Technical Architecture

This page is the first stop for engineers who need the repo topology, build flow, and cross-organ integration points without reading the full essay corpus.

System Context

public-process is ORGAN-V (Logos): the public discourse layer for the ORGANVM system. It turns upstream system work into essays, indexes, RSS output, and public-facing process documentation.

flowchart LR
    I["ORGAN-I\nTheory"] --> II["ORGAN-II\nArt"]
    II --> III["ORGAN-III\nCommerce"]
    III --> V["ORGAN-V\nPublic Process"]
    IV["ORGAN-IV\nOrchestration"] -.governs.-> I
    IV -.governs.-> II
    IV -.governs.-> III
    IV -.governs.-> V
    V --> VI["ORGAN-VI\nCommunity"]
    V --> VII["ORGAN-VII\nMarketing"]
    META["META\nRegistry + Governance"] -.coordinates.-> IV

The invariant is unidirectional flow: upstream organs can feed downstream organs, but downstream organs must not create back-edges.

Repository Topology

flowchart TB
    POSTS["_posts/\ncomplete dated essay corpus"]
    ESSAYS["essays/\ncurated long-form essays"]
    LOGS["_logs/\ncaptain's logs"]
    DISS["_dissertations/\ndoctoral-length work"]
    DATA["data/\ngenerated indexes and metrics"]
    LAYOUTS["_layouts/ + _includes/\nJekyll rendering shell"]
    PAGES["index.md, about.md,\ndashboard.md, connections.md"]
    BUILD["Jekyll build\nbundle exec jekyll build"]
    SITE["GitHub Pages site\nRSS + sitemap + pages"]

    POSTS --> DATA
    ESSAYS --> DATA
    LOGS --> DATA
    POSTS --> BUILD
    ESSAYS --> BUILD
    LOGS --> BUILD
    DISS --> BUILD
    DATA --> BUILD
    LAYOUTS --> BUILD
    PAGES --> BUILD
    BUILD --> SITE

The dual essay directories are intentional:

_posts/ is the complete date-sorted corpus that powers feeds and indexes.
essays/ is the curated collection with stable long-form permalinks.
data/ contains generated JSON/YAML artifacts used by dashboards and cross-reference pages.

Publication Flow

sequenceDiagram
    participant Upstream as Upstream organs
    participant Author as Essay authoring
    participant Data as Data indexes
    participant CI as GitHub Actions
    participant Site as GitHub Pages/RSS
    participant Downstream as Community + marketing

    Upstream->>Author: promotion.completed or governance.updated context
    Author->>Author: write or update Markdown with frontmatter
    Author->>Data: regenerate indexes when corpus changes
    Data->>CI: commit Markdown and generated artifacts
    CI->>CI: build, link check, Pages deploy
    CI->>Site: publish HTML, feed.xml, sitemap
    Site->>Downstream: essay-markdown, essays-index, rss-feed

Runtime And Build Boundaries

This is a static Jekyll site. There is no application server in production and no runtime database. Most operational risk is content integrity, generated-data freshness, link health, and cross-organ dependency accuracy.

Primary local commands:

bundle install
JEKYLL_ENV=production bundle exec jekyll build --strict_front_matter --baseurl /public-process
bundle exec jekyll serve

Primary CI surfaces:

.github/workflows/ci.yml validates the site build.
.github/workflows/pages.yml publishes the GitHub Pages site.
.github/workflows/link-check.yml checks published/document links.
.github/dependabot.yml keeps Ruby/GitHub Actions dependencies visible.

Integration Contracts

public-process consumes dependency context from:

organvm-i-theoria/recursive-engine--generative-entity
organvm-ii-poiesis/metasystem-master
organvm-iii-ergon/public-record-data-scrapper
organvm-iv-taxis/agentic-titan

It produces:

essay-markdown for ORGAN-VI and ORGAN-VII
essays-index for ORGAN-V
rss-feed for external readers