Problems that outlast an executor.
Six questions connect the tools and research. Each asks what another party needs to be able to do.
Commitments
What was agreed—and who may change it?
A replacement arrives after work has been accepted and part of a budget spent. The old order remains, but its next action needs current permission.
The decision: Establish the applicable terms, remaining limits, and who may authorize a change.
What existing approaches contribute
Contracts, mandates, access controls, spending limits, and durable workflow state already define parts of this position.
The contribution being tested
The handoff demonstrations keep accepted work, held operations, and current authority distinct. Buyer requirements make the evidence expected from a supplier explicit.
- A coding job across coordinator replacement — Constructed demonstration
- Buyer requirements — Research
- Compute Contract Drill — Constructed demonstration
Still unresolved: Can another operator establish the applicable mandate without rebuilding the original application’s context?
Evidence
What can another party establish?
An operator controls the only history. A recipient needs to assess a result after that operator becomes unavailable or disputes the account.
The decision: Determine which assertions the retained material supports, who held it, and what remains missing.
What existing approaches contribute
Operational logs, signed exports, backups, and independent custody can already retain relevant facts. A new record format must justify the additional work it creates.
The contribution being tested
Released Bulla checks supplied record integrity and reconciles receipts against explicitly supplied expected events. These checks establish neither authentic or complete history nor the completeness of the supplied expectations. Witnessing and retained-evidence research examine separately held histories, conflicting commitments, and later correction.
Access, purpose, and retention must be limited. A record is not permission to disclose its contents or retain them indefinitely.
- Bulla — Available software
- Witnessing — Research
- Retained Evidence — Constructed demonstration
- The economics of unfinished work — Research
Still unresolved: Which evidence is necessary for a later decision, and who can retain it at an acceptable cost?
Continuity
What must a successor carry forward?
An assignment ends with useful work accepted, an operation unconfirmed, and resources still committed. The next executor may not have existed when the work began.
The decision: Recover a usable operating position without purchasing accepted work again or treating silence as completion.
What existing approaches contribute
Durable workflows, journals, transaction identifiers, and reconciliation already support recovery within their operating boundaries.
The contribution being tested
The coding handoff shows a constructed recovery. The recorded buyer experiment tests continuation from retained evidence. Operating-account research examines what a successor needs to decide next.
- A coding job across coordinator replacement — Constructed demonstration
- Routed buyer continuity — Recorded experiment
- The economics of unfinished work — Research
Still unresolved: What conventional reconstruction work does a portable account actually remove for its recipient?
Open contribution brief
What may a successor do with an unfinished job?
A coordinator stops after one fix passes its checks, another fails, and a third purchase has no confirmed result. What can its successor establish from the retained account—and what remains unsupported?
What is the decision?
Identify accepted work, rejected delivery, unconfirmed attempts, remaining funds, and the next actions under the supplied order’s rules. Keep what the account records separate from what it authorizes within this constructed example.
What can you inspect?
Start a constructed $100 order with subcontracting permitted. Run the coding example, switch the coordinator, and stop before buying the replacement patch. Save both coding-job files at that same checkpoint and follow the existing local-review instructions. The published source archive and its manifest identify the supplied implementation.
What remains open?
The supplied implementation reconstructs this constructed account. It does not establish authentic or complete history, real-world authorization, payment, or whether another operator could use the account without substantial reconstruction work. Running the supplied checker is not independently implementing the rules.
What contribution would help?
Show a reproducible unsupported decision, identify a missing or ambiguous requirement, or describe a conventional implementation that preserves the same distinctions with less machinery. A precise explanation of why no justified decision is possible is also useful. Keep facts, requirements, and available evidence equivalent when proposing an alternative.
What would change our conclusion?
A demonstrated defect changes the bounded implementation claim. Missing information can invalidate the proposed handoff. A competent conventional solution can weaken the case for additional machinery. An anecdotal time saving does not establish economic superiority.
This is an open contribution brief—not a conducted study, verified external result, or authorization to act on a real order. The published external exercise remains preparation-only.
Recourse
What can be challenged or put right?
Later evidence defeats the basis of an earlier decision. Someone affected needs a route to review, even if they never received a receipt.
The decision: Identify what requires rechecking, who may decide, and which correction or remedy is authorized and resourced.
What existing approaches contribute
Appeals, dispute procedures, human review, and ordinary payment or refund systems contribute distinct parts of recourse.
The contribution being tested
Declared lineage identifies recorded dependencies for rechecking. Constructed accountability and effect-lock mechanisms test the separation between evidence, an authorized judgment, and a consequence.
Affected parties need a route to challenge without a receipt. An unverified accusation must not trigger an automatic economic penalty.
- Declared lineage — Research
- Accountability Circuit — Constructed demonstration
- Evidence-Locked Consequence — Constructed demonstration
Still unresolved: Can a challenge reach an authorized forum with enough evidence and resources to put a decision right?
Economics
Is the next step worth its full cost?
An output is finished, but an unresolved effect competes with new work for attention and resources. More evidence could change the best next action.
The decision: Compare production, investigation, repair, waiting, and stopping under current authority and the full burden of coordination.
What existing approaches contribute
Decision theory, management accounts, operational records, and ordinary financial arrangements already help make these choices.
The contribution being tested
Second Close makes the surviving position and revised decision basis inspectable in a finite model. Operating studies ask whether portable records remove enough work to justify acquisition, verification, interpretation, and maintenance.
Economics can reject machinery or deployment under the program’s constitutional commitments.
- The economics of unfinished work — Research
- Second Close — Decision model
- Financial backing — Research
Still unresolved: Does the proposed arrangement remove enough real work to justify its total burden, while preserving legitimate claims and challenge rights?
Exit
Can we change who we depend on?
Records can travel, yet the only institution able to recognize them or decide a dispute may remain indispensable.
The decision: Establish what can be transferred, what needs renewed recognition, and which obligations remain when a provider or institution is replaced.
What existing approaches contribute
Export rights, open formats, contractual transfer, and independent operation provide parts of practical exit.
The contribution being tested
Independent-enterprise research asks whether evidence, backing, and dispute procedures can be supplied separately. That question includes replacing Bulla and the institutions supporting it.
Exit must preserve legitimate outstanding obligations and the ability to challenge them. Continuity also permits authorized amendment, discharge, correction, and deletion.
- Who gets to operate in a machine economy? — Program essay
- Bulla — Available software
- Answerability Network — Constructed demonstration
Still unresolved: Can a replacement institution recognize the relevant position without becoming a new compulsory intermediary?