Greg Wilson

Comes a day, last kiss
Comes a day, last breath
Comes a day, and then another

Davey Alba’s article for Buzzfeed about DataCamp’s mishandling of an incident of sexual harassment has just been published. I think it’s a well-written summary of what has happened to date, and I am very grateful to her and to everyone who spoke to her for laying out the facts so clearly.

Yesterday, R-Ladies posted this great list of practical steps that people can take to build an R community where they work. I’m going to quote it a lot, but first I want to respond to this tweet:

My ten rules for teaching are too high-level to be immediately useful in the classroom, so here are ten others that I hope will help people in the moment as they deliver lessons on programming:

Mark Guzdial wrote a short series of tweets yesterday on how to reduce long lines at Computer Science office hours, which he has expanded in this post:

The biggest problem in data science is its bias against people who aren’t straight, affluent, white or Asian males. Its second biggest problem, in my opinion, is that nobody talks about how they sex chickens. After a long period of trial and error, people can learn how to tell at a glance whether a newborn chick is male or female. The problem is, most of them don’t know how they know: they’ve trained the neural network between their ears, but they don’t have conscious access to their decision criteria.

Mozilla has just earmarked $25,000 of funding for a research project aimed at getting R running in browsers under WebAssembly. The project is RQ5 in this list, and is to be conducted by someone at an academic institution or research focused non-profit. The deadline for applications is Friday, May 31, and they would like to spread the word as widely as possible—if you’re interested or know someone who might be, please pass it along. For more information, you can reach out to bcolloran at mozilla dot com via email or via https://gitter.im/iodide-project/iodide.

I just converted my introduction to R for Python programmers from Jekyll to bookdown, partly because I wanted to learn how bookdown works but also because I think that computational notebooks are one of the futures of programming. (We’re a pretty big field, so I think we’re going to have more than one.)

If you do a degree in English literature, you spend most of your time reading and critiquing other people’s work rather than writing new material yourself. I don’t know what the balance of reading and doings proofs is in Mathematics—I suspect it’s somewhere near 50/50—but if you do a degree in Computer Science, you spend far more time writing software than you do reading it critically.

There has been a lot of discussion online about DataCamp’s mishandling of a sexual assault case since Kara Woo bravely came forward on April 5. I published and updated some thoughts that same day, and summarized my final exchange with the company on April 15. Reading what’s been written since, I would like to clarify that I have not claimed I was fired for my concerns over the company’s mishandling of the assault on Kara. I was quite vocal in my unhappiness regarding that, but as I said on April 5, DataCamp’s stated reason for firing me was poor performance, and I accept that I accomplished less in late 2017 and early 2018 than usual.