2 ms·
Read papers, Not too much, Mostly foundational ones
- ggm 6y agoDidn't reference Michael Pollan, but the title is a riff on Pollan. Reading foundational works goes to citations and citation chains and depth. The risk is missing subsequent revising theory development. That said, I used a 1978 tapesort method for some things because I had disaggregated serialised inputs I wanted to make into a single ordered stream, two different contexts across five or ten years and it was supremely satisfying to go into an early CS work and find pseudocode which did this for me. I bet there are good, recent examples in things like kibana or logstash, more appropriate to our time. I liked using a tapesort.