LDD System Plan
Goal: turn the LDD set into a living, zoomable project knowledge system while keeping the static site stable, reviewable, and safe to deploy today.
This is the current approval page. The earlier rebuild proposal and AI responses are now background archive material. This page is the plan to vet, approve, and then execute.
LDDs remain canonical outputs. Claude controls LDD rewrites unless David explicitly overrides the lock.
The static site becomes data-shaped: sidecar maps, stable IDs, semantic status, hierarchy, backlinks, and visible rollups.
The future wiki can ingest the same model as projects, topics, documents, signal blocks, channels, revisions, and scorecards.
Core Concept
The LDD hierarchy should not be flat, and it should not be a rigid top-down sequence. It should be a roll-up / drill-down knowledge map that lets a human or LLM move between project atmosphere, system headers, focused LDDs, decision blocks, active research, and canonical outputs.
Working model
Input → Logic / Analysis → Output. Raw comments, product ideas, screenshots, sketches, and research enter the system as inputs. They are worked through in analysis spaces: collisions, scorecards, comparisons, and debate. Mature results then become formal LDD decisions or proposed LDD revisions.
Four Axes
Target Hierarchy
Barndo Project
Atmosphere / World Bible
Great Hall thesis
Calm industrial language
Korean cooking / social intimacy
Maintenance as psychological calm
Aging-in-place dignity
Header LDDs
Building Shell
Comfort / HVAC / Radiant
Gym / Athletic Great Hall
Mechanical Core
Living Program
Site / Civil / AHJ
Budget / Build Sequence
Focused LDDs
Clerestories
Flooring
Basketball
MUA
Plumbing
Lighting
Equipment / appliances
Working Buckets
Inputs
Research
System collisions
Proposals
Canonical outputs
Execution Plan
| Phase | Build | Approval question |
|---|---|---|
| 1. Knowledge map | Expand site/assets/ldd-index.json into a richer map with hierarchy, parent/child nodes, process stage, and relationship tags. |
Do the top-level header documents feel right? |
| 2. Header pages | Create top-level LDD header pages for major systems that roll up child LDDs, status, collisions, and open decisions. | Can Peter or an LLM understand the whole section from the header? |
| 3. Working buckets | Add Input, Logic / Analysis, and Output areas so brainstorming does not prematurely overwrite canonical LDDs. | Does the process protect curiosity without polluting locked documents? |
| 4. Zoom UI | Add jump paths, backlinks, rollups, and drilldowns to LDD pages using stable IDs and sidecar data. | Can users move up/down/sideways without getting lost? |
| 5. Wiki migration | Translate the static model into the future wiki schema: projects, topics, documents, revisions, signal blocks, comments, and scorecards. | Does the static model import cleanly without rewriting meaning? |
Approval Checklist
- Approve the top-level header categories.
- Approve the Input → Logic / Analysis → Output workflow.
- Approve separating experiential intent from technical implementation status.
- Approve using the static site as a published view, not the whole source of truth.
- Approve expanding the sidecar map before deeper UI work.
Decision Needed
If this page is approved, the next implementation step should be Phase 1: expand the knowledge map. That is the lowest-risk move because it does not rewrite LDD content. It gives the project a stronger skeleton that can drive both the current static site and the future wiki.