“The temptation to form premature theories upon insufficient data is the bane of our profession.” – Sherlock Holmes
“Whenever a theory appears to you as the only possible one, take this as a sign that you have neither understood the theory nor the problem which it was intended to solve.” – Karl Popper
The first idea gets in and then the mind shuts. Keep asking questions.
With LLMs, you can delegate problem-solving phase (coding). It’s about how to (re)define the problem clearer and faster, and also how to connect problems, and problem-solving agents. Only then comes the fun part—redefining what’s solvable.
Related:
- 1-1c4b You always have to solve problems, including the problem of what problems to solve
- 3-1c3e However, you will always be solving the problem of what problems to solve because of your unconscious. Don’t be a reductionist in approaching problems themselves.
- 5-1b5 Keep solving problems including the problem of what problems to solve
- On problem-defining phaserevisit
- 2-1a1a6 ‘Tendency to overgeneralize from small samples’
- 2-1a5 ‘Sampling’ - Increase your sample size with QUALITY data
- 4-1a4b2a4 Prioritize by weighing the value of additional information against the cost of not deciding
- 5-1b1a8d Knowledge doesn’t automatically make us wise—the most learned are not the wisest (or the richest)
- On problem-solving phaserevisit