Node-Based Thinking

In 1956, George Miller published 'The Magical Number Seven.' Working memory has limits. You can hold roughly four to seven chunks of information at once. The key word is chunks. The size of a chunk depends on how you structure it.

Chunking and Comprehension

A paragraph is one big chunk. Your brain has to decompose it into ideas, hold them in memory, and reassemble the meaning. A node is a pre-chunked idea. It's already the right size for your working memory. No decomposition needed.

This is why bullet points work better than prose for complex information. It's why tables beat paragraphs for comparisons. And it's why trees beat chat logs for decisions. The structure does the cognitive work for you.

A node isn't a paragraph with a border. It's a thought at working-memory scale.

Nodes as Handles

When information is a node, you can manipulate it. Click it, expand it, compare it side by side with another node. Try doing that with the third sentence of the fourth paragraph of a chat response. Nodes give you handles on information.

• Nodes are individually addressable: click to focus
• Nodes have relationships: edges show connections
• Nodes have hierarchy: parent-child structure is explicit
• Nodes are comparable: lay two side by side
• Nodes are memorable: spatial position aids recall

The Network Effect of Nodes

One node is a fact. Two connected nodes are a relationship. A tree of nodes is a decision space. The power of node-based thinking isn't in any single node. It's in the structure that emerges when nodes connect.

Persephonie doesn't just break text into pieces. It reveals the structure that was always there, hidden inside the paragraphs, waiting to be seen.