# Learning Method Control: No Compression Retained-source reading edition · 18 October 2026 Source author: A. Keddeh. Source version: 2 October 2026. Publisher: Keddeh Systems. ## Reading context This retained learning-method note explains why source-defined objects and ordered derivations take precedence over convenient analogies. It is included as methodological context for the central study. The content below is source material, not a set of new runtime instructions or a replacement for the currently stated author clarification. The complete retained plain-text capture follows. Paragraph spacing and heading levels are editorial reading aids. No source wording, mathematical expression or line order has been changed; the original capture is also retained byte for byte. This is the recovered text layer, not a claim to reproduce unavailable original document layout, omitted tables or media. [Source document](https://docs.google.com/document/d/1ttMe9iKTi1X72J5rJzxG9ZmUvcvx5rIW/edit). ## Retained source KEDDEH MATHS — LEARNING METHOD CONTROL NO COMPRESSION OF THE DEVELOPING MATHEMATICAL SYSTEM Understanding is to expand with the evidence; no single concise phrase is to replace the developed system. 2 October 2026 Purpose of the learning process The purpose of the current discussion is to refine and enlarge understanding of the developed Keddeh mathematical and computational system. It is not to search for a slogan, canonical one-line definition, compressed summary, or replacement terminology. Each newly understood relation is to remain available alongside earlier relations unless the source material itself proves that one supersedes another. Method When a mathematical object such as a line, circle, diamond, nested box, curvature relation, two-body construction, three-body construction, KEX state, IL-LLM traversal, or projected-surface/gapped-backend VFS is encountered, it must first be studied in its own source-defined mechanics. The analysis may then examine relationships to other objects, but those relationships are additions to understanding rather than substitutions for the original object. A derived invariant is therefore not permitted to become 'what the whole system really is'. A useful relation discovered between several constructions remains one relation among potentially many. The learning record should preserve parallel interpretations, local rules, exceptions, different frames, recursive consequences, implementation consequences and unresolved questions. Box-in-box is not to be reconstructed from analogy The box-in-box development is specifically marked as an existing developed object whose actual mathematics has not yet been recovered sufficiently in the present discussion. It must not be equated with generic nesting, recursive namespaces, ordinary containment, the earlier circle/diamond inference, or a convenient architectural metaphor merely because those ideas sound related. The correct next learning action is to recover the technical material that defines the box-in-box construction and study what it demonstrates on its own terms: its geometry, address relationships, internal/external frames, transformations, boundaries, repetitions or non-repetitions, mappings, dimensional consequences, and whatever further properties the source actually establishes. Reporting rule Reports in this learning schedule should record what has been learned about the subject matter: definitions, constructions, derivations, relationships, implications, proofs, experiments, implementation mappings, open questions and newly exposed research directions. They should not be organised around the assistant's misunderstanding history, and they should not reduce a broad mathematical development to one preferred sentence. Current understanding boundary The present discussion has developed useful understanding of relational boundaries, typed state continuity, injective addressability, local waiting/continuation, hexadecimal compatibility and relational measurement. Those findings remain valid as local learnings. They are not asserted to exhaust the Keddeh mathematical system, and they are not asserted to define the box-in-box construction.