Knowledge metabolism

It’s been interesting how my projects build on each other. I’m learning how I want to interact with AI to build rigorous processes. Recently I’ve been working on a knowledge base. The project is an outgrowth of my Lore Builder system. The idea is to draw cards to determine tasks that manage and maintain the knowledge base. The system draws a task and an element card (a verb and a noun), and acts based on the pair drawn.

• Mine + Timeline: Search through sources for information related to dates, history, or chronological trends.
• Refresh + Answer: Search for a stale old answer, and reverify. Is it still true? Are there any caveats to include?
• Bridge + Map: Connect two indexes that are currently unconnected.

I’m overcoming my instinct to collect knowledge and stack it like firewood. The deck enforces a metabolic process. Knowledge is ingested, processed, used to perform work, and at times, excreted. It sets limits by continuously linking ideas to projects, using novelty to manage projects..

• Ideas: Ideas are assessed on novelty. Is this something new? Yes, then save it. These are saved as questions and answers. Example: “How should knowledge particles be sized in a KMS?”
• Interests: Assed on improvement. Does an idea improve a feature I’m tracking? If so, link it to that feature. Example: “Note Atomicity.”
• Projects: Assessed on viability. Can the idea live in a project I’m working on? If so, link it to that project. Example: “Knowledge Management Deck.”

By doing this I keep the knowledge I gather focused on the actual projects I’m working on, and inversely, keeps me working on projects where I show genuine interest. I’ve still got a ways to go to develop the deck, but once done I should be able to let Claude work through the deck itself and shape the knowledge base to support my goals–while I sleep.