Notes from RStudio::Conf 2020

February 19, 2020

I enjoyed attending the RStudio Conference in San Francisco January 29th and 30th. I also attended the Tidyverse developer day on the Friday after the conference and did my first Pull Request (PR). Soon I will be able to type ?forcats::fct_reorder and see my own super-simple example added to the help for that function.

The videos from the conference are now online. It’s impressive how much video content RStudio has online at their resources site including video from all the past conferences. There’s a lot of useful stuff there.

My favorite talk of the conference was the keynote by J. J. Allaire, the co-founder of RStudio. It was relevant for the world way beyond R and RStudio. Starting at about minute 25 he gets into issues about the motivation and incentives of for-profit corporations. He gets deeply into the issues. One does not expect to see a slide with the title “Dodge v. Ford Motors (1919)” at a computer conference. The subject that is addressing is relevant way beyond the computer industry. RStudio is a fascinating company and J.J.’s talk made it clear how much of the personality of the company comes directly from its founder.

The other Wednesday morning keynote was also excelent. Fernanda Viegas and Martin Wattenberg of Google delivered a very polished and interesting talk on Data, Visualization, and Designing with AI. It got into the topic of issue of what “fair” and “biased” mean in the context of machine learning based decision algorithms.

Some other talks I enjoyed:

A talk on the rayshader package had a lot of graphical “Wow!”

I enjoyed William Chase on The Glamour of Graphics.

There were several heavy duty sessions on sports analytics including this one on NFL route tracking. And see this one as well.

I always find talks by Jenny Bryan to be very rewarding. She’s my go-to source for miscellaneous practical tips on a number of R subjects I care about, such as her Happy Git book and What They Forgot to Teach You About R. The title of her talk was Object of type ‘closure’ is not subsettable, a message that every R coder encounters eventually. The talk addressed how to debug R code.

Because many sessions were presented in four parallel tracks, I want to go back and watch some of the sessions I missed such as Making a tidy dress by Dr. Amelia McNamara.

Of course there were lots of sessions and not all of them were great. But there was a lot of good content.

A nice figure
Posted on:
February 19, 2020
3 minute read, 428 words
See Also: