Practical code archaeology
summary
Claim: about 70% of successful code archaeology comes from talking to people, not new tools. Use concrete techniques: git blame/TIG with a rainbow age color scheme to spot code eras, simple Unix tools (grep find sed), and safe sandboxes (Docker virtualenvs) to run risky artifacts. Noteable discoveries: missing product "B" remnants, a pervasive "bike" class from an old pivot, and that notebook metadata can reconstruct execution timelines — these practical hacks consistently work.
AI-written summary · about
chapters (16)
- 0:00Code archaeology intro
- 1:12Why code archaeology matters
- 4:05Frustrations in code archaeology
- 5:11Talking to people vs tools
- 7:19Code culture studies
- 8:23Legacy code stories
- 12:18Git blame and commit browsers
- 14:05Git blame by age
- 16:35CLI tools for archaeology
- 17:40Reading unversioned code
- 19:02Code reading skills
- 21:09Prioritizing code sections
- 21:31Pair programming detective work
- 22:59Safe legacy code runs
- 23:21Code archaeology toolkit
- 24:08Q&A
related talks
55mDigging into the Matrix: Practicing Code Archaeology
51mSoftware Archeology and The Code Of Doom
01h 02mDigging into the Matrix: Practicing Code Archaeology
33mHere's to history: programming through archaeology
23mHere,s to history: programming through archaeology
47mTreat Your Code as a Crime Scene
56mReading other peoples code
49mTreat Your Code as a Crime Scene