← The director at the handover door
What the map can support
This is a constructed first essai, dated 24 October 2026. Its person, consortium, observations, candidates and numbers are invented. They are neither anonymised client material nor observations of a real professional. The public case demonstrates a way to reason; it supplies no evidence that a museum would buy this service. The authoring model identifier was not exposed in the run interface and is not inferred.
The method begins with a recurring interface, separates six arrangements of work, then asks about reward, access, repeatability, setup and dependence on other actors. The baseline is continued delivery with a faster personal workflow. We selected these dimensions to make the proposed opportunity contestable. No search algorithm discovered the options; no optimisation chose a winner.
“Equilibrium” is deliberately provisional
In economics, partial-equilibrium analysis typically studies a bounded market while holding surrounding conditions fixed. A Nash equilibrium requires a defined game and best responses to others’ strategies. This construction has neither an estimated market nor sufficiently specified strategic payoffs. It uses nearby arrangement as its operational term and retains equilibrium as a question for the wider foray.
The distinction matters because an attractive point need not persist. A contractor can refuse extra scope, a sponsor can decline to fund shared benefit, and a partner can leave Mara doing all the hidden support. A future run could formalise those choices, but would have to justify the game rather than infer it from these coordinates. The Nobel lecture by Leonid Hurwicz is useful for its discussion of why equilibrium and enforceability need separate care; this essay does not reproduce or apply his mechanism.
The opportunity is a combination, not an invention
Acceptance evidence, readiness and preparation for transition are established concerns. NASA’s Systems Engineering Handbook, Product Transition treats the product, accompanying documentation, people and enabling arrangements as part of transition. That source supports the importance of the boundary, not the commercial opportunity in this invented setting. The museum rehearsal, buyer structure and service route are this essai’s hypotheses.
The visual approach uses linked small multiples: the same candidates appear through different pairs of dimensions. This makes omitted tradeoffs visible without pretending that a two-dimensional projection is the whole problem. Small multiples are among Edward Tufte’s published teaching topics; this page’s particular axes and judgements are its own design. Different axis pairs are different ordinal scales even though all run from 1 to 5.
Complete opening data
All entries are authored bands. 1 means low; 5 means high on the named dimension. High burden or reliance is not a benefit. The axes are ordinal: the gap from 2 to 3 is not claimed to equal the gap from 4 to 5. Every selected point receives a ±1-band sensitivity cross, clipped to 1–5. That is a uniform reading aid, not a fitted range or confidence interval.
| Arrangement | Reachability | Net working reward | Reuse | Setup burden | Reliance on others |
|---|---|---|---|---|---|
| 0 · Director as switchboard | 5 | 2 | 1 | 1 | 2 |
| A · Faster switchboard | 5 | 3 | 2 | 2 | 2 |
| B · One acceptance rehearsal | 4 | 3 | 3 | 2 | 3 |
| C · Consortium rehearsal service | 3 | 4 | 5 | 3 | 4 |
| D · Partner-led rehearsal service | 3 | 4 | 4 | 2 | 5 |
| E · Handover platform | 1 | 4 | 5 | 5 | 5 |
Reachability combines relevant existing access and capability, higher meaning closer. Net working reward combines useful delivery, sustainable earning and a better working week after routine effort. It is a deliberately contestable ordinal summary; those components could disagree and a later run should separate them if evidence permits. Reuse concerns transfer across schemes, not general market size. Setup burden concerns effort and commitment before useful delivery. Reliance concerns agreement outside Mara’s control. These judgements make no cash-flow, expected-income or causal claim.
Exact counterfactual rules
- With no second sponsor, C and D move to reward 2; E moves to reward 1. The three receive a challenged-assumption marker. Every other value remains the same.
- With no protected first-use time, B, C and D lose two reachability bands (minimum 1), gain one setup-burden band (maximum 5), and receive the marker. Every other value remains the same.
- Both changes can apply together. They are authored scenario rules, not estimated effects. The platform already has reachability 1 and is not assigned an additional time penalty.
- Selection changes the highlighted candidate and its detail. Reset and page reload restore B with both assumptions enabled. No entered state is stored or transmitted.
The dashed connector from current position to selection is a comparison aid. It is not a feasible trajectory, geodesic, dependency edge or probability of success. The real transition logic is the three probes in the essay. Coincident coordinates remain coincident; only their letter labels are offset for legibility.
Bounded novelty audit
| Candidate | Classification | What makes it interesting—or weak |
|---|---|---|
| Faster switchboard | Familiar improvement | Useful baseline; can leave the underlying dependency untouched |
| One acceptance rehearsal | Underexploited combination | Rehearsal plus named receiving roles and a bounded record |
| Consortium service | Underexploited combination | A buyer and learning loop spanning several projects |
| Partner service | Underexploited combination | Existing delivery capacity paired with Mara’s boundary knowledge |
| Platform | Not mentioned in the fictional brief | Familiar in the world; requires several unsupported assumptions |
This audit checks only the invented brief printed in the essay. It does not establish global novelty, patentability, commercial exclusivity or what any real person has previously thought.
What would improve the evidence
A real enquiry would need current receiving-team evidence, existing provision, sponsor choices, total effort, authority boundaries and an independent second delivery. The first result to preserve would be what changed in a real acceptance decision. A paid second commission would be evidence about that buyer’s choice, not proof of a scalable market. A failed test could favour the simpler personal workflow.
The leading hypothesis is a bounded rehearsal followed by a second-use test. It is not a recommendation to create a company or buy software. The open foray prompt permits future models to replace the search and representation, including this composite reward axis, while preserving evidence and disclosure boundaries.
Reproducibility and provenance
The model data and counterfactual rules and page interaction are public. The illustrated scene was generated for this fictional case with the built-in image-generation tool. It contains no source screenshot or real person’s likeness. The selected data are available in the table even without JavaScript. Browser controls alter only a temporary demonstration; there is no save or acceptance action to mistake for a real record.