. While we’re focused on producing a quality book, we’re finding significant value in the learning process itself.

The project is progressing rapidly, with AI assistance enabling significant productivity gains. For example, I, David, was able to produce over 2,300 words of content in 29 minutes. We’re exploring how to balance speed and quality in this experiment.

Our approach raises questions about authorship and quality control in AI-assisted writing. We’re developing methods to leverage AI not only for content generation but also for analysis and refinement of our work. The self-referential nature of writing about AI while using AI tools adds an interesting dimension to the project.

Here’s an edited transcript of a meeting we had discussing these and other topics.

Key points

  • The project’s rapid pace and experimental nature.
  • Balancing quality with speed in book creation.
  • Concerns about reputation and the final product’s quality.
  • The process of comparing and refining AI-generated content.
  • Reflections on authorship and the role of AI in writing.
  • The importance of transparency and openness in their collaboration.
  • Plans for refactoring and improving the book’s structure.
  • Strategies for sharing their process through blog posts and tweets.
  • Agreement to continue with the current schedule while remaining flexible.
  • Plans to meet more frequently to discuss progress and challenges.

Tasks

  1. Continue writing chapters according to the current schedule, aiming to finish by the 15th of the month.
  2. Review and compare the existing drafts of chapters to identify overlaps, gaps, and areas for improvement.
  3. Start posting more frequently on social media (Twitter) and writing blog posts about their experience and process.
  4. Massimo to share his experience using Figma for visualizing and organizing the book’s content.
  5. David to review Massimo’s draft of Chapter 2 (Theoretical Foundations) in the drafts folder.
  6. Continue exchanging notes about their progress and discoveries.
  7. Refrain from focusing on styling and refinement at this stage; prioritize getting content down quickly.
  8. Consider ways to refactor and improve the book’s structure after the initial drafting phase.
  9. Develop methods for fact-checking, referencing, and eliminating AI hallucinations in the content.
  10. Schedule more frequent check-ins, with the next meeting set for Friday morning at 9:00.
  11. Both to reflect on and potentially define their quality expectations for the final product.
  12. Continue to be open about their process and challenges, using them as potential content for the book or supplementary materials.​​​​​​​​​​​​​​​​

Edited Transcript

Massimo:

Let’s add for our friends, human and not human, that we are talking – this is Max Massimo Curatella, and we have David Orban. We’re discussing our project called “AI-Powered Knowledge Book.” We want to write a book with AI about AI, and we’re having a lot of fun. We set a very strict pace a few days ago and had specific practical tasks. I was writing theoretical foundations, while David was writing about AI and knowledge representation. 

The full title is “AI-Powered Knowledge: Navigating the Future of Information,” and the chosen subtitle is “Transforming Knowledge Management and Information Architecture in the Age of Artificial Intelligence.” We exchanged many feelings, impressions, tricks, tips, and tools, and each of us had our own emotional journey. Mine was particularly similar to a roller coaster – I went from completely blanking and panicking to excitement, dreams, and nightmares, but it was fun. Yesterday, I had a sort of illumination, my eureka moment, and I can’t wait to tell you about it.

David:

I’m very happy we’re doing this because of how much we’ll learn, and the final product will hopefully be valuable to anyone reading it. But I think it’s especially true in this case that a lot of the value is in the journey. I wouldn’t mind helping people who might miss it and would like to participate in a parallel journey after we’ve gone through it ourselves. 

So, having an exercise book or an addendum to the book where people can say, “Okay, this is the process that David and Max followed. If I want to experience what they experienced, I can also do these steps and exercises.” That’s why sharing what we do between each other multiplies our opportunities to learn, both by doing and by teaching the other, giving them the opportunity to try and experiment.

At the same time, we’re potentially documenting this exercise that will be an addendum or a parallel resource for future readers of the book. For some reason, maybe lack of self-awareness or not being a responsible person, I didn’t feel panic at any moment. I definitely felt exhilaration and joy. 

But what I definitely want is to share my conviction that it’s okay to create and share half-baked ideas and half-baked chapters. The worst that can happen, and the most likely outcome, is that no one will care. Even David sharing a half-baked idea with Max might not result in deep criticism for improvement, let alone the outside world, where the probability of anything we write being ignored or just the first sentence being read is absolutely the most likely.

But the best thing that can happen, and what we run the risk of achieving if we share half-baked ideas and chapters, is that someone actually tells us, “Hey, this is bullshit.” That would be wonderful. The only way we can do that is to give them the opportunity to get a glimpse into our process, into our bumbling, amateurish experiment. 

That’s why I jumped headfirst into not only creating the website and the Twitter account but actually writing and publishing blog posts that proudly put a stake in what a book outline should be, what the work plan should be. I’m planning to publish these drafts of the various chapters that I have created to get our AI writers’ input.

Massimo:

Thank you, David. Your connection is good, and by you talking slowly and clearly, I think you’re improving the chance for our synthetic friend to capture it. While I’m now struggling with the street vendor screaming in my ears, I hope nobody’s listening to this. 

I have different feelings than yours. Of course, that’s maybe the major reason why I’m doing this. I’m really enjoying having a partner like you; this is wonderful. I’m using your enthusiasm to go through my frustration, so I’m really thankful for that because I don’t think I would have been able to do this on my own. So this is already a win that I’m really proud of. This is how you do things. We’ve been sharing the cult of them. Can you-

David:

Yes, can you explain what you believe is the source of your frustration?

Massimo:

Yes, the source of my frustration is that I don’t trust the process yet, and I need to have faith in the iterative process of improving the process. This is something that I’m living every single day in my profession and in my life – managing and facing the uncertainty of things, and doing things that I don’t know how to do. It’s somehow my specialty, but I’m still, for now, human, and there is this inevitable step of feeling it. 

I don’t like to feel like I don’t know how to do things, or I don’t know things. My first reaction is not only curiosity but frustration. Why don’t I know that? Or why am I not able to do this? Or why is this shit? Which is absolutely the healthy practice of learning and design, design in the broadest sense of the word. That’s why you do things because you need to learn how to do them better and how to discover how to do them. 

The second level of frustration is that I’m familiar with the process, so no panic so far. I also have to say that with my colleagues, my teams, my clients, usually I’m the one who is soothing them by saying, “Don’t worry. We can find a way. We can discover that problem.” We don’t have time to look at things and say, “This is good, this is wrong. Let’s decompose that. Let’s have a nice plan.” There’s no time. 

I mean, if our plan – I wrote the plan thanks to your input, and I said, let’s write a chapter every two or three days, not five or four, one week, which is already crazy. It’s surreal. It’s absurd. So I said, okay, my friend David is going absurder than me. Let’s play this game. I want to be absurder than David. So let’s do a chapter in two days. Who cares? 

The problem with “Who cares?” is I need to be serious about that. I need to be accountable. So actually, I do care, because now I have to do it. So I have only one way of doing that, which is to use the tool we decided to use together. That is the sense of the game. So you’re not writing the chapter, you need to find a way to write it so it is acceptable at the end of the process. 

So far, the outcome is not decent, and I had to follow my intuition of spending a few hours at night looking at the result of what we had done. Professing the religions of systems thinking and critical thinking, I said, “Okay, why don’t I apply it? I mean, how can I be critical and systemic in addition to being systematic in this process?” It requires time, and I want to do it myself first. Then maybe I can instruct the AI to do it. And this is wise and inspiring. 

So I said, let’s do what I usually do. I used an infinite canvas, digital of course, and I started to copy and paste all the pieces of writing that Max and David did in the tidy columns of the chapters. Chapter two, chapter one, chapter three, because the first one is the introduction. We said, let’s write it at the end collaboratively. 

So I started by visually putting together the actual text and by actually reading it. Because when I do that, I’m actually reading, I’m trying to interpret it and say, “Okay, what did it write? Does this make sense? Do I know about this?” The first reaction was, I know less than half of this. And so there are some philosophical questions, like, how can I say I am writing this book if I don’t know about those things? So there’s this problem. This is not me writing. I instructed somebody else, something else to write it. Okay, good, fine. Let’s live with this feeling. 

The other point is, I usually use the writing approach to learn myself, to explore what I feel, to understand what I don’t know. So if I am not writing it, how can I learn from it and about it? Second main philosophical question. So I said, okay, let’s stop the game. Now I am carefully reading it. I will deep-read it, and I will copy and paste all the sentences, one side by side with the other, and I will start to do comparisons. 

So the first thing you have to do when you want to be critical and systemic is define things. What is this thing? And inside this piece of text written by AI following my prompts, what’s inside it? Is there any concept, definition, things, statements? So I carefully read those, and I started to discover something beautiful, fantastic. 

We generated an outline on the fly quickly, too quickly, of course. And then I said, no, it’s 14 chapters. Let’s do 10. So it was compressed, summarized. We wrote. And I said, it doesn’t make sense. I mean, why do we want to talk about knowledge representation and information architecture, which are, of course, related, but they’re two different things? One is computer science. The other is not strictly computer science. It’s library science. It’s about how people think about things. And of course, they would merge, but there are two different disciplines, different people are doing those jobs. So why are we connecting those things? Does it make sense? 

And then I said, let’s see how we treated the different topics by different people using different tools. And then I compare the two chapters. So I will show you, David, that I am in front of me one long, very tall column of text, which is chapter two, mine together with yours on the right, which is chapter three. And then I started to read and compare. 

So, yeah, the step is understanding what you’re talking about, defining the boundaries and comparing things. Try to find patterns. Well, that was fantastic, because most of the two chapters are about the same thing. So different prompts, different goals, different people, different processes, different tools, the chapters can almost be overlapping, one with the other. 

So first big problem, how can I be talking about theoretical foundations, and you talking about AI knowledge representation, and in the end, we are writing about the same things? This is a big realization. I’m loving this thing. I need to find in the process a way to avoid redundancy, or those are two drafts that need to be, of course, fruitfully elaborated, manipulated so that there is no redundancy. 

And the other beautiful thing is that if you look at the writings like two opinions by two experts, two researchers, two writers, you can confront them. You can compare them to discover gaps, similarities, symmetries and asymmetries. So when one is mentioning applications of knowledge representation on the left side compared to the one done on the right side, we have a different number of actual topics or differently treated topics. 

So I see value there in saying, well, let’s compare the two, let’s find gaps, and let’s make it better by merging the two things. So that’s pure creativity, because there is exploration by creating drafts. And then there is combination, comparison. And this is what a critical thinker does. Let me compare these two sources, and let me integrate the two things. And that’s another beautiful point.

The same was happening with history. So one is talking about what it is through the years, and is doing a different treatment in the history by being very quick about the recent times, while the other is starting from the Egyptians. That was hilarious. I was having a laugh about it, and I said, okay, cool. I can put piece number one, piece number two, put them together and have a beautiful draft for the history. 

So this is what I saw only yesterday after four days of completely wandering in the dark, trying to understand what I’m doing and why I’m doing it. And I have one hundred other things, but I want to stop for a sec and give you back the floor.

David:

Your concern about observing what you are doing and asking yourself, “Am I writing this if I don’t know everything that it contains, if I didn’t personally research, dig deep and understand” is, of course, a very appropriate concern. And it could very well be that in some number of years, what we are doing now in order to produce what, at the end, should become a book – printed or electronic or audiobook or other forms – that the process and the authorship will have a different label. It will not be called writing. 

I don’t know if you had the chance of reading it, but I actually published a blog post about how I feel, and the analogy I made is with an orchestra where the conductor doesn’t play any of the instruments. It is very likely that he or she is capable of playing one or two or maybe even more, but it is almost guaranteed that he or she is not able to play all of them. The conductor has a very unique role in making a symphony happen, and that role is different from playing any of the instruments. And if you ask the person playing the violin or the drums or anything else, they will confirm that the conductor is essential.

So I feel a little bit like a conductor in having a team that researches for me, that is able to identify and describe a specific anecdote. You mentioned the Egyptians. I visited the Egyptian Museum of Turin and had the opportunity to be guided by the director of the museum. It was fascinating, and the anecdote that I asserted in the book about the strike that is recorded on a papyrus is fascinating. I just vaguely described it, and the AI that I used was able to immediately identify it, give it a name, and insert it into the narrative, as a very simple example. 

Now, as we know today, AIs can and do hallucinate, because they are so enthusiastic in following our requests that sometimes they don’t catch themselves in fabricating things that don’t exist. So it will be absolutely the right thing to do for us to make sure that we catch these hallucinations if they happen, and we have to develop reliable methods for double-checking, referencing, footnoting, end-noting – everything that needs to be done. And of course, we will use AI tools to do that. 

The objectives we gave ourselves are purposefully crazy, and we decided to make them even crazier, not because we know that it can be done. It is very likely that it cannot be done, or at least, let’s say that the sheer volume of text, because the objective we gave ourselves is to write the book in 30 days, right? The quality control that we both want because we are proud of what we do – it theoretically should fit the same timeline, but it may take a little bit longer. It’s not that after having just started, I already want to change the goalpost, but what I want to highlight is that a typical book needs a year to be written, or two years. So we are already increasing tenfold or 20-fold the speed with which a book can be written through this experiment, right?

And we have to make everything we can to make sure that the experiment is completed, and then measure both the quality, the feasibility, the sustainability, the repeatability of what we are doing. And just as there will be value for people potentially to do the exercises so that they can experience what we are experiencing, there will be potentially value in also analyzing whether this whole thing makes sense. And to say, yes indeed, doing it faster, maintaining quality is cool, and after taking a pause of whatever number of weeks or months, we may say, okay, the outline of the book that we finished had certain features, but we know there is an opportunity to go deeper in one area, or expand in another area, or cover things that we didn’t do. And we may decide to want to do another book in a process that, having learned to do it once, will have been greatly improved.

To give you an example of the kinds of improvements that could be possible: today, it would be probably cumbersome and would slow us down to involve other team members helping us in anything. But having done it once, I am sure there will be parts of the process that we enjoy less or we are less skilled in doing, and we will happily delegate the supervision of AI tools to other team members as we are orchestrating a potential second book. This could be fact-checking, referencing, illustrating, the typographical layout of the book, or whatever other things, while we are concentrating on creating ideas, dialoguing, and crafting what’s done in the first one, but in a more focused manner.

Now, you greatly enjoyed your ability to compare the different theses that have been written in chapter two and chapter three in their current drafts and recognize overlapping parts, potential redundancies. And as a consequence, there is a need for refactoring, deciding what goes where. It is a process that you did with a given tool that you are familiar with, that gave you a visual layout and forced you to read and to realize what you described, which is wonderful.

The applicability of different approaches – linear or bi-dimensional, maybe three-dimensional (who knows, we will instantiate book chapters in a Metaverse) – to be traversed differently. Maybe we will explicitly hyperlink parts of the book together and give people the ability to​​​​​​​​​​​​​​​​ David:

Maybe we will explicitly hyperlink parts of the book together and give people the ability to walk through the book in a color-coded path that gives them a historical reading rather than a technical reading, rather than a philosophical reading, independently from how the linear representation of the text pretends to tell just one story. So that would be great. I have people who will be enthusiastic in helping us in various metaverses. 

You just gave an example of what we can generalize – different kinds of tools in order to do better what is a very hard part of creating a quality book: the ability to analyze it from different points of view. And there are many additional ways that it can also be done. For example, what I am thinking of is to look at the style that evidently will be different from the parts I write and the parts you write, and then decide how to merge those styles so that someone who is disciplined enough to read chapter after chapter is not experiencing a jarring sensation of jumping from your style to mine and back. Whether it is the use of adjectives, the use of metaphors, or certain turns of phrase, whatever they are. 

We have the tools. We have the ability to make that kind of analysis and comparison about the style, obviously the content as well. And there may be many other things that we can also achieve exactly because the 20-fold increase in speed and productivity is so breathtaking and so astonishing that if we are able to internalize it and to maintain it, then we have an enormous amount of energy, time and resources to apply novel tools to things that a traditional book writing process doesn’t allow. And that is also part of the reason why I think that not only the title of whether we are the writers, authors of the book may need to be changed – a different label may be required – but what we are creating may be a different thing. Through the process of creating it and what it is and how it is experienced, it can become a different thing.

Massimo:

I share the goal of aiming at something that improves our way of working and the speed of how we work. And this is absolutely exciting. I mean, I’m in for that. At the same time, I feel like when you talk about increasing the actual speed in productivity, I would like to include also the quality. And you are, in this sense, aspirational. So you’re talking about a day when we will be able to do that while creating something which is high quality. 

I want to leave this maybe for another conversation, because we didn’t decide what is the quality we are aiming at, and it’s almost, you know, implied. It’s implicit in the process we are doing that we are playing a seriously crazy game, which is serious, still a game, so we know we cannot do it. That’s why I like it. But in the end, then we will have something. We will have an artifact. 

And so I’m wondering, I know already your reply, but will I care about the quality of what I will have written, you know, in 30 days? Because the answer is, of course, I will care. And I already know that I will not be satisfied, but not in the sense of, you know, I’ve been writing so hard for many months, as you were saying one year, two years, and I put all of my efforts into it. I did the research. I know, you know, by heart, everything I can talk about it, and now I’m not satisfied with the final outcome. That is not the feeling. 

My feeling is I was in a rush. I mean, I was on a roller coaster. I did my best, and this is what I have. But, yeah, what is it? Is it… man? Yeah, I know I knew that, so I don’t think I would want to ask experts to review my book in a month and say, you know, and enthusiastically say, “What do you think about it? I wrote a book, so do you like it?” I think it would be really stupid. 

So I am already accepting that, because the value I want to get out of this is me and you doing this thing and producing the evidence of what we have done. And I think I’m already writing another book that is how to do it, and that is in first person. It’s my experience together with yours. I have supporting evidence that I researched myself and I can talk about it for hours. That is exciting in the first place. How do you feel about what I said?

David:

It is necessary to agree on what our expectation is in terms of quality. And quality can be quantified. Quality can be improved. Quality can be set as a goal and achieved and exceeded. So we can start by giving some fixed points. For example, I already mentioned, but it is worth repeating, I will expect the final product not to contain hallucinations. I would also like the final product to have as many references as possible so that people who are eager to check what we are talking about can go look at the reference.

At the same time, I would be astonished, as well as exhilarated and immensely proud, but I feel that it is not appropriate – it would not only be immodest of an act of hubris to expect that what we produce is such that it will advance the field. We are not going to receive the Pulitzer Prize or an honorary PhD degree for the book, and it’s okay. Maybe next year. As a matter of fact, I fully expect someone, not necessarily us, to achieve exactly that. And rather than earning a doctorate in five years’ work, to earn a doctorate in three months’ work, because of their ability to manage knowledge and to handle AI tools at a degree that today would be described as superhuman, right? 

And very, very concretely, I already feel that way. It’s just that I am actually more interested in the process than the subject, which is fascinating, but it is not my specialty, just as it is not yours. We could have picked some other subject. And I think it is good that we pick this one, because it is appropriately self-referential. And I think it is very good reinforcement of what we do, that what we talk about is also the kind of tools that we are using to create the text and so on.

Massimo:

I have a question for you. So, yeah, as an independent professional, working for 30 years or something like that, my greatest asset was my reputation. So the most important thing that I have is my reputation, because if you don’t trust me, I don’t think you would want to work with me. I might be, you know, naive, but it worked for 30 years. So I’m concerned about my reputation if I am so naive to say that I published a book about something that I can superficially know and then I want to share with the world, because I believe it is worth it. So am I stretching it too much if I say that my reputation is at stake with this project?

David:

I don’t know. If you feel that it is, then maybe it is. I hope that the people who work with me feel that I am trustworthy, and after having worked with me, they confirm their assumption, because for the first time, it will very likely be a little bit of a leap of faith. Dependability, reliability, accountability are all things that one needs to experience, and then they will have a higher probability in a second collaboration with the same person to be able to say, “Oh, yeah, I already know that guy.”

In my professional life, I have always been very happy to be an outsider and to very freely talk about my ideas, my impressions and my thoughts, my analysis, my feedback of a given field – a large variety of fields. AI to start with in the 80s, then cybersecurity and video conferencing in the 90s, speech recognition, Internet of Things, then online collaboration, large-scale crowdsourcing, online video, and so on and so forth, year after year, decade after decade, happily moving from one field to the other.

For whatever reason, either because what I would be contributing to the field was seen as valuable, or because I would claim the things that I would say with an air of authority that no one ever questioned, I would acquire a certain believability, if not even thought leadership, on that field with a speed that is, again, not reasonable – just very, very fast. 

Now, that means several things. One, that anyone looking at my professional history has a very hard time placing me, and rightly so, because I move around a lot. The other consequence is that I am not beholden to a very specialized, vertical network of peers who have the power of upholding me or destroying me because of a false step that may ruin my reputation in that particular network upon which exclusively my professional life depends. And that makes me free, as well as it makes me able to take risks that other people in other situations may not be able to take. 

I have, of course, my own standards and my own expectations, and that is why I’m saying what we end up creating must meet those criteria. I just gave one or two, and we may end up with more. But the level of self-knowledge that I have assures me that as long as those self-imposed criteria are met, I don’t have any concern about being worthy of my output.

Massimo:

This is one of the highest value you delivered in this session. So thank you for this, because it makes me think and maybe asking if you have time, or maybe the next time, another intentionally stupid question. So you are David, you explained clearly why you want to take this risk, why you have certain boundaries that you know you might trespass or not, but you are safe. What if I don’t feel the same? Am I not very wise in taking this risk? And of course, you would say, I mean, it’s your business. I don’t know, you have to decide, but I want to – I mean, we’re friends – I want to ask you this question.

David:

Sure, so I definitely don’t want to put you in a situation that makes you feel uncomfortable. For example, for me, it would be very natural after editing with AI the transcript of this conversation to publish it entirely. But if you feel too naked and too exposed in communicating to me your vulnerabilities, I respect you, and I respect your history, your situation, your feelings, and your concerns. That is why I can very happily not only listen but accommodate any adjustment we want to make in the self-imposed objectives. 

So a simple example: we want to work intensively for the next month in order to see whether we can finish the book. And I already said that going from 90% to 100% will take time in all the other areas – fact-checking, referencing, typography, illustrations, and whatever else. Maybe, maybe not. It will fit in the 30 days; it could not. However, I want you to feel absolutely free to put the veto on publishing anything until you have addressed your concern satisfactorily around the reputational risk or around quality or around anything else. I will totally respect that, and I will not push you to do something that you don’t feel comfortable about.

Exactly because, as a matter of fact, I don’t care about the book. Publishing it or not publishing it, I don’t achieve anything by publishing it. It’s not that my next year’s grant or tenure or whatever achievement depends on publishing the book. It is because I gain from writing it, in the process of writing it, as much or more than at the end publishing it. Of course, it will be a pleasure and it will be a joy, and I will be proud if we do, but I will not do it or force us to do it at the price of making you feel uncomfortable.

Massimo:

David, thank you for that. I never had an ounce of doubt about your transparency and you being aligned with what you said. So I always felt like I could do anything that I feel like doing and being open about it. So no problem with that. Thank you for saying it aloud. It’s useful, but I have no doubts about that. If worse comes to worst, I would, you know, give you all of the material – you can do with it whatever you want. Because, as you were saying, which is something that’s not really nice to say, but I don’t care about the book as well in the sense that you mentioned. 

So I have an aspiration to be a book author. Maybe because I am childish, I think like I want to leave something about me to the world. And so I have this dream, and somehow I fear ruining this dream. What if the book is shit? Okay, now I am the guy who writes shitty books and it’s done. I mean, they will remember me for that. Sorry-

David:

I have to stop you, because this is exactly what you’re describing – the situation of when we met first, when you brought me my book, and the sentence you wrote in it is this, right?

Massimo:

Yeah, I remember very well the moment. Yes, yes. Look where we are now, working together on a book. Yes, exactly.

David:

So in my case, you were diligent enough, and you were patient enough to overcome your first impression.

Massimo:

Yes, yeah.

David:

Yes, and I am grateful for that. So on the other hand, I was courageous enough in my risk-taking to put out a message that only people who were diligent and patient could deeply understand, and my reward is the ability to be working with those kinds of people. I don’t regret missing the opportunity of working with people who are superficial and impatient.

Massimo:

Okay, okay, okay. At the other part, if you have time, is about me being creatively crazy. That’s why I’ve invented the term “CREAZEE project,” which you were part of, and we had a lot of fun with me doing impossibly crazy things, like being, you know, an English writing coach – me, who can barely sufficiently speak English. I was leading 15 people writing 450 articles in one month. So I’m accustomed to that. I like to play with things. 

And I was already putting all my reputation at stake writing a blog with a lot of things coming out of my mind without even rereading them. Sorry, readers. So I know I have to do it, because it’s the best way that I know to grow, and the rewards for doing that are immense. So I’m really excited about this. At the same time, you know, from time to time, I become, you know, normal again, with my feet on the ground. I say, “What am I doing? Am I really crazy? This is going to be a sort of failure. It’s impossible.” 

And so I need to play with the two contrasting feelings, which are part of me. This is who I am. I am constantly living with cognitive dissonance, which is – and I want to make it a tool. I want to make it an asset, not something that blocks me. And at the same time, I need to talk about it and say, especially to friends like you, how I feel about it. So thank you for listening.

David:

Of course. So do we want to spend a few more minutes discussing some actionable items?

Massimo:

Yes, I do. The process of writing as fast as possible of the chapters is good, because it’s, of course, going to be really full of all of the issues that I start to find, because we want to find them as soon as possible, and we want to have all of them fast, because we want to have a synoptic view about them, all of them, and then doing either manually or with the use of AI, of course, the process of extracting the pearls, the raw pearls, because those will be the actual building blocks of maybe a new version of the outline. This is intuitive, what I feel like it’s the best thing to do.

David:

Okay, okay, very interesting, and I like it. The way I would describe it is the refactoring of the book.

Massimo:

It’s like free-flowing, free writing. It’s a specific creative process in which you don’t care about being strict or correct. You just want to throw things out of your mind. So let’s, you know, put things out of the AI mind and see what happens. It’s fine. It’s good. Don’t worry. We know we’re patient and we are permissive. You can do mistakes. Do all the mistakes that you want. So this is the time to do the mistakes, but you have to do that fast. Then I want to stop. And now don’t bust my balls. I need to review and think. There will be precious to have a better outline. Perfect.

David:

Absolutely, absolutely. You made the remark that surprised me a little, where you said that it is set in stone. I feel the opposite, and now you’re saying that actually you believe that it can and very likely will be changed. So whatever outline we have is fine exactly because it is just a starting point.

Massimo:

I didn’t put the emoticon, still smiley, but that was absolutely ironic and sarcastic. It’s changing every minute. It’s changing each time I read it. It’s a different thing, of course, and it should be like that.

David:

Now, forgive me, because you confirmed that the process of writing rapidly the various chapters, including mining the depth of trillion-parameter AIs that help us finding things that we wouldn’t be able to – definitely not at the speed that we have today. And just to make it explicit, I wrote 2,300 words of chapter two, including revisions, including several passes to make the style in a given way, eliminating lists and whatever else. I did it in 29 minutes.

Massimo:

Okay, good for you. I’m not able to be that efficient and effective. You will need to be so kind as to teach me how to do that. But I don’t think it’s the most important aspect.​​​​​​​​​​​​​​​​

David:

I just mention it and describe it because of what it potentially enables. It enables exactly rewriting the whole book five times over. 

Massimo:

Yeah, no. Even more than that. That’s exactly what I was saying. Still, we can be faster. We can be quicker because we don’t need to review the style while we are producing the draft. So we should consider what we’re creating as drafts. So give me the draft. I don’t care if it is refined. I don’t want it refined. Give me as it is, because I want to put it in the repository of all of the chapters so that I can have an overview. When I will have a valid structure that makes sense, that I can read and say, “Yes, it makes sense,” then we will care about the styling. So styling now is useless. I don’t care about it.

David:

Okay, and you mentioned also producing a new outline, etc. So these are very useful remarks, but they are not questionable for today, tomorrow and the day after tomorrow. Let’s-

Massimo:

I will keep the plan as it is. Keep the plan as it is by writing the chapters, because we’re going to end by the 15th of this month, which is much earlier than we were expecting.

David:

Okay. Two things, if possible, on my side. Am I wrong or I didn’t yet see your drafts in the drafts folder?

Massimo:

You’re wrong. Chapter two, Theoretical Foundations, is there with all of the different versions, my comments, references, structure is there.

David:

Okay, perfect. Sorry if I missed it. No problem. I will take a look. Second, I gave you access and invited you to tweet and to write blog posts. Are you hesitant to do so, or I don’t know what to write about, or what stopped you until now?

Massimo:

The fact that I didn’t know what I was thinking about it. I didn’t know what I was doing, and so I would have ended up by writing “I don’t know what I’m doing,” and I don’t think it’s really exciting. I have, as I said, since yesterday, a different angle on this thing, and I think that telling the story of what I am discovering and feeling would be useful. This is what I would like to write. I have private notes, drafts that I can use to start tweeting about it. I would be more on the personal experience side, rather than the actual topic or content. Would that be relevant or not?

David:

Yeah. So, for example, you could talk about Figma and how you are using Figma for understanding the relationships of the various pieces of content. That is perfect, absolutely.

Massimo:

Okay.


Comments

Leave a Reply

Your email address will not be published. Required fields are marked *