💃

Day 99-100: Small Steps

Sometime back in late April I made a strange decision to try to learn a new R package every day for 100 days. The goal was transparently ridiculous, and even at the time I was aware there was no way I could do it. Of course “a package every day” was never the primary goal. What I’d really hoped to do is get into the habit of trying out new R packages, just for my own curiosity. It’s part of a broader goal I have for 2018, which has been to broaden my intellectual horizons a little across a range of areas. And honestly, I just wanted to start thinking of programming as a thing to enjoy again, rather than a chore I had to do to make it through the day at work.

Looking back, I’m pretty happy with the outcome. In the last 100 days I’ve written 35 blog posts, and probably covered a similar number of packages that were (mostly) new to me. The posts covered a pretty wide range of topics and packages: blogdown, emo, h2o, magrittr, leaflet, lubridate, meme, default, lsr, rtweet, stringr, stan, sparkline, gaussfacts, mathart, purrr, refmanager, magick, non standard evaluation, here, pryr, fcuk, plumber, shiny, scico, dalex, acss, ggstatsplot, paletter, kableExtra, rcade, boring bits, skimr, Rcpp, workflow, and keypress. That was nice 😀

Not surprisingly most of the posts (about 75% of them) were written in the first half of the project. That’s partly the inevitable consequence of the novelty wearing off, but it also there have been a few other things that have come up along the way…

The other things!

One big thing that interacted with this 100Days project in positive way is my teaching efforts. I’m teaching an undergrad psychology class on program in R, so one thing I’ve had to do recently is focus on updating my lecture notes:

One thing I was pleased about is that although I didn’t deliberately start out with this in mind, many of the posts I’ve written for this series ended up influencing how I’ve gone about revising the R for Psychological Science page. It is still very much a project in progress, but I’m a lot happier about it now than I was feeling about the original Learning Statistics with R project that it builds from. So that’s nice too!

The other main “interruption” to the post series came when I was asked to fill in as a speaker at the useR!2018 conference in Brisbane. The long gap between the skimr post and the Rcpp one was spent writing my talk:

I think it worked out okay! I’m one of those people who can’t stand listening to how my voice actually sounds1, so I’ve not managed to watch very much of the recording 😬

In any case, apart from feeling flattered that I was asked to participate in such a wonderful conference, one thing I realised is how much this 100Days project helped me have the confidence to do it. I don’t think of myself as a particularly strong coder or statistician, but as a consequence of writing these posts I’ve been more active on twitter than I have previously. Contributing to the #rstats twitter community has made a big difference to me - it’s such a massive contrast to every other programming forum I’ve encountered and I credit that community for being the main reason I’ve regained some of my interest in R. I’m glad that some of the posts I wrote as part of my 100Days project have been of interest to other people contributing to that hashtag, but I absolutely feel like I’ve gotten way more in return than I’ve given. Yay! 🎉

Thank you, twitter!

What next?

I’ll probaly take a break from this blog for a while. There are some jobs I need to take care of with tidying up my R lecture notes; there are a few papers I’ve been neglecting; and I’d also like to put a little more effort into the cleaning up the templates I’ve been using to run online behavioural experiments. Similarly, I have a few posts I want to write for my other blog, and I’ve been toying with the idea of running a version of my “intro to R” workships for Sydney R-Ladies, since we have a fair number of beginners in the group! All of this means I have more than enough jobs to attend to over the next couple of months. Besides, I suspect I might enjoy updating this blog at a slower pace from now on, and maybe expanding the focus a little.

All in all though, I think this has been a really valuable experience for me, and I really appreciate the encouraging and kind comments from people who have helped me out. I’m looking forward to revisiting this blog in the not-too-distant future 😀


  1. To me it has an “uncanny valley” feel - it sounds enough like how I hear my own voice that I can recognise that it really is me, but it’s just off and it makes my skin crawl

Avatar
Danielle Navarro
Associate Professor of Cognitive Science

Related