Networked thought

January 28, 2026

The way we organize knowledge shapes the way we think. Hierarchical systems — folders, outlines, taxonomies — impose a tree structure on ideas that is fundamentally at odds with how knowledge is actually structured. Ideas don’t belong to single categories; they sit at the intersection of multiple contexts, and their meaning shifts depending on which context you approach them from.

Why Hierarchies Fail

Consider a note about “intermittent turbulence in stable boundary layers.” Where does it belong in a folder hierarchy? Under “atmospheric science”? “Measurement methods”? “Greenland field work”? “Fluid dynamics”? It’s relevant to all of these, but a folder system forces you to choose one, making the note invisible from the other contexts where it matters.

This isn’t just a filing problem — it’s a thinking problem. When you organize knowledge hierarchically, you tend to think hierarchically: within categories, along established paths, inside disciplinary boundaries. The most interesting research ideas often emerge at the intersections between domains, and a knowledge system that makes these intersections invisible makes them harder to discover.

The Network Alternative

A network-structured knowledge system — where the primary organizational principle is the link, not the folder — preserves the multi-contextual nature of ideas. A note about intermittent turbulence can be linked to atmospheric physics, measurement methodology, field work logistics, and mathematical modeling simultaneously. You don’t have to choose; you can navigate to it from any of these directions.

This is not a new insight. Vannevar Bush described something like it in 1945 with his concept of the “memex,” a hypothetical device that would store and link documents by association rather than index. Ted Nelson’s hypertext project extended the vision. The World Wide Web realized a simplified version of it. But most personal knowledge systems still default to hierarchical organization, because folders are familiar and links require more deliberate effort.

Associative Trails

The real power of networked thinking emerges over time, as the density of connections increases. Bush called these “associative trails” — paths through a knowledge base that reflect patterns of thought rather than categories of subject matter. When you follow a trail through densely linked notes, you often encounter unexpected juxtapositions that spark new ideas.

In my own research practice, I’ve found that the most productive moments often come from revisiting notes and discovering connections I didn’t see when I originally wrote them. A measurement technique developed for one field campaign turns out to be relevant to a completely different problem. A theoretical framework from one discipline illuminates a puzzle in another. These connections were always there in the material; the network structure makes them visible.

The Cost of Linking

There’s an important caveat: maintaining a densely linked knowledge network requires ongoing effort. Links, unlike folder assignments, need to be created deliberately. A note you write today should ideally be linked to relevant notes you wrote months or years ago, which means you need to be aware of what’s already in your system. This is where tools that surface potential connections — through backlinks, graph visualizations, or search — become genuinely useful rather than merely decorative.

The alternative — dumping notes into a system without linking them — produces a collection that looks large but is effectively inaccessible. Volume without structure is just noise.