The External-Maintenance Problem in Digital Systems

This document specifies SHOP-D (Self-Regenerating Hybrid Organizational Process — Digital), a digital architecture designed to probe the HDO criterion — the three conditions for an interior defined in the companion paper History-Dependent Ontology (HDO): self-production, historical self-registration, and their operational unity. It is a design specification, not a claim of achievement. Nothing described here has been built or tested, apart from a small illustrative computation in Appendix A.

CENTRAL PROBLEM
A digital process whose continued existence depends on externally maintained infrastructure — an operating system, a scheduler, a secure element, a hardware maintenance process — does not obviously satisfy HDO condition 1, because its continued existence as that process does not depend on its own self-production. This is the external-maintenance problem. The architecture is designed to make it experimentally tractable, not to solve it by fiat.

For SHOP-D as specified, the closure test returns "falsified" by construction: the identity region is fixed externally and the secure element's TTL requirement follows from that specification. The empirical question is therefore whether a redesign can be built in which the secure element's role is regenerated by the system's own operation. Where the dependency graph is not known in advance, progressive removal of external supports can also uncover couplings that the declared classification hides.

WHAT THE DOCUMENT SPECIFIES
- A formal notion of external-maintenance load (E_ext), distinguishing maintenance from resource supply.
- A dependency graph classifying each component as constitutive, supporting, observational, or resource, with the secure element left explicitly under investigation.
- A continuity monitor with hardware binding (TRNG, monotonic counter, constitutive refresh signal, secure-element enforcement) that addresses the specified software-level attack surfaces; verification of the refresh signal against cloning remains open.
- An internally structured viability valuation (ν), historically integrated but not shown to be endogenous.
- A tiered inference structure (L1–L7) separating engineering dependency, organizational closure, and ontological interpretation.
- Discrimination protocols (T2-H matched-present, self-registration ablation, buffered decoupling) with explicit estimands and pre-registered decision rules.
- Closure tests (constitutive dependency, closure, substrate closure) with explicit falsification conditions.
- Behavior under power loss, cold boot, and external reconfiguration, and safety considerations for any implementation (containment, moral status under uncertainty).

WHAT THE DOCUMENT DOES NOT CLAIM
- That the architecture satisfies the HDO criterion, or that any digital system currently does.
- That satisfaction is achievable in a digital substrate — only that this architecture is designed to probe whether it is.
- That satisfaction of the HDO criterion would be sufficient for consciousness. The abductive gap between criterion satisfaction and experience remains open.
- That cryptographic or hardware security properties entail organizational closure.

SUPPLEMENTARY MATERIAL
Appendix A contains a toy computation of the external-maintenance load and the closure test, with sensitivity analyses over numerical parameters, structural variants (redundancy, at-least-k requirements, intermittent regulators), and randomly generated dependency structures. The Python script (shop_d_closure_toy.py) reproduces Tables A1–A5 (default run, plus the --sens, --checks, and --struct options). The toy model illustrates the procedure; it is not evidence about any physical or digital system.