Software environmentalism
summary
Software grows ~10–20% year‑over‑year (doubling every ~5 years) and an estimate cites ~10k mainframes holding 200 billion LOC, so legacy systems persist and block recycling. Empirical samples show Chromium, IntelliJ, Elasticsearch, MySQL and Mercurial all doubled in years. Built a Pharo‑based object inspector with many small extension tabs (<10 lines each) and a graph‑layout clustering technique that splits a 3,000‑line theme (used across ~15,000 LOC) into cohesive parts, dramatically speeding assessment. The aim: produce recyclable systems.
AI-written summary · about
chapters (12)
- 0:00Software growth problem
- 5:41Plastic industry analogy
- 8:14Reading code costs time
- 12:53Introducing assessment
- 16:25Context matters in tools
- 18:12Emerging system structures
- 21:47Sustainable software development
- 26:44Conway’s law and tools
- 30:11Tool choice shapes behavior
- 33:06Rethinking inspectors
- 37:59Integrated development experience
- 40:51Software environmentalism
related talks
54mResolving technical debts in software architecture
35mSoftware Engineering - Development in 100 Years Time
01h 01mSoftware Design in the 21st Century
54mThe Future of the Past – Legacy Will Stay With Us
59mLiving with your Legacy
32mExemplars, Laggards, and Hoarders: A Data-driven Look at Open Source Software Supply Chains
23mCan Software Live for 50 Years or More?
52mChaos Development: Removing software as the inhibitor of Agile