A clear distillation of some cybernetic principles from Gordon Brander:
"If the error rate is too high, tighten the feedback loop.
If the variety is too low, loosen the feedback loop."
From other episodes
- The model and its harness together are like lichen.2026-04-13
- The time to reflect and write what you learned gives compounding momentum.2026-03-30
- Vertical AI is compressing the B2C-to-infrastructure arc into a single move.2026-03-23
- Find the smallest feedback loop that could work.2026-02-09
- An excellent post from Brendan McCord: A World Unobserved.2025-12-15
- Feedback loops create leverage.2025-10-27
- The inductive logic of folksonomies is not "from a standstill is this item good?"2025-10-13
- A key approach for agent learning: accumulating insights in a LEARNINGS.md.2025-10-13