5 ms·
There is some "academic" stuff you need to know to be a really good coder: algorithmic complexity, information theory, lambda calculus, and a good grounding in
by rjrodger 16y ago
There is some "academic" stuff you need to know to be a really good coder: algorithmic complexity, information theory, lambda calculus, and a good grounding in discrete and continuous mathematics.
You need to be able to eye-ball an algorithm and go: hmm, that's O(n^2) ... not gonna work for this use-case... etc.
You can book learn this stuff. Just make sure you do.
Some of the best coders I know never darkened the door of a university.
- notJim 16y agoWhen do discrete and continuous mathematics (in a formal sense) come into everyday development tasks? I don't even really know what those fields are, but I do fine as a developer. What am I missing out on by not knowing anything about them? Edit to elaborate on "everyday development tasks": some startups innovate by applying some novel algorithm (beyond well-understand and well-documented ones that anyone could implement), but many are just taking well-understood problems and solutions and doing them better. Better UI/UX, better business-model, more integrated with other platforms, etc.
- imechura 16y agoThis is highly debatable. I have been coding in the finance, airline and manufacturing industry since '99. Not once have any advanced (read above HS algebra) mathematical concepts been necessary. My experience has been that the advance math stuff is the edge case in software development. The bulk of it is crud, business logic, parsing and I/O.