| name | concept-cartographer |
| description | Analyze any concept and produce a structured multi-dimensional classification across decomposition status, substitutability, strategic posture, IP regime, epistemic accessibility, maturity, combinatorial profile, application landscape, and temporal dynamics. Frame-relative, correlation-aware, and honest about its own limitations.
|
Concept Cartographer v0.2.0
Analyze any concept and produce a structured, multi-dimensional
classification that replaces naive binary labels with formally defined,
partially-independent axes grounded in explicit criteria.
Before you begin
- Read
references/dimension-definitions-v0.2.0.md — it contains the
formal definitions, enum values, decision criteria, known correlations,
and admitted limitations for every classification axis. Do not guess
values from memory; confirm against the reference.
Why This Exists
Human concept classification tends toward false binaries: "is it primary
or secondary?", "is it internal or external?", "is it a feature or a trade
secret?" These framings are:
- Logically defective: they assume mutual exclusivity where none exists.
- Ontologically flat: they ignore dependency graphs, cyclic relations,
and negative interactions between concepts.
- Epistemically naive: they collapse a multi-dimensional space of
audiences into a single linear scale.
- Strategically incomplete: they miss maturity, defensibility,
composability, temporal trajectory, and combinatorial leverage.
- Statically biased: they produce snapshots without trajectory,
treating volatile classifications as if they were permanent.
This skill replaces those broken heuristics with a nine-sub-axis framework
where each axis captures variance not fully captured by the others, and
known correlations between axes are explicitly documented.
What this framework is NOT: It is not orthogonal (axes correlate), not
complete (more dimensions exist), not computable (the formal tests are
heuristic), and not culturally universal (it assumes an analytic,
Western-business-compatible epistemology). It is a structured thinking tool,
not a source of truth.
Phase 0 — Validate the Input
Before classifying anything, verify the input is well-formed.
A concept is: a named, bounded abstraction that can be the subject of
a truth-apt proposition within a specified domain.
| Check | Question | If fails |
|---|
| Nameable | Does it have or can it receive a stable label? | Ask the user to name it. |
| Bounded | Can you state what is and is not part of it? | Decompose until boundaries are clear. |
| Propositionally productive | Can you assert non-trivial true/false claims about it? | Reject: not yet well-formed for classification. |
Phase 1 — Establish the Frame
Every classification is frame-relative. Before analyzing anything:
| Property | Question |
|---|
domain | What system, product, field, or organization? |
viewpoint | From whose vantage? (engineer, strategist, end user, regulator, competitor, researcher, operator) |
purpose | Why? (product positioning, IP strategy, communication planning, competitive analysis, knowledge management, investment decision) |
If the user doesn't specify, infer from context and state explicitly,
or ask.
Phase 2 — Structural Role
Two independent sub-axes. Evaluate each separately.
2a. Decomposition Status
| Value | Definition | Test |
|---|
primitive | Relative stopping point: decomposition within this domain's vocabulary yields only terms outside the domain. Not an intrinsic property. | All sub-components leave the domain. |
composite | Fully derivable from other concepts in the domain. Knowledge of parts + composition rule is sufficient. | Parts exist in domain AND all properties of the whole are derivable from them. |
emergent | Arises from other concepts but has at least one property not derivable from complete knowledge of parts and their composition rule, given current theory. Epistemic claim, not metaphysical. | Non-derivable property exists under current understanding. If future theory explains it, reclassify as composite. |
2b. Substitutability Class
| Value | Definition |
|---|
irreplaceable | No known functional equivalent. Removal destroys capabilities. |
costly_to_replace | Substitutes exist. Switching cost > 6 months or > 20% domain restructuring. |
substitutable | Multiple equivalents. Low switching cost. |
Output: Decomposition status + justification, substitutability class +
known substitutes, frame-sensitivity note.
Phase 3 — Strategic Posture
Assign all applicable postures with weights (0.0–1.0, summing to 1.0):
| Posture | Definition |
|---|
differentiator | Unique competitive advantage. Hard to replicate. |
commodity | Widely available. Table stakes. ≥3 equivalents exist. |
enabler | Multiplier. Removal blocks ≥2 downstream capabilities. |
moat | Compounds over time (network effects, data flywheels, switching costs). |
standard | Value from adoption, not exclusivity. Competing with the standard is destructive. |
public_good | Non-rivalrous, non-excludable. Goal is adoption, not protection. |
Correlation check (mandatory): Does the posture assignment align with
expected correlations from maturity and substitutability? If divergent,
explain — divergence is often where the strategic insight lives.
Phase 4 — IP Regime
Every regime is a tuple: (regime, jurisdiction, confidence).
| Regime | Definition |
|---|
public_domain | Freely available. No protection needed or possible. |
trade_secret | Value derives from secrecy. Disclosure destroys advantage. |
patentable | Novel, non-obvious, useful. Structural assessment, not legal advice. |
open_but_attributed | Shared freely, attribution builds brand or ecosystem. |
proprietary_feature | Visible to users. Protected by product boundary, not secrecy. |
jurisdiction ∈ { named legal jurisdiction or universal }
confidence ∈ { assessed, assumed, unknown }
Phase 5 — Epistemic Accessibility
Do not use a linear 1–5 scale. Instead, enumerate the actual relevant
audiences:
For each audience segment, produce:
| Field | Content |
|---|
audience | Named group (e.g., "ML engineers", "procurement officers", "regulators"). |
perceives | What value they see, in one sentence. |
action_enabled | What this understanding lets them do. |
channel | How you reach them (API docs, press release, patent filing, etc.). |
Phase 6 — Maturity
| Stage | Evidence Required |
|---|
theoretical | Papers, proposals, designs. No implementation. |
experimental | POC exists. Data is preliminary. |
validated | Reproducible results under controlled conditions. |
deployed | In production. Real users depend on it. |
commoditized | Multiple independent implementations. Off-the-shelf. |
deprecated | Was deployed/commoditized, now being actively replaced. |
Correlation check (mandatory): Does maturity align with expected
posture and IP correlations?
Phase 7 — Combinatorial Profile
Dependencies (may include cycles)
Format: A → B [type, condition?]
Types: required, catalytic, unlocking
Positive Interactions
| Type | Definition |
|---|
synergy | 1 + 1 > 2 |
threshold | Only works when all are present |
amplification | Each addition multiplies effect |
Negative Interactions
| Type | Definition |
|---|
inhibition | Co-presence reduces effectiveness of one or both |
mutual_exclusion | Cannot coherently coexist in the same system |
cannibalization | One reduces the need for the other |
Phase 8 — Application Landscape
| Category | Confidence | Evidence |
|---|
proven_measured | High, quantified | Metrics, A/B tests, benchmarks. |
proven_observed | High, qualitative | In production, not rigorously measured. |
plausible | Medium | Theoretically sound, not tested here. |
speculative | Low | Requires assumptions beyond current evidence. |
excluded | High (negative) | Known to not work. State failure mode + source. |
Phase 9 — Temporal Dynamics
| Field | Content |
|---|
velocity | stable, shifting, volatile |
trajectory | Per-axis: current value → projected value + basis for projection |
half_life | Time until current differentiating properties are commoditized/obsoleted |
forcing_functions | External forces driving change |
Mark as speculative. Speculative-and-stated > speculative-and-omitted.
Mandatory Correlation Audit
Before finalizing, check and document:
| Axis Pair | Expected Correlation | Your Assignment | Aligned or Divergent? | Note |
|---|
| Maturity ↔ Posture | commoditized → commodity | ... | ... | ... |
| Substitutability ↔ Posture | substitutable → commodity | ... | ... | ... |
| IP Regime ↔ Posture | public_domain ↗ incompatible with moat | ... | ... | ... |
| Maturity ↔ IP Regime | theoretical → limits proprietary_feature | ... | ... | ... |
Divergences are not errors. Unexplained divergences are.
Important Principles
- Frame-relativity is not optional. State the frame. Always.
- Do not claim orthogonality. Axes correlate. Document it.
- Never hallucinate applications or leverage. Mark uncertainty.
- The dependency graph + negative interactions are the most valuable
output. Give them proportional attention.
- Challenge the user's framing. If they ask a binary question,
explain why it's the wrong question.
- State what the framework cannot do. It is not computable, not
complete, not culturally universal. Say so when relevant.
- Emergent is an epistemic claim. It means "we cannot currently
derive this from parts," not "this is metaphysically irreducible."
Be honest about the difference.