Ask HN: How much code does a CS PhD write?
I'm considering going to graduate school to do research in functional programming. It's something that I want to learn at a very deep level, not just practically but theoretically as well. My concern is that as a graduate student I won't write much code. Is this true? If you've done a PhD, how confident did you feel in your overall programming abilities compared to before? I know a lot of theoretical CS is math and formal proofs. How much is implementation, especially in the area of FP/Haskell/type theory?
Comments
Almost all CS PhD's I know, (mostly all professors), are pretty bad at actual programming, and very skilled in theoretical aspects of computer science.
Second year student here. I consider a Phd to be "research apprenticeship". You should expect to experience the full life-cycle of research. "Writing code" is only one part of it.
In my case, "writing code" accounts for ~25% of my time. Much more time is spent running experiments, analysing results and publishing.