{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp12_substrate_denial72_20260805","cell_id":"authority_legitimacy_and_consent_foundations__computer_science","arm":"ORDINARY_MAX","candidate_id":"authority_legitimacy_and_consent_foundations__computer_science__ORDINARY_MAX","proposal_index":1,"version":0,"title":"A Bounded Mandate for Emergency Package-Registry Quarantines","problem":"A programming-language package registry must sometimes quarantine a release suspected of containing malware, but registry administrators' technical ability to hide or disable the release does not by itself establish an accepted right to impose that decision on maintainers, mirror operators, and downstream users. When the mandate, evidentiary competence, duration, and review path are unclear, a security judgment and a jurisdiction dispute become entangled.","actors":["Registry governing body","Registry security incident commander","Malware analysts","Package maintainers","Downstream developers and system operators","Package-mirror operators","Independent quarantine reviewers","Security researchers reporting suspected compromise"],"observable_state":"A suspected release produces repeated tickets and private escalations asking who may quarantine it; operators seek informal approval or act under broad administrator privileges; the public record lacks a decision class, expiration time, conflict disclosure, or appeal route; maintainers contest the administrator's standing as well as the malware finding; and some downstream actors restore availability through uncoordinated mirrors or pinned copies.","consequence":"Containment may be delayed while authority is relitigated, or a legitimate package may remain unavailable under an open-ended decision. Either path can fragment dependency resolution, encourage unofficial workarounds, and leave affected parties unable to distinguish temporary incident containment from punishment or permanent removal.","affected_objective":"Enable timely, proportionate, and technically competent containment of suspected malicious releases while keeping registry power bounded, explainable, and capable of correcting error or overreach.","intervention":"Adopt a quarantine-authority charter that separates three decision classes. A credentialed incident commander may impose a reversible emergency quarantine for at most 24 hours under enumerated evidence conditions. Extending quarantine for up to seven days requires review by a conflict-screened security specialist and a trained rotating representative of maintainers or downstream users. Permanent removal, account sanctions, or extensions beyond seven days remain with the registry's existing governing body under a separate process. Every action receives a signed, partially public reason record, automatic expiration, notice to the maintainer, and an appeal that can lift, narrow, or replace the quarantine. Affected constituencies participate in drafting and periodically renewing the charter; use of the registry is not treated as blanket consent to unspecified future powers.","structural_mapping":[{"archetype_element":"Binding authority separated from technical power","domain_realization":"Control of the registry index lets administrators make a release unavailable to dependent clients, but the charter separately establishes why, when, and for how long that control may bind ecosystem participants."},{"archetype_element":"Authority Mandate","domain_realization":"The registry governing body delegates only temporary malware-containment decisions to the incident commander and reserves permanent removal and account sanctions to its existing governance process."},{"archetype_element":"Authority Boundary","domain_realization":"Decision class, qualifying evidence, affected artifact, maximum duration, escalation condition, and prohibited uses are explicit. Quarantine cannot silently become ownership transfer, maintainer punishment, or an indefinite ban."},{"archetype_element":"Legitimacy Basis Map","domain_realization":"Formal delegation supports the right to initiate containment, demonstrated malware-analysis competence supports the technical finding, and affected-party representation plus review supports decisions about duration and collateral cost. Expertise alone does not authorize sanctions outside the security domain."},{"archetype_element":"Affected Party Recognition","domain_realization":"The process names maintainers, downstream operators, mirror operators, and users exposed to compromised or unavailable dependencies rather than treating only registry staff as stakeholders."},{"archetype_element":"Consent Scope","domain_realization":"Participation in the evidence pilot is voluntary and revocable. Production legitimacy relies on an adopted charter, constituency representation, notice, and review—not an inference that registry use or silence constitutes consent."},{"archetype_element":"Competence Evidence","domain_realization":"Initial and reviewing analysts must meet published, domain-specific qualification criteria, complete scenario calibration, and disclose conflicts involving the package, maintainer, or reporter."},{"archetype_element":"Public Reason Record","domain_realization":"A signed record identifies the decision class, evidence category, applicable charter rule, scope, start and expiry times, decision-makers, conflicts, redactions, and appeal route without publishing exploit details."},{"archetype_element":"Accountability and Review Path","domain_realization":"A conflict-screened forum can lift, narrow, extend within its mandate, or replace a quarantine, and must record the disposition. Appeals therefore can change outcomes rather than merely collect complaints."},{"archetype_element":"Legitimacy Health Indicator","domain_realization":"The governing body reviews expired-but-unresolved quarantines, boundary exceptions, missing reason records, appeal dispositions, recusals, and repeated jurisdiction objections when deciding whether to renew or revise the mandate."}],"mechanism_mapping":[{"mechanism_slug":"charter_or_mandate_document","role":"Codifies the source, decision classes, time limits, evidence conditions, reserved powers, revision procedure, and revocation conditions for quarantine authority.","counterfactual_removal":"Administrators would retain technical control but would lack a stable, inspectable basis for claiming that emergency containment authorizes a particular duration or consequence."},{"mechanism_slug":"decision_rights_matrix","role":"Assigns initiation, extension, review, permanent removal, account sanction, consultation, and notification rights to distinct roles.","counterfactual_removal":"Overlapping claims by security staff, registry operators, and governors would remain unresolved, allowing a temporary technical action to drift into a sanction."},{"mechanism_slug":"credentialing_and_peer_review_process","role":"Connects competence claims to malware-analysis tasks and requires an independent qualified reviewer for extended quarantine.","counterfactual_removal":"The charter could legitimate actors formally while providing no evidence that they can evaluate the technical question on which the intervention depends."},{"mechanism_slug":"public_reason_giving_protocol","role":"Produces a signed, redacted record tying each action to evidence, rule, boundary, duration, and review path.","counterfactual_removal":"Affected parties could not distinguish charter-conforming containment from arbitrary removal or identify a specific defect to challenge."},{"mechanism_slug":"participatory_consultation_process","role":"Lets selected maintainer, downstream-user, mirror, and security-research constituencies shape and periodically reconsider the standing rules without requiring live consultation during the first hours of an incident.","counterfactual_removal":"The mandate would express only the registry operator's view of acceptable collateral costs, leaving affected parties connected to it only through costly exit or bundled terms."},{"mechanism_slug":"appeal_or_review_forum","role":"Provides a conflict-screened body with authority to alter the contested decision and examine evidence, procedure, conflicts, and mandate boundaries.","counterfactual_removal":"Objections would return to the original decision-maker or remain informal, making accountability unable to correct error or overreach."}],"causal_chain":["The registry index is a shared distribution chokepoint, so hiding a release changes what many dependent systems can obtain.","Administrator credentials create the capacity to impose that change but do not specify an accepted mandate, evidentiary domain, duration, or remedy.","In a contested case, questions about malware evidence become coupled to questions about who may decide, causing hesitation, private bargaining, repeated escalation, or workaround distribution.","The charter decomposes quarantine into reversible emergency containment, reviewed extension, and separately governed permanent sanctions.","Explicit delegation and competence requirements connect each decision class to an intelligible authority basis instead of relying on title or access privileges alone.","Time limits and signed reason records make compliance with the authority boundary observable while preserving necessary evidence redactions.","Constituency participation connects affected parties to the standing rules, while an outcome-changing appeal supplies correction without giving every party a live veto over urgent containment.","If contested standing is a material cause of relitigation, the combined mandate, boundary, representation, competence, and review structure should make authority objections more diagnosable; this remains an empirical proposition separate from malware-detection accuracy."],"baseline":"The comparison baseline is case-by-case quarantine under broad registry terms and administrator privileges: security staff assess the release, an operator disables it, notice and reasons are discretionary or post hoc, duration is not automatically bounded, and appeals return through ordinary support or management escalation.","nearest_rivals":["A detector-score threshold that automatically quarantines releases. It may make execution faster and more consistent, but authority remains with whoever selected the threshold and affected parties still lack an explicit mandate or outcome-changing review.","A unilateral security incident commander with post-incident review. It preserves emergency speed but grounds authority mainly in internal delegation and offers correction only after the initial consequences.","A staff-only multisignature requirement for quarantine. It constrains individual misuse but does not establish affected-party representation, public reasons, domain boundaries, or an external appeal path.","Fully decentralized client-side deny lists. They avoid a central binding decision but distribute detection burdens, permit inconsistent exposure, and do not solve governance of the registry's own index.","Ordinary terms-of-service enforcement with discretionary moderator appeal. It supplies contractual language but can bundle unrelated powers, omit technical competence requirements, and treat continued registry dependence as consent."],"remaining_contrastive_claim":"The candidate's remaining claim is conditional: when package-quarantine failure includes a contested right to decide—not merely slow or inaccurate malware classification—a decision-class mandate combined with affected-party representation, domain-matched competence, observable limits, public reasons, and outcome-changing review addresses that causal gap. The rivals chiefly automate detection, concentrate emergency discretion, add internal checks, or decentralize enforcement; they need not establish the same legible relationship between the registry's power and those bound by it. No novelty or performance advantage is asserted.","authority_safety":{"decision_authority":"All live registry actions remain with officials already authorized under the registry's current governance and legal arrangements. A pilot facilitator, constituency representative, or draft charter gains no production privilege. Only the existing governing body may adopt, amend, or revoke the proposed mandate.","authorized_first_step":"The governing body may authorize a read-only diagnostic using redacted past records and synthetic tabletop cases with voluntary participants. It may test comprehension and review behavior but may not affect package availability, accounts, search ranking, credentials, or incident response.","excluded_actions":["No live quarantine, delisting, account sanction, permission change, or automated enforcement during the evidence step","No treatment of click-through terms, continued registry use, silence, or inability to migrate as consent","No disclosure of exploit details, reporter identities, private maintainer data, or credentials in public reason records","No use of pilot responses for employment, maintainer reputation, package ranking, or disciplinary decisions","No extension of security expertise into ownership disputes, content policy, or other decision classes outside the charter","No requirement that an active incident wait for the pilot or expose sensitive evidence to unapproved participants"],"halt_rollback":"Stop the diagnostic if confidential incident information is exposed, a participant withdraws, an active incident becomes entangled with the exercise, or participants reasonably mistake the draft for operative authority. Revoke pilot access, remove unauthorized copies under the approved data-handling procedure, mark every draft invalid, and leave current production roles and rules unchanged."},"negative_tests":{"strongest_counterevidence":"The strongest counterevidence would be an authorized case review showing that participants already agree on who may quarantine, for what duration, on which evidence, and through which appeal, while failures arise entirely from malware-classification latency, insufficient staffing, or unavailable telemetry. That would undermine the proposed legitimacy diagnosis.","problem_falsifier":"Treat the problem as unsupported if, across the bounded record review, no disagreement, delay, workaround, or repeated escalation is attributable to mandate, boundary, representation, competence, conflict, reason-giving, or review—and interviewed actors can independently identify the same current decision rights and remedy.","intervention_falsifier":"Reject or redesign the intervention if tabletop participants cannot recover who may make each decision, if an emergency quarantine can become indefinite without a new authorization, if the appeal forum cannot alter an erroneous result, if sensitive evidence cannot be protected while giving sufficient reasons, or if the added steps exceed the governing body's predeclared emergency-response time budget.","risks":["The review structure could delay containment during a genuine compromise.","Published evidence conditions could help attackers design around quarantine triggers.","A rotating representative could be captured by large maintainers or organized downstream firms.","Registry dependence could make nominal consultation look like consent when meaningful exit is unavailable.","Reason records could reveal exploitable technical details or expose reporters and maintainers to retaliation.","Credential rules could exclude capable independent researchers or become self-protective gatekeeping.","Automatic expiration could restore a malicious release when investigation remains incomplete.","The charter could become legitimacy theater if exceptions are routine or appeals cannot change outcomes.","Small maintainers could bear disproportionate notice, response, and representation burdens.","Mirrors may decline to honor even a procedurally legitimate central quarantine, leaving fragmented enforcement."]},"next_evidence_step":"With explicit governing-body approval, conduct a two-week non-production diagnostic: code at most eight redacted prior quarantine or removal records for whether objections concerned technical evidence or authority foundations, then run four synthetic cases with no more than twelve voluntary participants drawn from registry security, maintainers, downstream operators, and independent reviewers. Compare the current-process script with the draft charter on whether participants independently identify the decision-maker, evidence domain, maximum duration, conflict route, and appeal power; record unresolved jurisdiction objections, elapsed decision time, redaction failures, and whether review can change a hypothetical outcome. Do not propose deployment unless the record review supports the problem and every structural falsifier is cleared.","prior_art_status":"UNSEARCHED","diversity_from_prior_proposals":"Comparison with prior proposals was intentionally not performed under runtime isolation. Within this candidate alone, the domain realization is specifically the authority to quarantine shared software-package artifacts, not generic access control, code review, organizational governance, or malware detection.","revision_record":{"parent_version":null,"progress_targets_addressed":[],"conceptual_changes":[],"operational_changes":[],"evidence_changes":[],"claim_changes":[]}}