How to write a short book in 20 hours

5 minute read

I’m selling a book about good habits that young scholars could foster while finding their path in academia, for only $ 5.

100% of the sales will go to a hospital fighting COVID-19.

There is a sample of the book, including the chapter that describes the first habit.


I wrote a short book in 20 hours.

Is this the best book I could write? No.

It this book really well-written? No.

Are you proud of this book? Definitely yes.

Some background

I am in the middle of a pandemic that is only getting worse. I have two small kids at home that require a lot of me. I only have the evenings free when they are already sleeping. I wrote for four evenings, about 2–4 hours every evening. I started writing on Thursday and finished on Sunday. Also, on Sunday I made the cover. On Monday morning, I started selling. I sell eight copies in the first day ($ 5 each).

Why I wrote this book?

Many reasons motivated me to write this book.

First of all, I have been thinking about writing a book for many years. I actually already started to write one book in the past, but I gave up not much later. For those that follow me on twitter, they may know that I am writing another book on open source licensing. But I thought I need another book to be ready sooner.

Second, because I am in the middle of a pandemic, and I am feeling incapable of doing anything useful. I have already canceled some research works because neither my students nor I can commit to them anymore (yes, I know I’m privileged). Many relatives, friends, and colleagues are suffering from this pandemic, and I can barely do anything to help them. I like my work, but I don’t see my work being anyhow useful to help them overcome this immediate situation. Nevertheless, I’m seeing many colleagues who are finding a way to help others, for instance, by creating tools that could increase the pandemic awareness. I eventually made some donations here and there, but, unfortunately, I cannot commit to these regular donations next month without compromising my family budget. Then I had the idea of writing a book and donating 100% of the sales to help COVID-19. My little experience pre-selling the other book opened my eyes that it’s possible to sell books on Twitter.

Third, having this plan in mind, the last point was to decide what I could write about (and very quickly). It did not take much to realize that I could write about one of the talks that I gave in the past. First because I believe the talk was well-received (both in-person and on-line). Second, because I was already thinking about putting the ideas of that talk in a blog post. However, I knew it would require me many hours to do it, so I was keeping postponing. By converting this blog post in a book motivated me to make it happen. And really fast.

How did I write it?

As I mentioned, I did not have much time to write this book. I decided to write the book on Thursday, April 23rd, and I was willing to compromise to make it available on Monday, 27th.

My first step was to find a book template. In my other book, I have already spent some good hours fighting with a template (and I still don’t like my current template). To write the other book, I’m using asciidoctor, a framework that exports documents to many writing formats. To save some hours, for this book, I decided to use LaTeX instead. I’m reasonably familiar with LaTeX, so I would not wast any time configuring it, running it, etc. I also knew there are some excellent LaTeX templates out there, so it was not hard to find a one I liked.

Since this book was about a talk I gave in the past, I have the structure already designed in my mind. I used the same structure of the talk to structure the book. On the first evening, I wrote the preface, and the first two chapters (around 2.5k words). The preface I rewrote in the end, but I did not make substantial changes in the first two chapters after my first pass. I kept the same progress over Friday and Saturday (about 2k words per day). The book was 90% on Saturday. On Sunday, I made an overall pass, trying to fixing some typos and strange wordings. Also, on Sunday, I made the cover (using, and I wrote a blog post to share the book on social networks. I already shared this blog post on hackernews and reddit on Sunday evening. On Monday morning, I wrote some tweets to release the book to the public. I made eight sales on the first day on Twitter and no sales through the other mediums.

More than 20 hours

Although the title of this blog post is correct, upon reflection, I spent many more hours working on this book.

First, because I probably spent 20 hours or so designing and creating the talk. When thinking about the talk, I took several notes, made some sketches, and also did some research. After that, I spent some good hours creating the talk on Keynote. Till minutes before the talk, I was making changes in the slides. I also spent 2–3 hours rehearsing the talk, which gave me additional ideas. When writing this book, I took for granted the time I spent thinking about and creating the talk.

Second, because I also took for granted what I learned in the past. For instance, I did not have to do much research on the topic (it was fresh on my mind), and I did not need to master the tools used (it took 10 minutes to have a book template and about 30 minutes to have a cover).

It would be a different story I had to write about a subject that I am not super comfortable with.

Finally, it’s also important to notice that, as a researcher, a good proportion of my job is devoted to writing. I have been writing a lot over the last 10 years lot, so I’m used to technical writing. I’m not a fluent writer (I am not fluent in English either), but writing is not a big issue to me.