An Interview with Jeff Geerling, Author of Ansible for DevOps: Server and configuration management for humans

Published Jan 11, 2016 by Len Epp

Jeff Geerling is a full-stack developer, author and blogger. In this interview, Leanpub co-founder Len Epp talks with Jeff about his career, open source projects, his book Ansible for DevOps: Server and configuration management for humans, and how he made over $25,000 in sales before his book was finished.

This interview was recorded on October 16, 2015.

The full audio for the interview is here. You can subscribe to this podcast in iTunes or add the following podcast URL directly: http://leanpub.com/podcast.xml.

Len Epp: Hi, I’m Len Epp from Leanpub, and in this Lean Publishing podcast I’ll be interviewing Jeff Geerling. Jeff is an experienced full stack developer, who creates websites and applications for organizations ranging from small businesses to large enterprises. Based in Saint Louis, Missouri, he’s involved in a number of open source development communities and maintains the Drupal VM project at drupalvm.com. Jeff also blogs, writing on various sites including his personal website, lifeisaprayer.com, and he is also a photographer whose pictures have been used in news publications, documentaries, and various places on the web. He currently works as a technical architect at Acquia, a cloud platform for building, delivering and optimizing digital experiences.

Jeff is the author of the Leanpub book, Ansible for DevOps: Server and configuration management for humans. In this interview we’re going to talk about Jeff’s professional interests and his experience writing and self-publishing a Leanpub bestseller.

So, thank you Jeff for being on the Lean Publishing Podcast.

Jeff Geerling: Thank you, Len.

Len: I usually like to start these interviews by asking people for their origin story. So, I was wondering if you could tell me how you first became interested in being a developer, and how you ended up with your particular specialties and interests.

Jeff: I like telling people about it because — and I love hearing also from other episodes of this podcast, other people’s stories — because it’s so amazing in our field of basically anything technical, how varied people’s backgrounds are.

When I was growing up, my dad worked for a radio station — actually a set of radio stations — as an engineer, here in Saint Louis. He would bring me to work sometimes and I got the opportunity to learn right as the internet was becoming a thing. The radio industry was picking it up, and they had a lot of local networking. They started connecting their offices together and they started needing more IT operations. Basically I went from a guy who liked playing games on an old Macintosh, to a guy who had experience with Novell and Microsoft and early Linux-based networking — in the course of a few years, when I was just barely able to work.

So he let me work on different projects that he had there, he helped me to build computers until the point where I could build them and I started selling computers. I was helping schools and organizations with their networking, and with their server set-up. Eventually that all led into being able to do some early web development back when it was using like Claris Home Page to build yourself a web page, and sticking a picture in there, that kind of stuff. I actually had a website with some pictures of mine on it, I think it was 1996, an IP address somewhere. I don’t remember where it was, but it was on an old InterJet. So I got into networking and computers mostly from the hardware side, that was my fascination back then.

As time went on, I went to seminary. I was studying to become a priest, actually. And when I found that that wasn’t for me, I was still doing stuff with computers at that point, and learning a little bit of programming. After getting a degree in philosophy, I started getting into Drupal, which is an open source CMS for building websites. And I just so happened to get into it at a good time, with the field of turning the web into content-driven websites that all interacted with each other. Drupal was hot at that point, and I kind of just hitched my ride on that bandwagon and started learning a lot more about Drupal.

And then, as time went on, I also was getting more interested again in infrastructure, and building sites and high availability environments on the cloud. And so, as time went on and I did shell scripting and I learned some Puppet and I learned some other configuration management tools and systems. Ansible came along right at the right time for me, because I needed to grow some of the sites that I helped with, a lot bigger. And once you get beyond a couple of servers, you need tools to interact with the services that you use and automate everything.

Ansible was introduced in 2012, and I got into it in early 2013. At that time there was no book on it or anything like that, so I decided, “I’ll start blogging on it,” and then the rest is history. The blog turned into a few chapters of a book, and that few chapters of the book turned into a Leanpub project, and now I’m here today with this book and a lot of involvement in both the Ansible and Drupal communities.

Len: You have a particular interest in open source projects. I was wondering if you could maybe tell about your first experience with an open source project, and how you became interested in that?

Jeff: Yeah, actually, I’ve been involved in a few here and there, smaller open source projects, and I’ve used Linux for anything server related for years so…. But I never really got involved in communities that much. I read The Cathedral and the Bazaar, and that got me interested in the whole idea of the movement of open source. I wouldn’t call myself a purist, by any means, but I do appreciate the value of having your work be open and free and available — free as in beer.

So when I got into Drupal and I noticed that the community had such a strong base of smart people working together, there were no real boundaries, like, you didn’t have to be a certain type of person, or you didn’t have to do certain things. You just had to have an interest in making Drupal better, and people would accept that. I found that to be awesome in that community. And I can’t say that every community has that, but a lot of the open source software communities do have that kind of atmosphere, which I love because you get to work with such varied backgrounds, and you get to do things that you’re interested in. As long as it’s something that’s helping the project, you’re going to have a future, and you’re going to get people to help you and you’ll be able to help other people.

So I really got involved in Drupal first, and that community is huge. There’s, I don’t know, like fourteen or sixteen hundred people that helped with the Drupal 8 release, that just is about to be released. They’re in release candidate stage right now. And Ansible also grew from when I started. There were may be a hundred or so different people contributing, to now there’s twelve or thirteen hundred people contributing, from a lot of different backgrounds, from big enterprises to little one-man shops. It’s really cool.

Len: You’ve brought up the fact that people can come to development and to different open source projects from different backgrounds a couple of times. It was actually something, when I was reading your resume before — when I was preparing for this interview, of course I noticed that you have a background in philosophy and theology, which is relatively unusual at least in my experience for developers.

This is slightly left-field, but I thought I’d take the opportunity to ask someone with a background like that, and with obviously deep IT experience, what you think about recent developments in AI and the discourse around AI. Because we hear from people like Stephen Hawking or Elon Musk, but not necessarily from people who have a deep background in different subject areas or — it’s a bit of a trivial way of putting it, but like you do.

So I was wondering what your thoughts are, given your background in philosophy and theology, on what’s happening with AI and where it’s going to go and what it needs?

Jeff: Yeah, it’s especially relevant now that were seeing things like Elon Musk announcing that A, we need to be cautious about AI, and B, all of his cars are now talking to each other and building up traffic networks and things like that. I think we’re at an interesting moment where AI is not just science fiction, but we’re seeing kind of small bits of it here and there being developed. And it might not be AI in the sense that something is going to take over the world and kill all the humans at this point. But it’s already to the point where we can create networks that evolve themselves, and do things like fix themselves or build something entirely new. But not to the point where you’d recognize that it’s human intelligence.

So you know, I have a background where I believe that the ends don’t justify the means, so I think, in my opinion, it’s always about taking it as far as you can go — but also taking the time in all your research and things that you’re doing to make sure, “Is this going to be supportive of the human race and the ability to be a human?” And really it all stems down to what you believe it is to be human. I could probably get into a long debate with most of the people listening to this podcast, because they’re all technical people, and technical people love getting in to the detail about, you know, morality and what it is to be human and ethics and things like that.

But it really comes down to making sure that you have some sort of morality that guides your decision-making, and I think whenever we run into problems, it’s somebody who kind of mutes that part of their life because they’re so interested in developing something further, going down a path.

On the flipside, you need to go down those paths, but you need to use caution and it seems that like every disaster in history is due to more hubris than anything else, the idea that we can take something further without worrying too much about the consequences because it’s science, you know.

Len: That reminds me of one of the potential consequences of AI. I’m sure that most people listening to this are probably familiar with the conventional understanding of the Turing test. Where you know, Alan Turing proposed this test that where if you’re interacting with a machine, and you can’t tell if on the other end, where there’s a thing in between you and the thing you’re interacting with, if it’s a person or a machine, then if it is a machine and you can’t tell if it’s a machine or person, then it’s passed the Turing test. I wrote a blog post about this from another direction a while ago, but when I was reading about you, and on your blog, I sort of thought that there’s an interesting metaphor there, for a priest being on one side of the confessional and not knowing who or — and potentially, in the future, what is on the other side of that?

For example, in a more trivial way, if we actually have computers, whether they’re actually thinking or not — if they’re sophisticated enough that they can mimic people, and we can’t tell the difference — what do you do, let’s say in a customer service situation? Or let’s say on a 911 call? It seems to me there’s a fundamental issue about how, and when you bring up, what makes us human. If we’re going to be in this very strange situation where relating to something that might or might not be a human, what kinds of decisions do we make about what to do? Especially if, let’s say, we can get kind of a denial of service attack or something, with all these things that actually look and sound like humans. But what do we do?

Jeff: That’s a good question. It’s one of those that I’ve — I try to spend more time not thinking about it, because, almost every road that my thoughts go down ends up in some sort of disastrous, crazy end-of-the-world scenario. But, I think it’s a question that all of us in the tech industry have to ponder at least somewhat, because unless you have your head in the sand, you see there are advancements everywhere — like the robotics, like the Big Dog, I forget what lab it’s from–?

Len: Oh -

Jeff: Boston Dynamics -

Len: Yeah.

Jeff: Their research, these robots are being built that can do things and manoeuver in ways that humans even can’t. And you mix that up with the networking and the brains that are being developed for other projects, and you start thinking, “It’s not going to be too hard to think of” — even if it’s not full AI, what happens when somebody can change the algorithm that detects human versus something else that needs to be cleaned up, or things like that?

For me, I’m always just trying to think, “Is what I’m doing right now going to advance the human race?” I don’t have that grandiose of a vision of this book, but you know in general, when I work on things I just, you always have to pull back and make sure, “Is this something that matches up with what I want the future to be — for me, for my family, for everyone?”

Len: There’s just one last question on this topic. Generally I know that there is a discourse amongst — sort of religious intellectuals, especially in the United States, about how the predominance of screens and the internet is leading to a form of isolation, where people are not interacting with each other as it were in meat space as much as they used to be. There’s a sense that there’s something about relating to these objects, even if there are people on the other end of them, that somehow distances us from other people. Do you think that that’s something that is happening and that it’s something we need to be aware of? Because all of this technology and this way of relating to things is actually quite new.

Jeff: Yeah, I mean, it definitely is a thing that is happening, and the question is, to me is — is it something that’s bad, or is it an opportunity? And I’m a Catholic, I’ll disclose that here, but both Pope John Paul II, Pope Benedict and even Pope Francis — who are the three most recent leaders of the Catholic Church, they’ve all seen this revolution coming. And it’s interesting because each one had a different perspective on it. But all three first identified it as being something good, and then they identified it as something that presents an opportunity for every individual to kind of — you can use that to increase your relationship with someone, or you can use it to close out your relationships with other people. It’s really up to the individual.

Also, as developers, we make these tools that can bring people together, but I think all three of them also emphasize the point that real human interaction is important. And that’s why all three of them also said — when you have things like the sacraments and the church, where you kind of come together as the church community, it’s important to have a physical presence, because there’s something different between a physical presence with someone else — and something through a screen or through an email or through text, that kind of thing. There’s something metaphysical there that just doesn’t happen. And it brings up some interesting discussion, because it’s definitely, everybody sees it happening today. People sitting at the same dinner table, all of them on the phone, none of them talking to each other. Or even worse, everyone on the phone talking to each other at the dinner table, with their phones instead of their voices. It’s a strange thing, and I think as a society we’re dealing with it and figuring out our best step forward at this point.

Len: When you speak about opportunity and looking at it with some optimism, is there anything in particular that you think of when you think about the opportunities for new forms of interaction that we’re being presented with now, or that we can develop in the near term?

Jeff: Yeah, one thing especially that I see is people who have any kind of mental delays, or physical problems that can interfere with their ability to interact, some of these people already are seeing the benefit of — they have these tools now that are so much cheaper than they used to be, or that they didn’t even exist ten or twenty years ago, to allow them to communicate and be more personable with people. And it’s the same thing with science and drugs and medicine. Coming up with new ways that people can combat situations and illnesses that have made them unable to be part of, like, the human experience, and have real deep human interaction. And again, there’s always two sides to it. If you take somebody’s — what we call an illness, that might also enable them to be somebody different, who they really need to be. You need to make sure that you’re not just muting that part of them; that you’re expanding their abilities to interact and be part of the human race.

Len: That reminds me of one time, actually relatively recently I was watching a research documentary about the history of the computer, that I think had been made in the early 90’s. And it included just an amazing scene that almost brought me to tears actually, in which a bunch of young people are in front of computers — and they’re actually able to read books on their own for the first time, because they were unable to hold them and turn pages. So reading with a paper book was basically an impossibility for them. But now, when they can read on the screen and they could turn pages with voice commands — all of a sudden this whole world of human thought was open to them for the first time in an independent way. And yeah, that just struck me as a really powerful way of sort of using technology to increase people’s ability to interact, not only with other people but with, you know, people from the past and our history.

Jeff: Yeah and a lot of it seems to come from the ability and the desire for — a lot of people have a lot of disdain for the idea of capitalism. But in a sense, some of the reasons why some of these technologies have been opening up people’s lives, especially people who might not be as well off as the top 1%, or whatever, is that since these devices are now in the hands of all consumers and are becoming cheaper and cheaper — devices that have what would be considered supercomputing abilities years ago — we can now write software and put the software in the hands of everybody to do these things and to enable new interactions. Especially for people who — like I said, ten or twenty years ago, they would have to spend $50,000 or more to have a device to let them communicate. Now that’s on your tablet in your pocket. Where you can mount the tablet to a new kind of mobile platform that you can use, if you have a physical debility — to get in places you never could before.

On one side, we see people sitting with their phones and blocking themselves out from other people. And on the other side, we see people with these phones that are opening up a whole new world. And being able to converse online and converse in person even, with other people.

Len: Just circling back to your book. You mentioned that it started out as a blog, and eventually you decided to turn it into a book. Was there any sort of particular moment? Like did someone ask you, “This would be great if you wrote a book about this” or “Can you expand on this because it’s really helping me do my work”?

Jeff: It was for me mostly a personal goal. I actually have been blogging regularly since about 2004. And probably around 2008 or 2009, I started thinking, “I really want to write a book someday. I have no idea what I want to write it about. But I want it to be something I’m passionate about and I’m interested in, and I feel like I could contribute something to people”. And for a long time I’ve done a lot of technical documentation. Like I said, I used to work with the local radio stations. One of the projects I always would get texts with is, “We have this new phone system, or we have this new whatever. Go sit down, sit through all the tutorials, and then write up a guide that will help people to quickly get up to speed on it”. Because the manuals you get with the manufacturer are pretty much junk.

So I’d spent a lot of time doing technical writing and documentation. And I’ve spent a lot of time blogging. Early on, it was more about personal things and religion, and philosophy, and those kind of things. But I started writing a little more about computing, and when I got into Drupal, I started doing more tutorial type posts on, “Here’s how you do this and that”, documenting how to do certain things in Drupal that I didn’t find were well documented. But there’s one platform for documentation and then there’s another platform for guiding people, really. And I’ve seen books are usually the best way, at least for me, but also for a lot of other developers to get into a technology.

So I think after I wrote four or five, six hundred to a thousand word blog posts on Ansible, I realized, “I can keep doing these short posts that kind of are disjointed, and don’t really guide somebody through things. But they give somebody a few nuggets to kind of get started with.” And instead of continuing to do that, I wanted to write more of a comprehensive guide of, “You’re a developer who might know a little bit of shell scripting and how to turn a web server on in a cloud environment”. I wanted to take that person and bring them through to, “Hey, it’s not that hard to build an entire infrastructure with fifty servers in Amazon or in Digital Ocean or something like that”.

So I figured the best way to that was a book and so I took those blog posts, kind of wrote a little introductory part to it, reorganized them. And that was the first published edition. Which at that moment, I said it was 30% of the book’s completion, when I first published it on Leanpub. Now, little did I know that after those, like, 30 pages were published, I’d be writing another 370.

Len: How did you find out about Leanpub and why did you decide to use us?

Jeff: I’ve been on Hacker News for a while, and I kept seeing every now and then somebody would post, “Hey, I have this new book and it’s on Leanpub” and I’m like, “That’s interesting, I’ve never heard of this Leanpub thing”. And a couple of them, I think it was actually right around the time where Node.js was getting popular. The Node — I forget was it the Node Craftsman book or The Node, one of the node books on there–

Len: Yeah, Manuel Kiessling, I think?

Jeff: Yes. Yeah, it’s one of the more popular books on Leanpub. I bought it on Leanpub and I’m like, “This is amazing. Like I bought the book, it wasn’t finished, but it’s already really high quality — and I get the free updates forever, and there’s no DRM. Like where’s the catch?” So it was really from a consumer perspective.

As a technical person I hate the idea of DRM on a book. Because it’s like, when I go to a bookstore and buy a book, I don’t have like a locking mechanism that I have to unlock to read it. And so I liked that aspect, and I liked the whole user experience of buying the book and getting updates, the author could communicate with me. So it really attracted me to the platform.

But that was probably six months to a year before I decided to start writing. And when I started writing I looked into the options and it was basically — Leanpub was awesome and everything else was like ancient history. And nowadays there’s other, there’s like Gitbook and some other platforms you could write with, but Leanpub is still, from what I’ve seen, it’s still the most like actual publishing workflow friendly tool out there. And I’m glad I’ve been sticking with it because it’s — right now I’m working on actually publishing a printed copy of the book and the whole process was basically, scroll through those couple of things, click a button, and you have the PDF, and you’re done, instead of having to reflow things and lay it out manually and all that kind of junk.

Len: Yeah, thanks very much for that. Yeah, we worked hard with a few of our early adopter authors to build our print-ready output features. We have one for PDF, and we have one for InDesign as well, which is a little less mature. But yeah, that’s a really important part of us. What we’re doing is that we want people, when they’re done their book, to be able to hit a button and then have something ready to put up on CreateSpace which I think you’re using–

Jeff: Yeah.

Len: Or Lulu or something like that, so that then they can do print. I have a question about that. Why are you making a print book?

Jeff: One thing that I’ve realized, for this book in particular, it’s more of a guide for somebody who’s never seen Ansible before. They can pick it up and by the end of it, they’ll be hopefully able to make their entire infrastructure and do everything automated — so they’re never logging into servers again. For that kind of a guide, I’ve had a lot of emails from people asking, “So when can I get a printed copy?” because they want to sit there and mark it up, they want to dog ear some pages and have it front of them while they’re doing their work. And that workflow is what a lot of people like for more of a guidebook.

If it’s more of a novel, where it’s a story about the software you’re using, or if it is actually a novel, a lot of people are more prone, I think, to get the ebook version. But there are some developers who like to mark their book up and like to have that physical interface a book has, that so far no kind of e-reader can give.

The other reason also is — I got started in web development through design, and so I’m a little bit more designer-oriented than developer-oriented, even though I don’t do as much nowadays. So I really wanted to get a physical copy, just so I could have it in my hands. It’s kind of like a physical reminder.

I actually just got a proof. The proof had some problems in it so I’m going to get another one. But I got a proof, and it was so amazing to know that two years of time that went into this book is in my hands. It’s like a tangible — it goes back a little bit in our conversation — it’s like when you have it in your hands, it’s different than having a file. You can flip through the pages that you spend so much time working on.

And the coolest thing is, with the feedback that I can get through Leanpub, it only takes a couple of minutes to build a new copy. So I can keep working on that physical printed book almost at the same rate that I can work on the online editions and the ebooks through Leanpub and CreateSpace. It’s a lot better I think even now than it was a couple of years ago when I started the book. I looked at the workflow back then and it would have been a couple of days’ worth of work just to do each new iteration, but now it takes less than a day to take something you wrote in a Markdown file or LaTeX, and then get a proof to your doorstep. It’s pretty cool.

Len: So were you interacting a lot with readers while you were writing the book?

Jeff: Yes, that was the other thing that really got me into the book, and turned it from 30 pages to 400 pages. When I started the book, I put up a little page that said, “I’m starting writing on this, sign up here if you’re interested”. And I think I got like 40 or 50 people’s email addresses. And that was already getting close to my goal. My goal was to sell 200 copies of the book. If I did that I think I would have been pretty happy. It would’ve been fun — write like 80 or 90 pages, be finished with it. But I got 50 then, and then when I published the first version — which only had like 30 or 40 pages and only a couple of introductory chapters — all of a sudden there were a couple of hundred sales. And as I’ve already sold the quota that I set for myself, and I haven’t even started the book barely. So that really motivated me to start writing more and giving a little more structure to the book.

Once I did that, I started getting a lot of readers who were taking my examples and putting them into situations I never would have thought of, and being like, “This broke. This broke. This broke”. So really they were being my QA and reviewers early on. And I have — I even put a list of the people who helped make the book better in the afterword. If you get the print copy, it’ll be in there. And it’s actually up, I just updated it last night with that. But those people are awesome. They, a few of them even sent me 10 or 20 emails throughout the writing of the book. Every time I’d put out a new chapter, they would sit through and find like every little detail, when I would have a grammatical error or whatever. And it was awesome, because that made my editorial process be a lot more of optimization rather than fixing my stupid grammar.

Len: That’s fantastic. I noticed you also include in your book a changelog, which I imagine won’t be in the print version. In your change log, with each new version you document what the sort of major changes were that you made and improvements in editions. I was wondering if that was something that you think necessarily belongs in the book itself, or if Leanpub could provide a feature like that, would you prefer to have that over a changelog that you manually create in the book itself?

Jeff: I can go either way. The thing that I liked about having it in the book itself, was that somebody didn’t have to refer somewhere else when they wanted to see what was new. It was also a good kind of barometer for the readers to see what’s been going on — or has anything been going on? Because there were a couple of points where I had some dry periods, and I didn’t write anything for a month. And some readers would go on there and they’d see — if they didn’t have a changelog, they wouldn’t know like, “Is this just abandoned, or what?”

I don’t know what the statistics are on your end, but I know for myself, if I start a book and there’s not a lot of interest, I might not ever finish it. But somebody who buys the book might not know that. So I wanted to give them that idea of, “This book is still active, and I’m still working on it.” And there’s an indicator on the page that says the last time it’s been updated. But I also like giving a tangible guide for, “Here’s what’s been updated.”

So if it were integrated into Leanpub itself, that might be nice. I’d just want to make sure that there’s a way that that can be updated easily. Because it’s really just for the reader to see, “Here’s what’s changed. Here’s what you might be interested in,” and sometimes the readers — like I said — there’s dedicated readers — would even go in the changelog, and they’d be proofreading my changes for me. I don’t know what drives them to do that, but they’re awesome people.

Len: That’s amazing. I hadn’t quite thought of that use of a changelog within a book before, that even if there’s been a dry spell, it gives you a very confident sense that this book is being worked on, and how it’s being worked on. That the person’s being rigorous about it.

One idea we did have for what we call our “library”, which is where your Leanpub books are kept — which we now see either in the browser or our iOS app, which is still a baby — it’s an interesting idea, like “What does a library of in-progress books look like?” We were thinking one interesting feature might be if you click on a book and kind of opened it up, that you would see a timeline. So it might be a line, and then there might be little circles along the line, that would indicate when the author had made a major improvement, or something they thought was significant enough to surface in a timeline, and then click on it or something and see basically, yeah, what those changes were. So that was what we were thinking, something along those lines. But I take your point about how important it would be to actually have something like that in the book itself. So that’s something that we’ll definitely think about.

Jeff: Even as motivation to me as an author, I know one reason I thought about going with a traditional publisher was just the forced pressure of — you have this deadline, you have to get the first draft in by here and that kind of stuff. Without those forced deadlines, you aren’t really motivated sometimes to get through the hard parts. There are some chapters in the book where I’m writing it more because people need to know about it, not because I love it. And reading through the book, you might see a couple of those kind of sections.

But having some sort of metric to show me too, “Here’s where I think I’m at and here’s where I’m going.” That’s pretty helpful. So I’ve even built my own little bash script that will count the words I’ve written in a given period, and give me a nice little graph of it. So I can see, I run it every month or so, and I just make sure that I haven’t just dropped off completely in terms of how much I’m writing.

As a writer, you have to motivate yourself to write something. Even if it’s horrible and you’re going to throw it all out, if you’re writing something your brain keeps going, and you keep moving forward instead of just stopping. And I know that from writing all the big blog posts I’ve done, and all the articles I’ve written. Even if it’s only a 500 word article, you can get stuck and then you just give it up. But if you just keep writing — I think there was a blog post I did a couple of weeks ago. I wrote it, four times I think in full, and deleted it each time saying, “This is just terrible.” And then finally it’s like, “Ah, there were go. I have it. I feel that that’s the right way.”

Len: Well, it’s a fantastic post and obviously we noticed it right away. And yeah, when it comes to motivation, I was just describing it to some of our developers, the post — and one of the things I really liked about it was — you talk about how successful your book has been, but you also talk about how much work went into it. I think you said something like 1000 hours? Is that right?

Jeff: Yeah, it’s close to that. Early on I was very strict about tracking the hours. Now I’ve kind of dropped off, and now it’s like one evening I’ll spend four hours on it, and then the next night I spend zero time. So there’s still five to ten hours a week at this point, as I’m trying to get the book on Amazon and iBooks and paperback and all that kind of stuff.

Len: Did those hours, did you include — I forget, perhaps you did mention this in the post — did that include time you spent marketing the book? Did you spend any time marketing the book?

Jeff: Yes, and I’ve actually spent pretty minimal time so far marketing the book. That’s one thing where some developers can — I’ll call it “pimp themselves out”. And I can’t do that as much. It’s hard for me to do it, but I force myself to do it. Because if you don’t do it all, you’re not going to get any sales. So I’ve probably spent maybe 2–3% of my time on the book doing marketing — either talking with a couple of different organizations, trying to get them to promote the book internally, and working with a few groups that are doing similar, doing related Ansible projects, trying to get them to cross-promote the book a little bit.

But the main thing for me in terms of marketing has just been having a Twitter feed just for the book. And I had that early on, I think the first week I was writing the book, I created it. And that’s up to at least 500 or 600 followers. And that’s been a good platform for getting people to retweet information about the book and getting reposts.

But really the number thing has been that email list. Early on I had 50 or so people subscribed, and as the book has gone on and more people buy it on Leanpub — every time I do anything through email, I get an immediate small spike in sales. It might be one to two sales per day for a while, and then I’ll do an email and it goes up to five to ten for a few days. And so it seems like, at least for things like books, reminding people, “Hey, this book is out here, and here’s a coupon for half off for your friends,” or something like that — those are the things that get the most in terms of tangible spikes in sales.

The other thing is getting closer to finish. I think a lot of people are less comfortable buying before the book was 70 or 80% finished. But once I reached some point, the sales started going up a little bit, to the point where it was more steady over time. There weren’t little spikes here and there from the emails. It was more steady with a small spike.

Len: Okay. I’m curious — if you were approached by a conventional publisher who wanted to buy the rights for Ansible for DevOps — is that something you would consider doing?

Jeff: It’s something that I would consider depending on what they were talking about. One of the best advantages of having the book through Leanpub — and especially Leanpub, because it’s really the only platform that’s like this — is, Leanpub is more about getting the book to the readers, and it’s not about making money for Leanpub. At least that’s — every organization that sells something, you have to make some money. And I get that, but Leanpub has the least restrictive licensing terms, the least restrictive operating agreement with authors, out of any place that I’ve seen short of just posting the book on Github. And so that’s really what attracted me there, and the fees are nominal basically, and the capability to take your book and go to CreateSpace without having to do any extra work. That alone is worth whatever fees that I’m paying, which is again pretty nominal for this book.

Len: Just to be clear to anyone listening. We pay a royalty rate of 90% minus 50 cents for every sale, so it’s not exactly — we don’t, what I’m trying to make clear is that we don’t actually charge authors fees. What we do is we take a cut of sales, if they get any.

Jeff: Yeah.

Len: And so that’s 10% minus 50 cents per sale. [Editor’s note: the fee is 10% plus 50 cents per paid sale, which includes the transaction fees we are charged from PayPal. There is no fee for free sales.]

Jeff: Another cool thing about the platform is Leanpub seems to have a little more uptake with a technical audience. And so a lot of books on Leanpub are good opportunities to work together with other authors. I have a few bundles that my book is in, and those bundles have also helped to produce extra sales. So it’s cool, because sometimes a bundle — like somebody with another book in the bundle will promote their book, and then I’ll get a spike in my sales because people are, “Oh, that’s a cool book,” but I don’t care about this one. So it’s neat to have that. It’s kind of an informal community of authors at Leanpub.

Len: Oh, that’s awesome. I’m so glad to hear that. Yeah, our work on community is something we really want to focus on in the future. And yeah, we’ve got a feature where if you’re a Leanpub author you can reach out to another Leanpub author and ask them to make a bundle with your book, and they can accept or decline. That way you can have people make connections with each other — especially if they share interests or specialties, and then they can kind of cross-promote. And it can be a really fun thing to do.

I guess my last question for you is is there anything that stands out with your experience with Leanpub that we could improve? Is there anything that isn’t there that you think should be there? Or something that’s there that hasn’t been done the right way? Or another way of putting it is — If you could have your ideal “Leanpub give me this one thing”, what would that be?

Jeff: I think one thing that would make things a little simpler for me is the ability to more quickly preview changes, especially to the full book. If there was some local tool that could produce a PDF copy, so I could have a continuous integration process for my own book, that would be cool. I wouldn’t expect that, because that’s the bread and butter pipeline that probably takes the most work to maintain on Leanpub.

But that being said, the other thing that’s been tough — and the reason why I asked for that is, sometimes, when I’m writing in Markdown, which is my preferred format for writing, sometimes the way that the syntax behaves can be a little surprising. And even when I look at the Leanpub manual, there are some times where the examples don’t exactly match up to what I’m trying to do. Right now I think the Leanpub manual has a formatting that’s a little funky if you look at it in the front end, in the browser. So things like, when you put code blocks in lists with a certain language, with text surrounding it, with multiple indentations in that list — then those kind of things are where you find these little edge cases. And it just takes my brain a little bit out of the writing when I’m trying to work on the syntax stuff. But the ability to quickly preview those things, and maybe even have like a tool that can show you, “Here’s how it’s gonna look.” You can like paste a sample in, that kind of thing would be cool. But the thing that I–

Len: Oh, that’s very interesting. Yeah, thanks for that suggestion.

Jeff: Yeah, the thing that I like most about — in the time that I’ve been writing this, Leanpub’s interface has improved, at least three times, in a major way, to the point where it looks so fresh, the author dashboard is a lot easier to navigate, and has more useful information on the front of it now. When I started it, a lot of times I would go in there to change something, and I would sit there and I would click through every single menu, because there was no organization to it. So now it’s a lot easier to find, “Oh, it’s under ‘Writing’, and then this section.” So that’s been really helpful.

Len: Thanks very much for that. That’s really great feedback. The explanation for — there’s sort of a two-part explanation to why Leanpub was that way, and why it’s been improving. The first explanation is we’re a bootstrapped startup, so Leanpub is our baby that we would love to be spending 110% of our time on, and hopefully we will be in the future. The second explanation is we really believe very deeply in customer development. And so often, adding a feature and making it work for an author who’s desperately in need of it, is more important than necessarily optimizing the design or organization of things. But as we’ve grown more mature, we’ve been able to then focus. When things become more stable, we’ve been able to focus more on organizing things properly, and presenting them well. So I’m glad to hear that the work we’ve been doing on that has actually been an improvement. Because when you’re sitting on this side of things, you’re often just — you just hate yourself because things aren’t perfect. So it’s nice to hear that.

Jeff: Yeah, it’s been much better.

Len: Okay, I just wanted to say, thank you very much for being on the Leanpub Publishing podcast. This was a great discussion, and thanks for being a Leanpub author.

Jeff: Yeah, thank you very much.

This interview has been edited for conciseness and clarity.

– Posted by Len Epp

Originally published at leanpub.com.