The Leanpub Podcast 🎙 Feat. MaxGuernseyIII, Author of No Direction

Len: Hi, I’m Len Epp from LeanPub, and in this episode of the LeanPub Podcast, I’ll be interviewing Max Guernsey III. Max has over three decades of experience as a software engineer, architect, and consultant, and for more than a decade, he has helped leaders diagnose and remove the root causes of software delivery problems. Max is the author and co-author of a number of LeanPub books, including Fixing Software Delivery, The Engineering Leader’s Field Guide, The Other Half of Coding, What They Didn’t Teach You, and the novel No Direction, The Severed World Part 1. In this interview, we’re going to be talking about Max’s various book projects, how he’s automating the release of chapters in his serial novel. We’ll talk specifically about Fixing Software Delivery and The Other Half of Coding, and how he manages multiple book projects at the same time. And then we’ll just have a general discussion that might include some disagreement about books and writing and AI, and what that world looks like from Max’s perspective as a self-published author. So thank you very much, Max, for your repeat appearance on the LeanPub Podcast.

MaxGuernseyIII: Thank you for having me.

Len: I guess we could start by talking about how you’re automating the release of chapters in your serial novel.

MaxGuernseyIII: Yeah. So, you know, I have a — I don’t know how common this problem is. I’m guessing it’s fairly common in our industry. People who want to publish books incrementally, but maybe we have a day job, right? But some weeks I’ll write like three chapters. Some weeks I won’t have time to write one. And I’m trying to have my novel released every single week. So I set up a GitHub Actions workflow that runs every day. And it checks for little files that are like, you know, a books.txt.schedule.date, right? Something like that. And if it finds one, then it goes and it copies it over books.txt and pushes that and then runs a release using your API, which is really cool. It works really well. It’s pretty straightforward. And, you know, there are little details, like if there’s a release email with the same date, then it’ll consume that. So it kind of is this secondary agent that’s in there doing the serial release part of the book. And it’s gotten to the point where when I talk to someone about what’s published, I don’t even think about what’s actually published. I think about what’s in that hopper to go out.

Len: Oh, that’s really interesting. And so how are you contacting people about this? You mentioned an email.

MaxGuernseyIII: Yeah, I’m just sending out the emails with the API.

Len: Oh, okay. You know, it’s funny. I’m not sure that you could do that.

MaxGuernseyIII: We’re working on that part, right? But I think that might be there now. Right. But yeah, I see what you’re saying. Like when I reach out to someone to give them — I can’t call it an advanced reader copy because the book isn’t done yet. So I don’t know what you call it, but like a friendly reader license coupon. Right. Someone I want to read the book. I want to get their opinion on the book. I want to get their feedback on the book. I’m talking to them and they say, oh, you know, it’s published through Chapter 14. And then when the conversation is over, I’m like, no, when I was talking to them it was published through Chapter 12 with two more in the hopper. Right. So, yeah.

Len: And it’s interesting — I’m going to do something. This is the very first time this has ever been done on the LeanPub Podcast. There’s the cover. So do you have a kind of length-per-chapter kind of thing, like a word length that you shoot for or something like that? I’m just curious how systematized the process is.

MaxGuernseyIII: It works out to a rough mean. I can’t remember what the length is, but in terms of the PDF, I think it ends up usually being about 12 pages, can go up to 20, be down to 10. Right. But really, what I’m trying to do is have one small narrative arc that advances the larger narrative within the larger story. Right. Each chapter is like a little story that leads you towards the conclusion of, at this point, Act One. Right. And then when the act completes, that will lead us towards a big transition that moves us towards the conclusion of the book, which itself is an installment in a four-part series.

Len: Okay, that’s very cool. So it’s kind of each — sort of like a cliffhanger at the end, kind of driven kind of thing.

MaxGuernseyIII: Yeah, something like that. There’s an objective I want to meet. And however long that takes is how long it takes. And it varies by, you know, up to 100%, right?

Len: Okay, okay. Yeah, that’s very cool. One thing I wanted to share with you — I don’t know if you’ve actually seen this on your landing page — but we’ve got this little progress thing that we show now. And this was — so for anyone watching, if you’ve ever wondered if LeanPub is responsive to authors, I think Max actually asked for something like this.

MaxGuernseyIII: I did.

Len: And we went and built it. So yeah, so now there’s something that’s showing a little — forget what these used to be called in Excel, but there was a term for them — but the little arrow going up. And yeah, and then just some word progress and stuff like that in the last 30 days, which is particularly useful for people who are like, oh, has this been updated? Is this author really working on their book, et cetera? And you can see this author is very actively working on their book.

MaxGuernseyIII: Yeah, that makes me happy. I’m very happy to see that badge. I knew that those badges were coming because you told me they were, but then I didn’t see the badge, right? And I’m like, oh, I don’t know what it is. And then I saw that it had to do with 30 days and I’m like, well, I’m going to have to wait until the book’s been around for 30 days before I see that. Yeah, that’s really cool. I like that.

Len: Yeah, good. Thanks for the really good suggestion. We need to have things like that pulled out of us sometimes, and that was a really good one. And so just the last question I have about that kind of thing — so you’ve automated it using our API and stuff like that. And do you have — so you said it’s kind of like indeterminate how much time you’re going to have in the week to work on your book, which is pretty common for busy professionals and stuff like that. So do you have a — just because people are interested in this — do you have a place where you go to write? Do you like have a way of getting away from your family or something like that?

MaxGuernseyIII: No, I just write in my office. Same place I would do anything else. So the same place I would work on a GitHub project or for my boss. I’m not like that in general, right? I remember — this is quite a diversion, but I’m going to tell you anyway. I remember maybe 15 years ago, people were talking about, I work in large blocks of time. I need a lot of focus. I need this, that, and the other thing. And I would listen to them and finally said, I never understand it when people say this. Do I just not focus the way other people do? And so he said, yes, you don’t focus the way other people do. And it’s because I work in such little cycles that I actually focus apparently more than most people. He said he’s actually seen someone really important walk up to me to shoulder tap and get completely ignored for like 90 seconds while I finish whatever it is I’m doing. And then I’ll turn to pay attention. And because he knows it’s only going to be 90 seconds, he doesn’t get uppity about it. And so yeah, I just don’t — I don’t have trouble tuning out the environment the way a lot of people do.

Len: Yeah, yeah, that’s really interesting. And it’s a testament to the fact that there really isn’t a single model for being a productive person or anything like that. There’s a theoretical physicist named Sean Carroll who’s got a podcast called Mindscape that I like to listen to. And in addition to Mindscape, he does a bunch of other things, including writing books and being a professor and going to the Santa Fe Institute and stuff like that. And people ask him like, how do you do it? And he’s like, I don’t know. He doesn’t — he’s like, I don’t have a system. I just, you know, I constantly feel like I’m failing at being productive enough, you know? So maybe that’s the secret or something like that. But I’ve definitely, personally, tried it out, like kind of separating tasks or types of tasks by environment and stuff like that. And it never really seemed to make much of a difference to me. So yeah, I’m kind of with you on that.

MaxGuernseyIII: Okay. I can imagine like if my kids wouldn’t leave me alone, right, like if they were coming in and spilling macaroni on my keyboard all the time, I can see how I’d need a place to go.

Len: Right, right, right. Cool, so let’s move on. And let’s talk a little bit about your book, Fixing Software Delivery. Let me try my new toy out here. Okay, there we go. So what progress have you been making on this book lately? You and Lilino.

MaxGuernseyIII: To be honest with you, that one I haven’t added to lately. So, you know, maybe a few typos that I noticed or whatever. Somebody would tell me there’s a typo and I’ll go fix it. This one, where I’m really at, is I’m just trying to figure out how to find — not how to find its audience, as in how to identify them or select them, but how to get in front of them and say, hey, this is the book for you, right? Like how do you — I still struggle with cracking the nut of getting in front of an executive and getting my foot in the door and saying, check this out, you don’t want to read a whole book? Great. This is a book that you don’t have to read all of. This is a book that’s structured so that you can quickly scan the table of contents and then quickly scan a one- or two-page executive brief. It’s just been terrible at the outreach part of marketing.

Len: So it’s really hard and it’s definitely one of those things where if you feel like you’ve always been on the outside and other people are on the inside, but it’s a mystery, then it can be very discouraging. So one of the things that, if I can try and offer some advice to you and to anyone listening, because that’s basically every author’s problem and it’s a very hard one — one of the things that we haven’t been good enough about encouraging authors to do is to keep track of conferences that are in your area, that your book is about. And typically, and this is very specific, conferences will often have dates, start date and end date of the conference. It’ll typically be in a place, but they’ll have like social media hashtags and stuff like that. And so what you can do is you can organize a coupon discount campaign around the time of the conference and say, hey, I’m not affiliated with the conference in any way, but I wish I could be there, or I’m a big supporter and stuff like that. So anyone who wants to can use this coupon code to get my — and the coupon code could be like softwareconf2026 or something like that.

MaxGuernseyIII: I actually remember you pointing that out in your tip of the day and just never put that to action, but that is a really good technique, right?

Len: Yeah. And then you’re riding an existing wave of discussion on social media.

MaxGuernseyIII: Exactly. Exactly.

Len: Well, so that’s something else I can — I’ll actually be doing a tip of the day about when I get a little bit more developed on it, but I’m working on something along those lines. But definitely, like, there will probably be — and the interesting thing about piggybacking on conferences like that is that there will be more than one.

MaxGuernseyIII: Yeah. And there’ll be all around the world.

Len: So you can kind of create these campaigns, and you can — with LeanPub coupons, as everyone knows, you certainly know — but you can set the start date and an end date. You can set a maximum number of uses, for example. So like, you know, big supporter of the conference, just for anybody there, if you enter this code, the first hundred people get my book for 50% off or something like that. And on that note, I wasn’t planning on doing this either, but we actually released this recently. This is leanpub.com/coupon. So LeanPub always has — our coupons have been coupon links. So clickable coupon links, but the coupon link has a little code at the end, and it can either be automatically generated and be kind of like unguessable gobbledygook, or it can be something that you decide, like softwareconf2026 or something like that. So now what you can do — and links are great, they’re easy, you can click on them and stuff like that — there’s one big problem with them: you have to be using a device. And so now what you can do is, if you’re talking to somebody, if you’re on a podcast for example, you can just say, go to leanpub.com/coupon and enter this code, and you can get a discount on my book and stuff like that.

MaxGuernseyIII: Yeah, that’s interesting. So is it a distinct code from the code that you enter? Like, is it something that distinguishes, like, also has the idea of the book in it or something?

Len: No, no, it’s just exactly the code that you enter when you create your coupon. So what happens if, um, like two people are both trying to pick up the same —

MaxGuernseyIII: This is the great thing about LeanPub authors being programmers — they’re smart in that particular way.

Len: Both will show up. So if multiple LeanPub authors are using the same code to put on the end of their book to promote the same — to associate with the same conference — every LeanPub book that’s on sale in association with that conference, if you enter that conference’s code, they’ll all show up.

MaxGuernseyIII: So then that’s incentive for us to like talk on the forum and coordinate, right? If there’s a bunch of us who are all trying to hitchhike on the same conference, it behooves us to all use the same code.

Len: Exactly. Exactly. And all get the same — share the exposure. Okay. Oh, you just reminded me of something. Do you know we have brand new forums?

MaxGuernseyIII: I do know you have new forums. I haven’t looked at the new forums, I’m sorry.

Len: I think I did a tip of the day about that, but anyway, that’s enough of me sharing this stuff. But yeah, so when it comes to getting things — and now, for example, because we have this feature, just to go back to what we were talking about — if you do meet an executive, you can say, hey, by the way, go to leanpub.com/coupon and enter this, and you can get my book for free or something like that. You’d want that to be a unique one, but anyway, hopefully that kind of thing helps. But definitely when it comes to getting in the door, and the thing about conferences is that there are people of all levels within their organization who are at the conferences. There’ll be keynote speakers, stuff like that. And you can use social media in various ways, like, hey, so-and-so, I see you’re giving the keynote. I’ve made this coupon for you. Or you can use your creativity along those lines. But definitely one thing is conferences. Definitely speaking at conferences is another really awesome thing, and one thing we encourage authors to do is definitely apply to conferences, see if you can get your foot in the door, see if you can speak at conferences, and on your last slide you are going to want to have a cover image of your book and a QR code, which you can make with LeanPub coupons as well. So that’s another thing to do. One thing that I think maybe we’ve spoken about — I don’t remember — but I heard about this from someone years ago, which was basically make print copies of your book and send one to the CEO of the top 100 Fortune 100 companies, or whatever the equivalent is in your industry, address it to the CEO. Maybe it’ll get through to them, maybe it will probably get blocked by their secretary or their EA or something like that. But you can’t win if you don’t try. And they might be — what’s this, interesting — someone sent me a free book, just a free copy of their book. So that’s another really good way of doing it.

MaxGuernseyIII: That really gets to the heart of it, right? So I’ve got a friend who’s, I think, really good at breaking in. He’s in a completely different industry, and admittedly that industry is like physical operations and the doors are open and they can go in. But there is just something about people who are really good at outreach, that they just get in there. They just get in there, and I don’t know if they’re comfortable or not, but if they’re not comfortable, they’re willing to accept the discomfort of just insinuating themselves into somebody else’s conversation or life. And that’s just a challenge for me, even if I know it’s in their best interest, right? I know this person’s life would get better if I can get in front of him and get him looking at his problems a different way and get him solving the real problems, right? And so I can tell myself that all I want, but there’s a psychological challenge that has to be overcome to just kind of do that kind of cold outreach.

Len: Oh, I mean, I identify with that 100%. I can go to a conference where people are apt to meet and talk to people about their subject, and I will not talk to a single person, and somehow feel like I’m doing something wrong if I try. And so I’ve definitely got that psychology myself. This is kind of a moonshot type thing, but another specific piece of advice that we’ll give to authors is keep an eye out for journalists who are writing on your topic. And if you see someone write an article that’s adjacent to, or right on, the topic that you’ve written about, send them an email making it clear that you’ve read their article. Not a canned email — making it clear that you’ve read their article — and saying something interesting to them that you bet they don’t know. Because as a journalist, what they want — they get a million emails a day, very few get through their filter because they know what the crap looks like. But if you can get through and you give them something interesting that’s new to them, and they’re like, oh, I didn’t know that, why didn’t anybody ever tell me that? They might reply to you or they might not, but now you’re on their radar. And what you’re really aiming for there is to become the person that they contact when they want a quote about something. And then your name starts to become — in that sector or whatever — your name gets in the public awareness space. And then when you’re at that conference or something like that, people are like, oh, you’re so-and-so, oh, I see you in the news all the time, or in the tech journals or whatever, talking about this and that and the other thing. I actually got a long letter published in the Atlantic about LeanPub that way. And it was basically — someone had written an article about KDP and pricing, Amazon KDP and pricing or something like that. And so I wrote an article that was essentially promoting LeanPub and saying like, we pay higher royalties than they do. But the way I got in was by saying something actually genuinely interesting that the person didn’t know, which was in this case, I think it was something like, did you know that they actually pay a lower royalty rate when your book goes above $9.99? They recently increased that to like $12.99 or something like that. But that was interesting to people, right? Why is Amazon punishing people for charging higher prices for their books? And what does that actually say? So that’s a very highly recommended, very low chance of success approach. But if you read the article anyway and it is something you’re interested in, keep a journalist list with their names and email addresses. And then when something comes up — and they’re usually easy to contact and find — send them like a short email with a quote, basically a quote they could use about something. They probably won’t reply. They probably won’t read it. But that’s another way to kind of long-game your way in your career into that kind of world. And the cost is basically nothing, right? So if you send out 1,000 of those and one of them turns into a quote, you’ve paid 1,000 emails and you got a quote in.

MaxGuernseyIII: Yeah.

Len: Yeah, I’m just reminded actually, just to give an example. So I have a little hobby of being interested in autonomous vehicles and electric vehicles and stuff like that. And so years ago, I got an article published in TechCrunch about basically how nobody understands that when there are real autonomous vehicles out there, it’s not going to be the same as it is now where the cars are driving themselves. There’s going to be all kinds of vehicles. Like there’s going to be mobile autonomous ice cream trucks. Obviously, you know, like when you think about it, right? There’s going to be mobile R2D2s for the cops. You know, if there’s an outdoor event, there’s going to be autonomous vehicles selling soft drinks and whatever. And just use your imagination to think what else could be on the road, once things can drive themselves, that’s useful. And I immediately started getting — I think it was that one — I started getting basically people, journalists, approaching me as an expert on autonomous vehicles and stuff like that. One article published. I also got on the mailing list for an anti-autonomous vehicle, anti-EV outfit based in Michigan that was pretending it was associated with the University of Michigan. And so they would start sending me their anti-autonomous vehicle propaganda to try and influence me. But anyway, that’s another way of saying too — one of the reasons to follow these journalists and read magazines in your space is like, often you can actually find your way into publishing your own article. Like I just submitted it to TechCrunch and got on there. So that’s another way of thinking about approaching things.

MaxGuernseyIII: It’s a really good idea. And the idea of being approached by an organization in Michigan that’s anti-stuff is pretty appealing to me.

Len: Yeah, that’s more on brand for me. Okay, so the next thing — let’s just move on to talk about your book, The Other Half of Coding, What They Didn’t Teach You. Have you been working on this book lately?

MaxGuernseyIII: Well, I mean, it’s finished, right?

Len: Right, okay.

MaxGuernseyIII: I haven’t been working on it very much because again, I’m sure it will change. This is another one where I’m sure I’ll have to do some kind of major revision, but I’ve got to figure out how to — I wouldn’t say I’ve got to figure out. I think I’ve got this one’s marketing figured out, right? But here, all the management costs right now is YouTube. YouTube is easily my most successful attempt at marketing. Although Pinterest has been shockingly useful lately. But YouTube is still number one for me. It’s like you put out a little video that’s a refactoring video or something on the subject of The Other Half of Coding or adjacent to the subject of The Other Half of Coding, and then you run it through like Opus Clip or whatever — one of the few uses of AI that I’m friendly towards, which is selecting from a larger work — and you push a button to schedule that all out and it just drips it out and you get, you know, one jackweed saying this is a worthless video, and 10 people who are like, oh, this is pretty cool, and a couple dozen subscribers out of the deal. Right. So it’s like, that has been like a straightforward, just simple — other than a few bucks for Opus Clip — a simple, like, input time, output subscribers. I imagine when I get my subscriber base up large enough and I start really pushing it, right, and I start writing videos where I say, well, you know, in The Other Half of Coding I say whatever, yeah, right, then it’ll start driving traffic over to that. And then once I’ve got traffic there, I’ll get more people who are like, I don’t really think this is very important, or hey, you missed this critical aspect. Probably some — it’ll be like, hey, why did you write a whole book about doing this to databases and then not mention it in a book that’s about the theory of the same thing, right?

Len: Speaking of YouTube — when we were corresponding before this, I think you mentioned that you’re setting up or thinking of setting up a YouTube channel specifically around the Severed World series. Yeah. So for a lot of people listening, especially if they’re like self-published authors or aspiring self-published authors, they’ll be curious how to do that. So what’s — I mean, you’ve already obviously got, you know, you’re publishing YouTube videos and stuff like that, but now you’re setting up a dedicated channel for a serialized fiction book series. What’s your current thinking or approach to that?

MaxGuernseyIII: Well, so first of all, I am doing a lot of reconnaissance on Twitter first — or I guess X is what I’m supposed to call it now. Because this is a hard thing, right? So as we’ve talked about in the launch video at least once, it’s easy to take even an unorthodox nonfiction book and talk about it. You’re the expert. You just talk about the stuff. You say, here’s the structure, here’s how it works, here’s what it’s for, whatever it is you want to say. But when you’ve got this whole book that’s built on this kind of Robert Jordan-esque — or at least a pale shadow of Robert Jordan-esque — slow burn, drip reveal kind of model, how the hell do you talk about it without giving stuff away?

Len: Yes. Yeah, yeah.

MaxGuernseyIII: So I’ve been working on that. Like trying out different ideas on Twitter, and trying to build up in my mind like what works, what gets people’s attention. Bonding with like-minded people who I think are the audience seems like the number one thing that’s working. And so I’ll probably try to do that with YouTube in a way that’s compliant with their terms of service, which might be a little bit of a challenge, but I’ll figure it out. So, you know, creating an affinity between them and me in the first place is one thing I’ll have to do. Talking about the themes — I’m starting to try to figure out how to do that. But like, you know, themes like when you try to gain leverage over someone and then you find out that you became the lever for someone else, right? Like that’s a trope almost, right? Like it’s used. And so it’s not giving anything away about the book, but it’s not just a theme in books — it’s a thing that happens in real life, and that’s why it’s a theme in books. So you could talk about that. That’s like a safe subject to talk about. And then I’m also trying some stuff out on Substack, see what sticks there. And then when I’ve got a good enough idea of what I can make a video about, I’ll start doing it. My original idea, honestly, has been pretty much thumbs down by everyone I’ve run it by. But my original idea was — it used to be that when you were an author, you would take a book to a bookstore and you would read some of it. I mean, I would never have done that with Test-Driven Database Development, right, but a fiction author would read a segment of it that didn’t give too much away and that got people interested in their style and the world that was being built. I thought, well, what if I did that? What if, once a month, I did a reading from the book with some sort of video element to it. And I still haven’t taken that off the table, even though like everyone I talked to just says, dude, that’s a stupid idea, you shouldn’t do it. I still have a gut feeling that maybe we can bring back the reading and resurrect it in a digital form.

Len: So I’m curious — the people who’ve rejected it, do they ever give you any kind of reason why they think that’s a bad idea? Like it’s just lame, or who would want to watch you read your book or something?

MaxGuernseyIII: Yeah. I mean, honestly, probably that’s — that’s really what the answer is about to tell you, is how bad I am at market research and why I lean so heavily towards making stuff rather than trying to research markets. Because I did not ask them, and I probably should have. It doesn’t occur to me to do anything that might smack of pushback on feedback. But I don’t know what it is. I think it might be that — you know, my voice sounds a little bit to me, at least when I hear recordings — sounds a little bit like Eddie Murphy saying, you’re not going to put the banana in the tailpipe, you’re not going to fall for the man in the tailpipe. So my annoying voice might be an issue. It might be that like the structure of the book, they don’t think it’ll lend itself to that. But I think I could still cut passages out that might work. I don’t know.

Len: Well, I think maybe you don’t think it’s a stupid idea.

MaxGuernseyIII: No.

Len: Well, the thing — I mean, the first thing that comes to mind to me is that one of the ways Mark Twain and Charles Dickens made the money they did was public readings. Like they would fill auditoriums. And that was a famous way for authors to make money in the past — reading tours. You’d show up in a venue and you’d sell tickets and you’d get money from people who bought tickets to come and watch you read. So in the olden days, that was a very popular thing. You’d go on a reading tour. And now you have to be kind of — obviously they were famous — but they got famous by doing things like that partly, and you get better at it over time and stuff like that. But the public reading — I mean, to this day too, it’s not like you’re going to fill a stadium or a big auditorium probably like you used to. But if you’re a big famous novelist, you could certainly fill up like a medium-sized auditorium for a book launch reading or something like that. And that kind of stuff does still happen. But it’s interesting to think about translating that to — and I’m sure there are lots of people out there who’ve tried this kind of thing — but I think that if you did a reading of each chapter as it came out and people knew they could go to YouTube and, you know, watch you do it — and actually this mostly probably people would be listening — but if the chapter comes out and there’s a reading, you can point from each to the other. People then might discover your book through your video. And you know, some people — my girlfriend for example, who starts books by reading the last chapter first — there are people who like wouldn’t mind, they wouldn’t be concerned about any reveals. They might hear, oh look, this self-published author has a cool fantasy novel that they’re serially publishing. Here’s a reading. And it might be you talking for half an hour, but they might listen to it and then be like, oh, that sounds good, and go get the book. So yeah.

MaxGuernseyIII: Yeah. That’s kind of what I was thinking, right? Like that’s the kind of reading I was thinking of — not a famous author filling an auditorium, but a fledgling author getting 20 seats in a Borders Books filled, right, to get their books known. So yeah, I mean, maybe I will try it. Maybe I’ll still try that no matter what the other sources of expertise say.

Len: Yeah. Well, that reminds me — one thing that we’re kind of shifting towards is trying to put more of our attention into helping people get attention. Not right now, but in the near term, that’s kind of one thing at LeanPub that we’re thinking of doing. And the thing that I think a lot of authors and self-published authors kind of somehow just don’t internalize is that a book manuscript is a kind of seed for multiple things, right? Like if you’ve got a whole — the analogy, you know how some forests are one tree? You can think of the manuscript as like the first tree and its roots go down and then they come up as new trees. And you can think of a whole forest over here. Like for example, a book could come up as a screenplay, and then from that root manuscript, you’ve got a whole forest, which is the movie. And then out of that, there can be merchandise, which is like, you know, a call of the dream scenario for a lot of fiction and fantasy fiction authors, right. But then there can be one of those offshoots, and a new little forest over here is all the public readings that you do. You know, another could be like posts on X or wherever that could be like a line from your book — interested in this, learn more, new chapter — like every time a new chapter comes out, there’s a bunch of life that can spring from that. And the thing about those things these days is that like a post stays up. It’s not in the algorithm probably, but it can be referred to, it can be linked to, it can be favorited, it can be, you know, what have you. So thinking — if you’ve gone to all the work of making a book manuscript, think of all the things you can grow from it.

MaxGuernseyIII: I didn’t see it as a theory, but I’ve seen a practical expression of it that now that you’ve provided the theory, it fits perfectly, right? So for No Direction, one of the things I had to do as part of world-building was construct a language, okay. And in constructing the language I needed to construct the script, you know. Me being a developer, when I built the script, the way I did that was I built like a program where I give it a string and it generates the images in the script, right? So I’ve got all these images that get generated in the script, and then I’ve got the chapter ornaments, and then I’ve got the, you know, this and that. And I — I ripped all of the AI out — the only AI was the art — I ripped all that out, did it by hand. You can tell it’s not done by AI. And then, you know, I was looking for advice on how to market it and what social media to use and someone told me, well, you’ve got all these assets, all these visual assets you generate, right? So why not throw them up on Pinterestand build like an organized library as you build these assets? You already have to build them for the book, right? I already have to build new glyphs, I already have to do all this stuff, right, to support the book itself. I generate this stuff and then I put it on Pinterest, and it’s — I mean, that’s worked like gangbusters. I mean, that’s just been getting huge numbers of impressions per day, driving a small amount of traffic to the book.

Len: Yeah, that’s a very interesting idea. That’s cool. Okay, an expression of what you said.

MaxGuernseyIII: Yeah, right. It’s — I already built the asset, and now I’m just like, how do I take that asset I already built and put it to another use?

Len: Did you know that we have chapter ornaments now?

MaxGuernseyIII: No, you didn’t tell me.

Len: Yeah, it’s in Markua now, I think, and it’s in our visual editor. So for anyone listening, if you don’t know what chapter ornaments are — if you’ve read like, you know, sort of fantasy novels and other kinds of things, there might be like, you know, chapter title and then like a little image, a unique image or a repeat image or something like that, at the top of each chapter — that’s called a chapter ornament. Max requested this a while ago and one of our co-op students went off and built it. So I haven’t actually promoted that yet. It’s kind of like we’ll do a tip of the day about that at some point or something like that. But yeah, that’s out there too. And I can actually — that’s a very compelling thing, because one of the things I was thinking was like, boy, it’s a lot of work to make these chapter ornaments. And of course so many books are works of love — you know, people enjoy what they’re doing anyway, so they don’t see it as kind of wasted time. But the idea of being able to surface your chapter ornaments on Pinterest is actually a really excellent idea. And then you kind of like a Pinterest board or whatever they’re called of your chapter ornaments, and like adding one — if you’re serially publishing, you know — adding one as your new chapter comes out, so your Pinterest board updates as well. I could also see you guys exposing them — you’ve already got the metadata, right?

MaxGuernseyIII: Yeah, checkbox — include chapter ornaments in table of contents on the web, right? So yeah, somewhere. Yeah. Yeah.

Len: That’s a very interesting idea. So the last thing I wanted to talk to you about today — I was hoping to talk to you about — is AI. And this is just because I know you’re obviously, you know, you’re a software engineer and all that kind of stuff. So the technical side of things, using computers to do things — this is something you do for a living. But you know, AI in the book publishing world — if you follow the book publishing blogs and articles and all that kind of stuff, it’s just been all about AI for a long time now. And I was just wondering what your opinion is or your approach is to AI as a self-published author.

MaxGuernseyIII: Well, I do use it, but not the way I think a lot of people do. This is also true of software — it’s less true of software. I don’t use it basically at all in software. But I’ll use it for research. That’s one thing I use it for, and yes, I accept that research is a kind of work, and that’s the kind of work I do with AI. I think it’s exceptional for that, and I think it makes sense why it’s exceptional for that, right? It’s curve fitting. The curve that it fits is massively complicated, right, with many, many billions of variables, but it is still a curve fitting machine, right? And search is kind of a curve fitting problem, right? We’re finding a pattern and finding things, right? So I use it to extract knowledge from the internet. It’s a great source for that. The other thing I use it for is anything kind of reflective. So I have a chapter review process that I’m putting everything in No Direction through. And then I put everything in a different process — but a process that I put everything in Fixing Software Delivery through. I don’t rely entirely on it. I still have human review. I review it. I get other people to read it for me, but it cuts down on the dumb errors, right? So like no spelling errors, that stuff. Where I don’t think it works is in generation. I don’t use it for coding at all. I don’t think it can code in terms of working code. I don’t think it can generate code faster than me. Certainly in terms of like doing the thinking — like a lot of people are like, oh yeah, either they think it works, they think it’s magic and it works and it’s faster than them, and I don’t believe them at all. Or there’s this other crowd that’s like, well, if I spend a bunch of time really getting my prompt right and really understanding what I want it to type for me, and then I type all of that into the prompt, then it’ll type this other thing out for me. And I’m like, well, but that’s the hard part, right? You’re saying if I do the hard part, it will do the typing for me at an exorbitant price and, you know, generating enough heat to fry an egg. So that’s kind of my position.

Len: And so you don’t use AI to like write?

MaxGuernseyIII: No, I don’t use AI to write. On very, very, very rare occasion, when I’ve put it through the chapter review process, the second to last step is readability. It will give me a turn of phrase that I really genuinely think is better. And I will copy that one turn of phrase in. But generally speaking, I don’t use it. I certainly don’t use it to write any kind of rough draft or draft type stuff. I’ll use it, like I said, as an editor feedback thing.

Len: Thanks very much for sharing that. That’s really great to hear. I think — you know, we’ve made a Ghost AI kind of skill in LeanPub and we’re very curious to see how people are going to use it. One of the contributions I made to that was the research. There’s a research kind of dimension to it. And, you know, I wrote a doctorate years ago — 20 years ago, 25 years ago I started — and that’s how old I am now. And if I’d — all my research was in folders on my computer. But if I’d been able to use like an AI type skill to kind of, oh, remember that quote about blah blah blah, you know, where was that, where was that from, who said that? And I could query my own research files in that way. Like that would have been an amazing thing. And so like so much of what you do in sort of academic writing is like finding quotes to back things up, for example, often. And so that kind of thing would be very — it’s not writing it, it’s finding it.

MaxGuernseyIII: Yep.

Len: You know, that kind of thing would be incredibly useful. Obviously there are things like bibliography, for example — this should be automated, 100%. Things like that should — bibliography should be automated. Yes. You know, thesis descriptions basically should be like for university libraries, basically be automated, in my view. Things like that, because they’re things that people search. And so you want a certain kind of uniformity in the way they’re generated, and stuff like that. Personally, I don’t like — just to give my point of view — I wouldn’t use AI to write things for me, like paragraphs and stuff like that, especially if I were writing a book. I actually wouldn’t use it to do that. Just because — well, not just because, but like in so many ways, like then it’s not my book, then I haven’t written the book, is kind of the way I’d look at it. And if you ask me questions about it, I wouldn’t be able to answer them, probably. And I would — this is my concern — I would be too distant from what it actually was. Now, if I was like, you know, carrying out some kind of campaign to achieve some kind of practical outcome in the world around me, my qualms or concerns about that would go way down. Because, you know, like I’m on a mission to achieve something. Like let’s say I want to campaign in my community to get a bike lane or something like that. I definitely would consider using like, rewrite this — and this is not a majority thing, this is like a standard for clarity and transparency on government websites — rewrite this at a grade six level, right. You know, that kind of thing is something I would definitely do. But anyway, my personal view is that things like research — for things like — we’ve got a thing in our skill which is like, act like you’re — I’m going to describe the role, play a reader of my book, the ideal reader I’m looking for is like this, what do you think of my latest chapter? You know, and that kind of thing can be very useful.

MaxGuernseyIII: That’s very interesting. Yeah. So I mean, that is interesting to me, right? That is how I use it now. Right. So that would be what I would use it for if I ended up upgrading. Ending up with credit. My other concern is — I don’t know how you — I don’t know if you run this on your own metal or something, but I believe the prices are heavily manipulated right now. They’re heavily subsidized, right? We’ve got all these sweetheart deals, you’ve got Nvidia putting money into OpenAI so that OpenAI can buy stuff from Nvidia. And it’s just this very — it’s got that smell, that bubble smell to it. And I suspect very strongly that at some point the rubber is going to hit the road and all the capital is going to — I mean, either the capital is going to pull out, or if it’s coming from the government, like the federal collapse or something, and it just won’t be able to prop those prices up anymore. And then I’m worried that the prices will go up by a factor of 10 or 30. And here’s the thing — I might pay 30 times what I’m paying now for research, but I would never pay 30 times what I’m paying now to have something write for me, or write code for me, right. You know, the high investment to go in where I don’t have knowledge and, like you said, find that quote, right — that costs, if that costs you 50 cents instead of a nickel, that’s worth it, right. But the generative part of it, I think, is just going to — I think price-wise, the actual generative part just can’t stay.

Len: Yeah. Thanks for sharing that opinion. I share a lot of that perspective. You know, being a former investment banker who worked on multi-billion dollar M&A transactions, like looking at the numbers, it’s like the first question is, where is the money going to come from to recoup your investment? And like the kind of M&A I did was like in utilities, right? So it’s like, well, we projected increased electricity usage because of population growth in that region. You know, but like literally, where’s the money going to come from? Like if you’ve spent $300 billion on your AI venture, who is going to be giving you their money to recoup that investment, let alone make it profitable? Like literally who’s going to be doing it. And if it’s consumers, then oh my goodness, well, they’ve got limited budgets. So now they’re not spending it on something else. So that’s like, just looking at it from that analytic perspective, it’s like, oh my God, what other industry is going to die because people are now spending all their money on AI, right? And it’s like, so are they going to cancel their Netflix subscription because they want to use that money on ChatGPT, like, you know? And so like, it’s — you know, there are huge numbers floating around. And partly — this is not an original thing — but like one of the consequences of the dot-com bubble bursting was the internet. It was built — there was a bubble, but people built all this actual physical infrastructure and did all this really hard engineering. And you know, like all these companies collapsed, but we were left with the foundation that our world has been built on for the last, you know, 25 years in the aftermath. And it’s possible that one outcome of all of this is that there’s left behind an incredible amount of computing available to be used to do things. And even if the companies collapse because they can’t service their debts, the things they built will remain behind and then someone — it’s capitalism — will come in and buy it for a song. And then their capital requirements or their investment requirements to get a return on investment are going to be way lower than the people who lost money building it. And then they might find something to do with it after. And that compute can be used for all kinds of things, right? It is being used for neural networks, but it could be used for heat transfer simulations, right? You can do all kinds of things with that kind of low power, massively parallel processor.

MaxGuernseyIII: I agree, there’ll be an outcome for sure. There always is. Whenever a bubble bursts, right? There are a bunch of beautiful houses on the hills of Kirkland that were empty for a while, right? But the houses were still there when things bounced back after the crash.

Len: Yeah, thanks. I’m on board with that. So I guess the last thing is, do you have anything you’d like to log roll? Anything specific that you’d like to promote or point people at before I let you go?

MaxGuernseyIII: I mean, you already did it, right? No Direction is the thing that I’m working on now. No Direction is the thing I’m trying to get to people. It’s not for everyone. You might not be able to handle it. It might be too much for you. So don’t go read it if it’s not for you. Let’s see if that works.

Len: Awesome. Okay. Well, Max, thanks very much for taking the time to appear again on the podcast and for being game for such a wide-ranging, kind of random discussion about stuff. But I’m sure, especially for LeanPub authors, there’s going to be a lot of interest in here.