The pet projects of Dr Frank Einstein
summary
An extensible on-demand language called XL language aims to reduce syntactic noise but undergoes repeated rewrites and sits at ~4,000 commits with about 10 authors and only ~267 non-owner commits, illustrating a “failing forward” research project. A 3D slide tool Tao3D accumulates ~9,300 commits across 38 submodules yet has effectively one active user and suffers LLVM bitrot. A build tool make-it-quick (≈1,500 lines) yields up to 16× faster parallel builds versus autoconf but gains little adoption. A C/C++ flight recorder (~360 commits) helps debug Heisenbugs but has low traction. Successful contrasts: git-publish (246 commits, 24 authors) and qboot (84 commits, 13 authors, 624 stars) show simple problems plus an existing community matter.
AI-written summary · about
chapters (15)
- 0:00Start
- 2:55Concept Programming
- 3:53Pseudometrics
- 4:03Syntactic Noise
- 4:50Bandwidth
- 7:55Lisp
- 11:05How Did Excel Fail
- 14:19Conflicts due to Plugins with the Base Semantics
- 15:02Llvm
- 18:17Tau 3d
- 19:15Idea for Tao3d
- 28:42Recorder Library
- 28:54Why Is It So Hard To Know Why a Program Crashed
- 30:27Flight Recorder
- 31:20Git Publish
related talks
16mMaintaining an OSS Project (Babel)
57mTypelevel in 2016
32mScaling Open Source Projects from 0 - 1000 Commits
31mScaling Open Source Communities
13mKeynote: Giants, Standing on the Shoulders Of
42mFast & Furious: From Zero to Open Source Community in 9 Weeks
58mResponsibly Abandoning Open Source Projects
41mThe Open Source Effect