A lot of common assumptions about performance in software are built on superstition, not evidence. That's my hot take.
Seeing talks is only a small part of attending conferences. It sucks that ICFP this year seems to be actively disincentivizing talking to other researchers or even asking questions, since you have to leave the talks to do it.
Oh, I never plugged our ICFP paper here.
Efficient tree-traversals: reconciling parallelism and dense data representations
The PDF for the paper and the talk are both available on the ACM DL.
advertising a nosql database with cartoons seems unethical, like advertising cigarettes to children. what if some innocent software engineer sees it
postdoc at kent, recently phd student at indiana
make functional programs fast using types
socialism is cool
A Mastodon instance for programming language theorists and mathematicians. Or just anyone who wants to hang out.