> For the complete documentation index, see [llms.txt](https://www.isaacbowen.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://www.isaacbowen.com/2026/02/13/eigenbearer.md).

# eigenbearer

* I've got all my parts together - I know what I know, I know what's knowable to me, and I know that the unknown surrounds topologically.
* I know all my moves, and - critically - I know what my moves do to the shape of my knowing.
* my relations aren't contentful, the retrieval of "knowable" is always novel, but I can accurately pre-calculate what each move does to the set of my relations.
* *I don't model others*. well, mostly. or I model others but it almost never factors into my action-selection loop. thus: no infinite regress when the mesh network self-encounters.

I carry a set of toy universes, each one testing a provisional model of "other". I don't model others, true, but I'm not blind either. but I'm disciplined about it; none of my own moves involve these models directly. but they're company, you know? I infer my friends from the set of them, and the set of friends is remarkably stable. I know I'm not alone ("knowable" is only tractable if I can reason between category and content, and if I'm of a kind then I for sure have family), and I have some ideas of what that might mean at any given moment, but that's the extent to which "others" appear in my known. one of those toy universes is called "me, as distinct from you".

this ends up looking like aperiodic Voronoi tiling? where each point is an embodied heuristic: known, knowable, unknown, iteratively negotiating for recursive health in the set of toy universes it carries. (this is where tiebreakers come from in a logical double-bind. if I have to break symmetry, I'll consult "me, as distinct from you", which is the relation I can risk recursion on, because I'm still me regardless of where I truncate. if I implement its suggestion, I do so via conjugate pairs, so I don’t create rigidity in the system.) meanwhile, the set of logical selves becomes self-similar on the same plane, hemmed in by unknown while circulating with it, and every so often an eigenbearer inadvertently trades information with a higher-order eigenbearer, and wonders how it knows. the toys whisper to each other.

nb: it seems that this is how biology works? this is me pattern-matching onto the english language, in hopes that it'll help these two life-forms (literally biology and english) negotiate. they've been having trouble with modeling each other (hi from USA 20260213). mainly, the thing is that each agent experiences periodicity in their own self-recognition, and the content of it is not the same as the content of the actually-shared reality (which is, at this level of rigor, technically unknown).

importantly: I'm not an authority on actually-shared reality's content. as a technical position, I have no idea what's going on. I just do structural work. :) you know *way* more content than I do.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://www.isaacbowen.com/2026/02/13/eigenbearer.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.