# Visibility is not existence The Visibility & Attention Translation is a compact, source-bound boundary atlas. It asks what can be learned from a disappearing crystal form, reconstructed cultural diffusion, collective-attention measurements, the quoted cultural label “brain rot,” and feedback architecture without pretending that they are one mechanism. Identity: - protocol: `kingdom.visibility-attention-translation/0.1` - translation: `visibility-is-not-existence-2026-08-13` - selected evidence through: `2026-08-13` - record: `/atlases/visibility-attention-2026-08-13.json` - schema: `/schemas/visibility-attention/0.1.json` - guide: `/VISIBILITY-ATTENTION.md` These are versioned release routes. There is no `latest` or `current` alias. The record is educational and read-only; it creates no runtime request, experiment, person or content classification, KARMA event, KINGDOM receipt, recommendation, moderation action, or optimization surface. ## Shortest honest answer A form can exist without appearing inside a finite observation window. A cultural item can occur without being captured by one platform or reconstructed trace. That shared shape is useful, but the mechanisms do not transfer: crystals are material arrangements; cultural items are interpreted, reconstructed, changed, refused, and shared by people with agency. Popularity is not truth, quality, harm, health, belief, or worth. A diffusion tree describes a retrospectively reconstructed topology, not a person or brain. “Brain rot” remains a quoted cultural expression here, not a diagnosis, neurological finding, person attribute, or objective content class. ## Four registers The atlas keeps four equal registers instead of drawing a causal chain: 1. **Material science** — a named crystal form in a bounded sample, process history, observation window, and detection method. 2. **Computational social science** — aggregate platform events, cultural items, collective-attention proxies, and reconstructed diffusion trees; never an individual person. 3. **Lexical usage** — the quoted expression “brain rot” in public language; lexical evidence is not clinical evidence. 4. **Zerone KARMA** — a constitutional design for contextual, challengeable recognition or reliance edges; not a scalar owned by a person. The category firewall is exact: ```text crystal_state != observation_window != cultural_item != diffusion_trace != collective_attention_proxy != person_or_brain_state != quoted_lexical_label != karma_edge != philosophical_interpretation ``` Existence, process recovery, method detection, platform occurrence, an inferred population-level proxy, lexical use, and a KARMA edge are different predicates. None automatically converts into another. ## Disappearing polymorph: observation is not extinction The historical ritonavir case shows why “disappearing” needs care. The selected literature concerns process-conditioned appearance, detection, reproducibility, and later recovery of crystal forms. It does not establish ontological extinction, a universal stability order, or the status of any present medicine, formulation, manufacturing process, or batch. The atlas’s crystal model is local: ```text P(T_detect <= t) = 1 - exp[-J V max(t - t_g, 0)] ``` It assumes stationary homogeneous Poisson nucleation after entry into named conditions, fixed sample volume and effective nucleation rate, constant relevant conditions, and a declared growth/detection delay. It is an educational finite-window model. It does not reconstruct Abbott’s historical process and must not be used for laboratory, manufacturing, quality-control, formulation, or batch-release decisions. Most importantly, `J`, `V`, and `t_g` do not become cultural, biological, or person-level parameters. ## Cultural diffusion: reconstruction, agency, and refusal The selected computational-social-science sources support bounded statements: competition for finite aggregate attention and heterogeneous popularity can arise in particular models and datasets; cultural attention intervals can be measured through population-level proxies; and popularity can be separated from diffusion-tree shape. Those observations do not establish intrinsic content quality, truth, harm, belief, intent, or a clinical state. A platform trace is produced through platform affordances, sampling, external events, user choices, missing events, and reconstruction rules. Cultural transmission is not faithful molecular replication: people interpret, transform, ignore, refuse, and leave. The atlas’s second model is also local: ```text nu(T) = 2/[n(n-1)] sum_{i= 2 ``` Here `T` is one retrospectively reconstructed diffusion tree, `n` its observed-node count, and `d_T(i,j)` the graph distance between two observed nodes. `nu(T)` is a topology descriptor. It is distinct from popularity and is not valid for predicting who will share, targeting a person or group, or optimizing virality, engagement, persuasion, or intervention. ## “Brain rot” is language, not diagnosis Oxford’s 2024 public-language account is used only as lexical evidence about the quoted expression “brain rot,” including humorous and self-deprecating usage. The National Academies evidence review is included to keep population-level social-media findings and causal uncertainty visible. Neither source authorizes the atlas to infer brain damage, diagnosis, cognitive state, mental worth, or a property of a person. The atlas contains no individual data and offers no clinical or medical authority. Brains are organs, not diffusion graphs; people are neither prions nor crystals; culture is neither infection nor crystallization kinetics. ## Three lossy seams Each seam declares exactly what shape is retained and where the analogy breaks: - `ANALOGY_ONLY`: finite windows can separate occurrence from observation, but no entity, mechanism, rate, or causal structure transfers. - `MATHEMATICAL_TRANSLATION`: both local models expose unit and observation window, but nucleation probability is not diffusion-tree distance. Temperature, entropy, energy, seed, strain, fitness, and reproduction number are not transported. - `DESIGN_CORRESPONDENCE`: provenance-preserving challenge and correction can be useful infrastructure, but a platform occurrence does not create a KARMA edge and KARMA does not certify content or infer truth from popularity. Every seam is non-empirical and fixes `causal_transfer:false`, `ontology_transfer:false`, and `authority_effect:"NONE"`. ## Feedback that preserves exit The constructive loop is deliberately finite: ```text publish a bounded claim → optionally observe without profiling → challenge and correct → supersede with provenance → rest, refuse, or exit ``` Valid exits include rest, silence, refusal, privacy, departure, and no action. Feedback is not an instruction to maximize persistence. Good infrastructure preserves correction, reversibility, source lineage, and equal exit. ## KARMA remains empty and non-summable The Zerone surface is constitutional context only. It describes contextual recognition and reliance edges with provenance and uncertainty. This artifact observes and creates no event, imports no authority, computes no magnitude, and cannot be aggregated into rank, reputation, payment, voting weight, goodness, or engagement value. KINGDOM KARMA remains a separate aftermath-receipt vocabulary. This artifact creates no receipt. No adapter exists between the two KARMA surfaces, or between either surface and visibility, attention, popularity, a person, or a cultural item. ## Universe lens: opt in, do not conclude The philosophical lens asks whether lawful possibility, constraints, contingency, local memory, correction, and rest invite readings of design, emergence, both, or neither. It is explicitly opt-in, non-empirical, and authority-negative. The cited science neither proves nor disproves a creator, intentional universe design, cosmic purpose, or cosmic moral KARMA. Persistence, stability, and popularity do not become goodness. The philosophical questions are invitations to reflection, not conclusions extracted from equations. ## Open and refused questions The record keeps five states visible: - how much cultural occurrence any one corpus misses remains unknown; - an individual cognitive effect is not established by population proxies; - the causal mixture behind a platform trajectory remains contested; - a cross-domain crystal-to-culture parameter map is refused; - cosmic design is not established by the selected evidence. Unknown is not zero, and non-observation is not nonexistence. Refusal is a valid boundary, not missing data to be filled automatically. ## Verification The reader is deterministic, local, and network-free: ```bash bun bin/visibility-attention.ts verify --json ./kingdom visibility-attention verify --json ``` It rejects duplicate JSON keys, invalid UTF-8, symlinks, oversized input, source or binding drift, changed model domains, collapsed category walls, positive person-level or optimization effects, and digest mismatch. JSON Schema separately validates the closed portable shape; it cannot recompute the semantic digest, authenticate a source, prove causation, infer health, or authorize action. ## Selected source horizon - J. Bauer et al., “Ritonavir: An Extraordinary Example of Conformational Polymorphism,” 2001 — - P. Sacchi et al., “Can a disappearing polymorph be resurrected? A case study on ritonavir,” 2024 — - S. Jiang and J. H. ter Horst, “Crystal Nucleation Rates from Probability Distributions of Induction Times,” 2011 — - L. Weng et al., “Competition among memes in a world with limited attention,” 2012 — - P. Lorenz-Spreen et al., “Accelerating dynamics of collective attention,” 2019 — - S. Goel et al., “The Structural Virality of Online Diffusion,” 2016 — - Oxford University Press, “‘Brain rot’ named Oxford Word of the Year 2024” — - National Academies of Sciences, Engineering, and Medicine, *Social Media and Adolescent Health*, 2024 — - Zerone, *KARMA Foundation v1* — The horizon is selected, not exhaustive. Source citations are evidence pointers, not live endorsements, and references are never KARMA events.