Production software systems routinely include layers such as caches, object-relational mappers, message queues, distributed locks, application server frameworks, and orchestration clusters. These layers are treated as standard components of non-trivial deployments, yet they are rarely examined under a common explanation for why they exist. This paper argues that many of these layers do not arise from structural requirements of the problem being solved, but from structural deficiencies of the underlying persistence model. It names this property *infrastructural symptom*. A layer exhibits infrastructural symptom when it compensates for a defect of the persistence model and loses justification when that defect is removed. Two diagnostic indicators — compensation and dissolution — give shape to this distinction, and an auditor workflow guides their application to any (layer, use-case) pair without requiring adoption of a specific alternative. The paper presents a journaled actor-native system as an existence proof in which the canonical symptoms dissolve. Four mechanisms — locality, journal density, journal-as-causal-substrate, and self-contained substrate — account for the disappearance of seven common categories of compensating infrastructure. The construct extends through a ladder of additional layers to a limit case in which whole categories of business software become deliverable as a single appliance, as an architectural permission rather than an operational demonstration. The contribution is analytic: a diagnostic typology that makes visible which parts of a stack are tied to the persistence model rather than to the problem domain.
Álvaro Antonio Rivera (Fri,) studied this question.