generated from coulomb/repo-seed
9b5defa6d8a7c68ca3ecf66fde23110184d71452
SHARD-WP-0003 — nine wiki-engine deep dives chosen against the synthesis coverage review: high-insight (Federated Wiki, Wikibase/Wikidata, TiddlyWiki, ikiwiki), the INTENT-named git-forge wikis (Gitea/GitLab/ GitHub), and four for breadth/lineage (Salesforce Quip, Oddmuse, MojoMojo, UseModWiki). Each task = persist a deep-dive memo + promote/enrich the use-case catalog + log adapter-contract notes for SHARD-WP-0002; post-batch synthesis-v3 decision. SHARD-WP-0004 — eight computational / interactive-knowledge systems as design prior art for executable/computational/live page content and one-source-many-projections: Knuth WEB/weave/tangle, Mathematica & Jupyter notebooks, Processing/Processing.js, Strudel.cc, Squeak, Glamorous Toolkit, Pharo. Carries the question of whether a page can be a live computational artifact; closes with a "computational page model" synthesis. Both workplans authored file-authoritative (no state_hub IDs; sync will create+link workstreams and tasks). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
shard-wiki
Git-based Markdown wiki orchestrator and federation layer.
shard-wiki joins heterogeneous wiki-shaped page stores (shards) into a
coherent information space while preserving provenance, capabilities, and
history. It is an orchestration layer, not a wiki engine.
Status
Early-stage: Python scaffold, intent and specification groundwork, active design
workplan. See SCOPE.md for current maturity.
Documentation
| Document | Purpose |
|---|---|
| INTENT.md | Aspiration and boundaries |
| SCOPE.md | What we achieve now |
| AGENTS.md | Agent working guide |
| docs/repository-layout.md | How this repo organizes information |
Quick start
pip install -e ".[dev]"
pytest
Layout
research/ explorations (yymmdd-prefixed)
demand/ inbound unreviewed requirements
spec/ implementation specifications
workplans/ State Hub–registered tasks
docs/ stakeholder documentation
wiki/ collaborative knowledge (wiki UI when connected)
issues/ ticket mirrors
history/ archived material (yymmdd-prefixed)
Languages
Python
89.6%
HTML
4.9%
CSS
1.9%
Perl
1.6%
Makefile
1.1%
Other
0.9%