I recently listened to the Greater than Code episode with Felienne Hermans. A lot of interesting points are raised during the episode, but the one that really stuck with me was the question of how the way programming is perceived in society can drive people (in particular women) from it. Felienne says that the more people perceive skill as coming from innate ability, the more women are driven away from it.

This is of course related to the idea of fixed vs. growth mindset and the work of Carol Dweck. A person with fixed mindset believes that skills are innate, whereas a growth mindset postulates that people learn skills mainly through effort. I still haven’t read the research (or its criticism, for example Scott Alexander’s several long posts on the subject), but I want to offer my own view of it before I’m inevitably proven wrong by hard numbers.

I believe most people have a mixture of both mindsets, in a specific way: they believe in the growth mindset in general, but when it comes to concrete skills, fixed mindset sets in. The belief in growth mindset is exemplified by inspirational phrases and quotes such as “Genius is 10 percent inspiration and 90 percent transpiration”. These phrases come in particularly handy whenever people see others being a lot more skilled than they are at various tasks.

But where growth mindset serves as a coping mechanism, people think with a fixed mindset when choosing which skills to learn, or, more importantly, which skills not to learn. When I hear non-mathematically inclined people talking about their experience with mathematics, they most commonly say things like “I was never good at math” or “math wasn’t for me”. There’s never any mention of the fact that they at some point made an effort to learn the subject, and that at a later point they chose to stop making that effort.

Mathematics might be an extreme example of fixed mindset domination, but it really shows up whenever people decide to stop learning a skill. Math is just something everyone who went to school had an experience with, with many finding it unpalatable. The fact that the fixed mindset effectively excludes certain skills from people’s radar should be of more concern, especially if you’re (as I am) less interested in measuring how well people learn and more interested in how and why people choose different things to learn.

This ties us back to the discussion on Greater than Code. Regardless of how the effectiveness of teaching is influenced by the two mindsets, it seems clear that in order to teach people technical skills like programming, the growth mindset must be encouraged. It’s embarrassing that it’s never been this easy to start programming, and yet the programmer class is still very much a closed, homogeneous club. As Felienne notes, the way forward is removing the barriers that act as gatekeepers but aren’t really fundamental to programming. She talks a bit about the more algorithmic, CS-y aspects of it, which I agree are massively overrated; I’d also point to the emphasis on tools, which generally get a lot more attention than they deserve.

It’s no coincidence that these two aspects relate back to two other male-dominated fields, namely math and (real) engineering. Fortunately programming allows for a lot more variety than that, and I’m always excited to see what programs people come from neither of these backgrounds are writing. I particularly like the parallels Felienne draws between programming and writing (beyond the obvious ones); certainly we want everyone to be able to write, not just the ones who come from literary backgrounds.