When my co-author Jacqueline Nolis and I wrote our book, Build a Career in Data Science, we wanted to provide a comprehensive resource of the non-technical skills and knowledge needed to get into and succeed in data science. I’ve been really pleased with the positive feedback we’ve gotten from the people it and our free companion podcast has helped. But while I think most people will find something new and helpful in it, one book can never be the only word on a topic as broad as career advice.

[This is my final post on publishing a technical book. If you’re wondering if you should write one, how to find a publisher, or what the writing process is like, start from the beginning.] It’s really cool to have a physical copy of something you wrote! I still sometimes just flip through the pages, marveling at how professional it looks and how much content we created. The final book ended up being 350 pages, so although it’s no mammoth technical reference, you can certainly spot it on your bookshelf.

[This is my third post on publishing a technical book. Check out the previous posts on why you should (and shouldn’t) write a book and how to find a publisher.] Finding the time Overall I’d estimate writing the book, including the proposal process, took us each around 400 hours over a year and a half. This doesn’t count the work promoting the book after it was out, which I’ll talk about in part 4.

[This is my second post on publishing a technical book. If you’re not sure if you should write a book, check out part 1.] You’ve probably heard of some book publishers before like Simon and Schuster or Random House. Those companies don’t usually get involved in smaller technical books. Instead there are several specialized publishers for books in the tech field; the biggest are O’Reilly, Manning, Pearson, and Chapman and Hall, along with some specialized ones like A Book Apart (which publishes short books).

Following the sage wisdom of brother Dave to write a blog post when you’ve given the same advice three times, after I had a few people reach out to me with questions about writing a technical book, I’ve put together my thoughts in this four-part series a year after publishing Build a Career in Data Science with Jacqueline Nolis. In April 2018, Jacqueline reached out to me about whether I would be interested in writing a book.

I recently started playing Pokémon again - “Pokémon Let’s Go Eevee” on the Nintendo Switch to be specific. In the classic Pokémon games, you have a team of 6 Pokémon that you use to battle against other trainers. In battles, type match-ups are very important, as some types of moves are “super effective” against other types. For example, fire moves are super effective against grass Pokémon, which means they do double the damage they normally would.

In this post, I try something new and share an analysis I did without stopping to explain the code along the way (with a few exceptions). I analyze a dataset on Bob Ross paintings from last week’s Tidytuesday, an initiative by the R for Data Science online learning community. Each Monday, a new dataset is posted on GitHub with a short description. You can see some analyses and visualizations people have done by searching for the #tidytuesday hashtag on Twitter.

Have you ever had a “first this then that” question? For example, maybe you’re an e-commerce business and you want all the times people clicked on an item and then added it to their cart within 2 days, or the last page they visited before registering. Or you work with pharmaceutical data and need to know what drugs people took before drug x and which drugs they took afterward and when.

I recently completed Colin Fay’s excellent DataCamp course, Intermediate Functional Programming with purrr (full disclosure: I work at DataCamp, but part of why I joined was that I was a big fan of the short, interactive course format). Although I’ve used the purrr package before, there were a lot of functions in this course that were new to me. I wrote this post to hopefully demystify purrr a bit for those who find it overwhelming and illustrate some of its lesser known functions.

In early 2018, I gave a few conference talks on “The Lesser Known Stars of the Tidyverse.” I focused on some packages and functions that aren’t as well known as the core parts of ggplot2 and dplyr but are very helpful in exploratory analysis. I walked through an example analysis of Kaggle’s 2017 State of Data Science and Machine Learning Survey to show how I would use these functions in an exploratory analysis.