Research in the emsemioverse is a relational practice. It proceeds by building structures — pages, links, formal objects — and testing whether they cohere with what already exists. The test is not incidental to the research; the test IS the research.
The relational account
Each new piece of content enters the vault undetermined. A triage note, a slop draft, or a freshly written page has a claim — it asserts something, defines something, describes something — but it does not yet participate in the relational field. It sits outside the web of relations that gives the vault’s content its meaning.
Validation is the act of testing whether a new element holds together with what exists. The vault’s skills perform this testing concretely:
- Style checking tests whether the element’s language meets the constraints of PTGAE — whether it says what it means without vagueness or misdirection.
- Frontmatter fixing tests whether the element declares itself correctly — whether its type, tags, and metadata place it in the right categories.
- Cross-linking tests whether the element connects to what it should connect to — whether the relations it implies actually hold.
- Reference auditing tests whether the element’s connections resolve — whether the things it points to exist and point back.
When validation succeeds, the element is integrated. It participates in the relational field: other pages link to it, its terms appear in definitions elsewhere, its claims support or constrain other claims. It has a place.
When validation fails, the failure is informative. A broken link reveals a missing page. A style violation reveals an unclear claim. An orphaned page reveals a gap in the relational structure. Each failure identifies something undetermined — and that undetermined thing becomes the next research question.
The research cycle
This practice follows the cycle described in The Research Cycle:
Philosophy → Mathematics → Technology → Content → Questions → Research → Philosophy
Philosophy produces claims about what exists and how. Mathematics formalizes those claims as structures in the semioverse hierarchy. Technology implements those structures as specifications and tools. Content instantiates the specifications as actual pages, links, and metadata. Questions arise when content fails to cohere — when validation reveals gaps. Research addresses those questions, producing new philosophical claims, and the cycle begins again.
The cycle is not a sequence of phases. All stages operate concurrently. Writing a page is content work; discovering that the page has no home is a question; creating the directory structure is technology work; understanding why that structure should exist is philosophy. A single session of vault maintenance can touch every stage.
Relation to information-theoretic formulations
emsenn’s original formulation of this methodology, in a paper titled “Directed Relational Research,” used information-theoretic language. In that framing, new content increases entropy; validation reduces divergence between the vault’s state and some coherent target; convergence of validations marks equilibrium. The paper described research as gradient descent on a divergence functional.
These correspondences are real. The semiotic universe admits information-theoretic interpretations, and the closure operators that define it have analogues in divergence-minimization. But the information theory is a mathematical correspondent, not the primary description. The practice is relational: building structures and testing whether they hold together. The information theory describes one way to measure the process — not the process itself.