This is a great paper that says it's about code complexity metrics but is really about the superstitions that can abound in a young, poorly-understood field: https://www.tu-chemnitz.de/informatik/ST/publications/papers/ICSE21.pdf
This is a great paper that says it's about code complexity metrics but is really about the superstitions that can abound in a young, poorly-understood field: https://www.tu-chemnitz.de/informatik/ST/publications/papers/ICSE21.pdf 5 comments
@mhoye me, I can’t wait for the ci system that hooks humans up to mri machines and fails your PR if their brains light up too much @antifuchs @mhoye I think a good CI system would wait about 25 years, interview a bunch of people about the legacy of your software, and then declare whether the code was good or not based on those responses |
We've built elaborate tooling around code complexity metrics, to try and rein in the difficulties of managing complex codebases, but do those metrics actually correlate with programmer's brains working harder? It's a small sample in this paper, but so far the answer is: nope. Not at all.