A key problem for developing Tools for thought is that Mental states of deep software development and deep research thinking are mutually exclusive, and Switching costs are high between software development and research thinking. A natural response is to hire help—designers, engineers, project managers, etc. But a challenge I’ve noticed here is that because Insight through making prefers bricolage to big design up front, it’s very difficult to externalize a long runway of legible work which can be accomplished by a typical contractor or intern. Most of the time I can only see a short distance ahead of myself. I’m sure I could get better at looking around corners, but I suspect there’s some immutable component of this too.
And so, rather than a “full-time” intern or contractor joining me for some months at a time, it’d probably be better to work with such folks on a hourly or weekly basis, whenever I have something ready which needs help. Of course, this comes with significant switching cost!
All the way at the other end of the spectrum, the alternative is to bring on a creative partner, with the necessary skill and context to autonomously identify and solve problems. But that’s awfully expensive, even if I could recruit such people!
There’s a substantial chasm between these two points on the employment spectrum: hourly contractors, and deep full-time partners. I’d probably make a lot more progress if I could figure out how to make good use of a collaborator for months at a time.