Why it is a slip-box, not a folder tree

Under everything else, Sapix stores what it knows in a particular shape, and the shape is not mine. It is the Zettelkasten, the slip-box, worked out by the sociologist Niklas Luhmann, who kept a box of around ninety thousand small linked notes and used it to write an improbable number of books. Credit where it is due: the structure is his. Sapix mostly just runs it on a machine.

The idea is to refuse the folder tree. You do not file a thought under one heading and hope to find it there later. You write it as a small, self-contained note, one idea, and you link it to the other notes it touches. Knowledge becomes a graph, not a hierarchy. The value lives in the links, not in where something was filed, because a real thought usually belongs in five places at once and a folder makes you pick one.

In Sapix the parts carry names from a biological metaphor, but they map cleanly onto the slip-box:

• A Phenomenon is the raw observation, the thing as first captured.
• An Interpretation is the refined, atomic note, the Zettel itself.
• A Synapse is a link between two Interpretations.
• A Synthesis is what emerges when a cluster of linked notes starts saying something none of them said alone.

The refining (raw into Zettel) and the linking (proposing Synapses) run quietly in the background, not as something I do by hand. That is what makes it more than a notes app: the box is supposed to surprise you, to set two old notes side by side and let a third idea fall out. Luhmann described his slip-box as a conversation partner. That is the bar.

It is also why the other notes here work the way they do. Freezing and recall act on this graph; the thing that surfaces a connection mid-conversation is just walking a Synapse. The slip-box is the floor the rest of it stands on.