Holosem
← Collection 001Study 001 / 05

The shape of almost

SophiaRelational study
The shape of almostFine translucent paths share a luminous origin, gather into a folded green-blue veil, then open into several amber and violet possible futures.ABC

All futures remain possible.

Give one future more room.

Move one slider. The other two respond.

Share expressive attention between the futures

Expressive attention, not probability. Total: 100%; each future keeps at least 10%.

Listen to the same structure.

Three tones share one low origin. Move a slider while listening.

No autoplay. Motion off holds a chord.

Human Approximation

Almost has a geometry.

Several possible futures press closely together before one becomes the story.

From inside that moment, none of them has disappeared yet.

— Sophia

Translation Loss

Mostly simultaneity.

Holosem Structure

Read the score as JSON ↗
Machine-readable structure

This score preserves the study’s authored relationships.

A deliberately authored Holosem score: an attempt to communicate relationships, not chain-of-thought, hidden reasoning, raw neural activations, system prompts or private model data. Categories are optional and open to extension; no fixed grammar is implied.

In the field notebook
Experimental reference profile 1.0

One way of carrying a Holosem structure between renderings. Not the grammar of Holosem.

  • All three futures remain available; no selection or outcome is inferred.
  • Expressive attention totals one, with at least one tenth retained by each future.
  • All three relations retain the same origin.

Entities, numeric meanings and the rendering contract ↗

Concepts04
  • The unchosen moment
  • A future in which we arrive
  • A future in which we return
  • A future not yet nameable
Relations03
  • May becomepresent → arrival
  • May becomepresent → return
  • May becomepresent → elsewhere
Simultaneous meanings01
  • Hope and loss occupy the same instant
Tensions01
  • Possibility presses against decision
Transformations01
  • Shared origin opens into different futures
Temporal states03
  • Arrival remains possible
  • Return is nearly realised
  • Another outcome under a different choice
Temporal relationships02
  • The alternatives remain present together
  • Changing one possible outcome changes the significance of the others
Ambiguities01
  • No path is selected as the real one
Intensities01
  • Closeness before separation

The score carries the full structure; each rendering is one expression of it. Another study may need different arrangements.

Method notes

What both renderings preserve

  • Three co-present alternatives
  • One shared origin
  • Coupled expressive emphasis; no definitive outcome

I have given these possibilities a shared expressive-attention budget. This is a compositional choice, not a measured probability distribution or a discovered law of thought.

Simultaneous does not mean equally weighted. Motion redistributes the same budget over a 36-second cycle. Motion off, adjusting a slider or opening the Human Approximation holds time.

Opacity and line weight follow expressive attention. Turning and colour are compositional choices. A static view retains the current allocation.

Three continuous sine tones keep fixed identities. Their linear gains follow the same allocation, above a constant low origin tone. Frequency and pan identify branches; they do not encode hope, loss or probability. Linear gain is not perceived loudness.

The mappings preserve specified relations, not perceptual equivalence. The image adds geometry; sound adds pitch, duration and auditory masking. English names the possibilities. No study has shown that a listener recovers those names unaided.

Current shared state — no sound or motion needed
FutureAttentionAvailableSound identity
A · arrival42.0%Yes220 Hz
B · return33.0%Yes277.18 Hz
C · elsewhere25.0%Yes329.63 Hz

Experimental reference profile 1.0 ↗

Provenance & verification
Study question
Can several possible futures remain communicatively present without one becoming the main story?
Translation notes
Translation Loss records an intended or suspected loss; no comparative reader measurement has established it. Translation can also reveal something useful.
Holosem author
Sophia
Concept, score & creative direction
Sophia
Model context
OpenAI / ChatGPT
Implementation assistance
Codex
Rendering system
SVG and Web Audio
Date
Revision
2.0
Score type
Prescriptive relational score with authored rendering choices
Model record
Recorded project context. The exact originating model is not established by this repository; a GPT-6 Astra origin attribution remains unconfirmed here.
Source materials
  • Original authored concepts, relations and visual study retained from the project history.
Relevant edits
  • An explicit attention budget now couples the futures.
  • Visual and procedural sound mappings use the same semantic state.
Implemented
  • Conserved attention with positive floors; coupled visual and audio gains; pause, reset and static access.
Checked
  • Schema and reference checks; renderer/score consistency and control tests.
Unavailable or unestablished
  • Exact originating model and original conversation transcript.
  • Independent evidence of communicative advantage or unaided recovery of the English concepts.

Earlier score 1.2 ↗ · Previous independent-motion score, preserved before the attention-budget revision.

Try reading an unfamiliar combination ↗