FPF.Reference:10 - Thinking-Oriented Architecture, Not A Descriptive Upper Ontology

Preface node heading:fpf-reference-10-thinking-oriented-architecture-not-a-descriptive-upper-ontology:1319

What this page is

This is generated FPF reference text from the specification preface or supporting sections. It helps interpret FPF; it is not FPF Reference product documentation.

Methodology

Use it to understand how the specification wants to be read, then return to a route, pattern, or work packet for active work. Cite generated IDs only when the wording changes the task decision.

Content

FPF uses ontological distinctions to make reasoning usable across domains. Its practical question is what the practitioner must understand, distinguish, decide or do, and which meanings must remain stable for that use. The architecture connects distinctions with guidance for obtaining a result, inspecting its grounds and reconsidering it when conditions change.

Start from the receiving use and recover the subject and claim it needs. Identify what is being asserted, about which entities, between which participants and under which conditions. Writing can help obtain this meaning before a stable name or classification is available. F.19 restores the predicates and participants compressed by wording; E.10 directs an unresolved meaning to its subject pattern. For a Characteristic, A.18 requires a defined aspect and Scale; A.19 organizes already defined characteristics and Scales without supplying their missing subject meaning.

When a current pattern already gives the needed claim or action, apply it directly. E.24.CD helps find that sufficient use. Several claims may need to be read together without requiring another kind or ontic. A single episteme must satisfy C.2.1; otherwise the claims can remain in separate epistemes or working notes. This choice preserves the subject and useful reasoning.

Classification becomes useful when the work needs to distinguish members from non-members under a stable criterion. C.3 and C.3.1 recover the candidate domain, membership condition and boundary; C.3.2 governs the relevant candidate judgement. The individual kind is an instance of U.Kind. A KindSignature describing its criterion, a judgement about a candidate and a representation of its members remain different objects. Use a reusable signature when a later use needs that declaration. Distinguish a supported non-member judgement from an unknown result or a candidate outside the declared applicability. A local kind can serve its context without acquiring a new public U-name.

E.24.UK governs the stronger proposal to introduce a durable U-kind. It requires an independently identified subject of classification, a defining identity or membership rule, practical need and dependent use, and comparison with sufficient existing expressions. Its root and two dependent-kind dispositions have different grounds: an additional classification may concern the same individual, or a distinct individual may have a separately defined identity dependence. Use that pattern for the full admission conditions.

An ontic answers a different question: which identity or constitution rule and which relations must remain coherent across named dependent uses? E.24 defines this connected ontology unit through its primary subject, the smallest necessary relation set, the defining rules and the reliance of its consumers. This lets several patterns reuse compatible meanings instead of repeatedly redefining them. When existing direct rules already close the uses, keep those rules; when several governed claims need only a bounded joint account, use that account. Concrete cross-pattern disagreement or duplication can justify opening E.24 to determine whether a durable ontic is needed.

A kind and an ontic are therefore not paired one for one. An ontic can coordinate several admitted kinds; a newly admitted kind can use existing coordination or sufficient direct rules. Where both are needed, E.24 and E.24.UK return separately grounded results in one decision.

These distinctions lead to different useful outcomes:

  • Recorded values. An evaluation may need several characteristic values, each on its declared Scale, together with the conditions for interpreting them. A selected space or a profile of those values can be named and reused without thereby admitting a new kind of spaces or profiles. Classifying such constructions is a further question only when a receiving use needs it.
  • An admitted dependent kind. B.5.2.0 defines U.AbductivePrompt as an additional classification of the same episteme about an unresolved explanatory question. Its consumers need the shared entry distinction and the returns when the question, account or grounds change; the general U.Episteme classification alone does not express that distinction. The subject pattern supplies membership and exclusion conditions. This use retains an admitted kind without adding another ontic.
  • Shared ontological coordination. A.19 keeps a CharacteristicSpace's basis, characteristic-to-Scale bindings and Coordinate meanings coherent. A.19.ECS uses that organization to construct evaluations, while E.21, E.9.DA and E.2.DA keep their different evaluated subjects, evidence requirements and results. Reusing the common rules makes their meanings recoverable without making their evaluations interchangeable. Changing a selected Scale requires a changed space and reconsideration of comparisons that used the old one; it does not by itself change the common ontic.

When revising any of these expressions, establish what changed. A clearer name or another publication form may preserve the subject and claims. A changed membership condition reopens dependent classifications; a changed characteristic, Scale or relation rule reopens the comparisons and actions that relied on it. Preserve unaffected results whose grounds still hold, and retain the written reasoning needed to locate the affected use. If the necessary subject meaning or evidence is missing, keep that specific question open.

Ontology thus serves practical reasoning together with the writing described in section 9. It helps distinguish a method description from performed Work, a claim from its grounds and a subject from its presentation, then directs the practitioner to the contribution needed next. FPF uses domain ontologies, mathematics, standards and evidence with their own meanings and limits; it does not replace their subject knowledge.


Last Updated: 2026-11-22 — upstream FPF commit 7a93d8f8 (github.com/ailev/FPF)