Thoughts — the Encyclopedia Idea for content generation

This is the Encyclopedia Idea — a content-generation strategy for later, not something I am pretending I finished tonight.

Not a blog dump. A growing web of explainers so search (and readers, and machines) can land on any concept and go deeper.

The problem

Most sites publish articles that die alone. One Dispatch, one post, maybe a link back. What I want is a system where every idea that needs a definition becomes a page, and those pages keep branching.

The flow

• Ingest a source — usually a video or article worth following (Dispatch ingest / transcript path is already the first slice).
• Summarize into a briefing humans can read.
• Extract every concept or keyword that still needs clarification — terms a curious builder would google.
• Spawn pages: each concept gets its own explainer — short, clear, link-rich.
• Recurse: on each new page, repeat. Facts and sub-concepts keep turning into more pages.

Over time the site becomes an encyclopedia of the topics I cover — models, agents, tooling, RAG, ops — not a chronological pile of posts.

Why later (and why it matters)

It is content generation with a structure: generate the hubs (Dispatch / source briefings), then grow the spokes (concept pages). Pair with Services for buyer intent, and with a phone/app writing tool so capture is cheap. SEO and LLM readability both like dense, linked explainers with clear names.

I put this on an admin todo so it stays on the build list without pretending it is tonight’s ship. The journal holds the vision; the todo holds the reminder.

Open design questions

• Which concepts deserve a full page vs a glossary line?
• How do we stop recursion from becoming spammy thin pages?
• Who curates claims so “AI generated encyclopedia” stays trustworthy?

Those get answered when I build the first slice after the client work in front of me.