I’ve been working for years on “life garden” document, trying to come up with an at-a-glance view of where my life is working and where it isn’t. I’ve been trying to apply Christopher Alexander’s principles of living structure.
And I’m stuck.
My document has all kinds of structures and centers and boxes and arrows. But I don’t think it has enough of me in it. Or the categories I’m using don’t resonate. Or the relationships I’m building between notes is off. But it simply doesn’t have a good feeling which was one of the main things Alexander looked for in living structure.
Now I’m using Tinderbox with an MCP connection to Claude, so it would be very easy for me to fire up the LLM and try to use it to diagnose some of my problems. But I found myself hesitating on this.
Recalling that sociologist Niklas Luhmann interacted with his extensive collection of notes as if he were having a conversation, I decided to ask my notes to find an answer.
And the results were pretty interesting. I started with notes about Alexander’s use of feeling and introspection to identify “centers.” This lead me to Alexander’s thoughts on Self, which in turn lead me to a quote from Howard Thurman suggesting we are not single persons, but contain many persons, and the Self is a rallying point. From here I jumped to a note about the Internal Family Systems model which seeks to use the Self to mediate and heal the various parts of our persona.
Spending an evening in a bar alone with my notes turned out to be extremely worthwhile. Now I’m adding small notes for “parts of me” that I can identify – Curiosity Seeker, Smart Ass, Imposter, Care-Giver and the like. I don’t quite understand how this integrates with the “done-list” section of my life garden, but it has a better feel to it.
I think the answers that I got from my own notes are much more contextualized to my life and interests than anything I might have gotten back from Claude. Not that Claude is bad or AI is evil. It still doesn’t know me as well as I know myself, apparently.
Now what if I hadn’t taken such a bonanza if tiny notes about Self? I was actually expecting that would be the case. In that event, I planned to go straight to Christopher Alexander and have a conversation with him.
SEE ALSO
Hmmm…looks like I wrote a very similar post here: Why You Shouldn’t Ask ChatGPT about Information Architecture (or Anything Else You Want to Learn) in this case the LLM delivered precisely the wrong information. The models have improved considerably in three years. They are remarkably good at finding what you’re looking for. My notes, on the other hand, delivered something unexpected, topically unrelated but remarkably relevant.
Is Your Zettelkasten Alive? How Structure Impacts the Life of Your Notes
