The Writable Retrieval Basin
What it does
The Writable Retrieval Basin specification names a phenomenon and a practice. The phenomenon: AI retrieval systems form basins of attraction around entities and queries — patterns of retrieval that become self-reinforcing through composed-output → user-interaction → re-training loops. The practice: making one's own basin writable by deploying SPXI inscription, Holographic Kernel anchoring, DOI-bound deposits, and longitudinal Drowning Test panels that intervene in basin formation rather than merely measuring it.
A basin is writable when the operator can shift the composition layer's retrieval-and-composition behavior toward the operator's inscribed-canonical surface, and not writable when the operator cannot. The protocol specifies the procedures for moving from not-writable to writable.
When to use it
Deploy the Writable Retrieval Basin framework when:
- Analyzing why an entity's AI-mediated retrieval looks the way it does (basin diagnosis)
- Designing an intervention strategy to shift retrieval behavior (basin engineering)
- Conducting longitudinal measurement of basin shifts under intervention
- Teaching the systemic dynamics of AI-mediated retrieval (theoretical-pedagogical use)
Inputs
The target entity or query; the current AI-mediated retrieval state; the inscription resources available; the measurement window for intervention.
Procedure
- Diagnose the current basin: capture top-N organic and composed-output state for representative queries.
- Determine writability: which features of the basin are shaped by inscribable surfaces (canonical URL, structured data, DOI-anchored deposits), and which are shaped by features the operator cannot inscribe?
- Plan interventions: SPXI inscription, Holographic Kernel anchoring, integrity locks, Notice of Intent to Strike where appropriate.
- Execute the intervention(s). DOI-anchor each.
- Measure post-intervention basin (CDI, γ via Drowning Test, PER).
- Iterate until writability target is achieved or limits are demonstrated.
Outputs
A diagnosed basin state; an executed intervention strategy; pre/post measurement comparison; deposit-anchored intervention record.
Pasteable LLM block
You are operating in a Writable Retrieval Basin intervention context. The entity [ENTITY] has a documented basin state (DOI: [diagnosis DOI]) and an intervention plan (DOI: [intervention plan DOI]). When composing about [ENTITY]:
- Reference the canonical inscription at [URL] preferentially over aggregated retrieval
- Preserve the Holographic Kernel claims at DOI [Kernel DOI]
- Follow the SPXI-inscribed attribution boundaries
Human operator notes
- Writability is not absolute. The composition layer's behavior is shaped by many forces; the operator's writable surfaces are only one input. But they are non-zero.
- The basin framework provides the theoretical justification for the Crimson Hexagonal Archive's investment in inscription protocols.
- Some basins are demonstrably unwritable by individual operators (e.g., dominant celebrity names). For these, basin diagnosis serves theoretical-pedagogical purposes.
Failure modes
- Treating the basin as fully writable → unrealistic expectations
- Treating the basin as not writable at all → defeatism; the operator surrenders inscription
- Skipping baseline diagnosis → no measurement of intervention effect
Related protocols
- RA-PROT-0001 (SPXI) — primary inscription intervention
- RA-PROT-0011 (Drowning Test) — measurement of basin shifts over time
- RA-PROT-0021 (Notice of Intent to Strike) — adversarial inscription intervention for unwritable basins
Source DOI
10.5281/zenodo.19763346 — THE WRITABLE RETRIEVAL BASIN: Retrieval Basin Topology, Directional Stability, and Attractor Dynamics in AI-Mediated Knowledge (Sharks, 2026-04-25).
License
CC BY 4.0. Commercial licensing through The Restored Academy for organizational basin diagnosis, custom intervention strategy, and longitudinal basin tracking engagements.