Personal study notes, mostly written while reading textbooks, papers and source code. Each section is a folder of short markdown files — one topic per file, kept small enough to re-read in a sitting.
The point is to understand, not to look exhaustive. Notes that go stale get rewritten rather than padded.
| Section | What's in it |
|---|---|
os/ |
Process model, memory, scheduling, syscalls, Linux internals |
network/ |
TCP/IP, congestion control, HTTP, TLS, modern QUIC/BBR |
algorithms/ |
Classic algorithms, data structures, complexity |
rust/ |
Ownership, lifetimes, async, unsafe, ecosystem deep-dives |
harmonyos/ |
ArkTS, ArkUI, Stage model, performance notes |
distributed/ |
Consensus, replication, CAP, time |
database/ |
Storage engines, query planning, PostgreSQL internals |
security/ |
Auth, crypto, common pitfalls, threat modeling |
- One file per topic. Filename =
nn-kebab-case.mdso they sort. - Each file opens with a one-line summary and a "why I'm reading this" line.
- Code snippets are runnable when reasonable; otherwise marked
pseudo. - Sources are cited at the bottom under
## References.
Seeded. See INDEX.md for the live table of contents as files
land.
CC-BY-4.0 for the prose; code snippets are MIT unless otherwise noted.