# INTENT > This file captures **why this repository exists**, > the **direction it is moving toward**, and > the **kind of system it is meant to become**. > It is intentionally **aspirational and stable**, not a description of current implementation. --- ## One-liner **The Railiance architecture-definition home — turning evolving repo families, workload models, and substrate realities into a coherent framework vocabulary and boundary model.** --- ## Why This Exists As Railiance grows, implementation repos naturally solve immediate problems: infrastructure, cluster runtime, platform services, forge operations, application releases, and ecosystem modeling. What those repos cannot safely do on their own is define the framework's shared meaning. Without a canonical architecture home: * the same terms drift across repos, * new repo families appear without clear boundaries, * execution models and workload wrappers get mixed into the wrong layers, * and migrations happen by local habit rather than deliberate framework design. This repository exists to define the **shared architectural language** of Railiance before that language hardens accidentally inside implementation repos. --- ## Operating Context Railiance does not exist in isolation. It is part of a wider, coevolving ecosystem around **Railiance**, **Net Kingdom**, and the **Helix Forge** software factory. That ecosystem is intended to support the full path from: * product ideation, * discovery, * delivery, * operation, * marketing, * and monetization. The architecture therefore has to work across experimental and production-grade realities at the same time. That matters because repo boundaries in Railiance are not only about code organization. They shape how humans and agents coordinate responsibilities, reason about workload placement, and evolve operating models without losing the ability to scale into more mature security and delivery expectations. --- ## The Mission > *Where we are going.* To become the **canonical home for Railiance framework architecture** — where repository taxonomy, architectural boundaries, execution-model concepts, managed workload patterns, substrate concepts, and migration direction are defined once and referenced everywhere else. This means: * Railiance vocabulary is defined through **explicit architectural documents** * New repo families and framework terms are introduced through **clear decisions** * Ownership, execution mode, workload identity, and substrate identity remain **separate and composable** * Explorations mature into **stable decisions** before they spread across the wider repo landscape * The framework can support both **early-stage experimentation** and **production-grade operation** without collapsing those needs into one vague structure --- ## Core Principles ### 1. Architecture Before Proliferation New repo patterns should be named and bounded deliberately before they multiply across the ecosystem. ### 2. Shared Vocabulary Is Infrastructure Terms such as `railiance-*`, `rail-*`, `rapp-*`, and `reef-*` are not cosmetic. They shape ownership, tooling, and operator understanding. ### 3. Separate Axes Cleanly Ownership, execution architecture, workload packaging, and substrate reality must not collapse into one ambiguous repo type. ### 4. Decisions Need A Canonical Home Framework-level architecture decisions should live in one source-controlled place rather than being reconstructed from scattered repo-local assumptions. ### 5. Stable Meaning, Evolvable Model The framework vocabulary should stay understandable over time even as specific rails, workloads, and substrates evolve. ### 6. Guide Implementation, Do Not Shadow It This repo should define structure and direction, not absorb the operational content that belongs in the implementation repos themselves. ### 7. Stabilize The Default Path First Railiance should not introduce multiple top-level rails speculatively. The default path for platform services and managed applications is `rail-kubernetes` until there is a concrete workload need and a sound argument for a distinct rail. ### 8. Managed Packaging Is Not Ownership `rapp-*` repos exist to wrap third-party or self-built workloads so they can be run as fully managed Railiance workloads in the wider Railiance and Net Kingdom context. They do not replace the ownership repos that define why a capability exists. ### 9. Substrates Need Purpose, Not Just Names `reef-*` repos should represent compute resources organized around a defined purpose and operational boundary. They should not be created merely because a named machine exists. --- ## What This Is (Conceptually) This repository is: * an **architecture-definition home** * a **repository taxonomy authority** * a home for **framework-level ADRs and conceptual models** * a place to define how Railiance repos **compose across multiple axes** * a **migration map** from current repo reality toward cleaner framework structure --- ## What This Is Not This repository is not: * the infrastructure substrate * the Kubernetes runtime or workload execution layer * the shared platform-services layer * the forge runtime * the application release surface * the implementation of the ecosystem graph registry * a dumping ground for operational runbooks that belong elsewhere It is the **place where the framework explains itself**. --- ## Direction of Evolution This repository is expected to evolve toward: * clearer **repository-family definitions** and lifecycle rules * stable **rail**, **rapp**, and **reef** contracts * stronger **boundary guidance** for new repos and migrations * better linkage between architecture decisions and `railiance-fabric` declarations * a repeatable path from **exploration -> ADR -> adopted framework pattern** * a clear model for when experimental multi-rail substrates are acceptable and when production-grade separation is the better default --- ## Guiding Question > **How can Railiance define its own structure clearly enough that every new repo, rail, managed workload, and substrate increases capability without increasing ambiguity, even as the ecosystem grows from exploratory operation into production-grade responsibility?**