sarahdurks/sarah-durkin-okf

A hand-authored Open Knowledge Format (OKF) v0.1 bundle for sarahdurkin.ai — a small-scale learning example of the format

★ 0Forks 0HTMLGitHub ↗Compare

README

sarah-durkin-okf

A small, hand-authored Open Knowledge Format (OKF) v0.1 bundle representing the public career and personal site of Sarah Durkin.

It exists as a learning exercise — to see what OKF feels like at human scale, not just dataset scale. Every concept here was written by hand; no enrichment agent was involved.

🔗 Live interactive viewer

👉 sarahdurks.github.io/sarah-durkin-okf

The viz.html file is published via GitHub Pages — open the link above to explore the bundle as an interactive force-directed graph with full concept rendering, search, and backlinks. No install, no backend; everything is embedded in the page.

What is OKF? OKF is an open, vendor-neutral format for representing knowledge as plain markdown files with YAML frontmatter, organized in a directory tree. It was introduced by Google Cloud as part of the knowledge-catalog project. The format is human-readable, agent-readable, diffable in git, and portable across tools.

Browse the bundle

Or start from the bundle index and click through the directories like a wiki — GitHub renders the OKF markdown beautifully on its own.

What's in it

20 concepts across six directories:

Directory Concepts What they represent
people/ 1 The subject of the bundle.
roles/ 7 Professional roles, newest first.
companies/ 4 Organizations where roles were held.
projects/ 1 SarahBrain.ai, the AI community.
interests/ 4 Books, and Harold the dog.
references/ 3 External authoritative sources (LinkedIn, Forrester Wave 2019, Better Collective acquisition announcement).

Cross-links between concepts (roles → companies, companies → references, sarabrain_ai → person, etc.) form a 65-edge graph rendered by viz.html.

Source of truth

All content is derived from https://www.sarahdurkin.ai. The bundle is a deliberate re-shaping of that site into OKF — same facts, different representation, suitable for ingestion by LLM agents, graph viewers, and anyone who prefers cat to a CMS.

Reproduce the visualization

The viz.html file is checked in. To regenerate it:

# Clone the upstream knowledge-catalog (where the visualizer lives)
git clone https://github.com/GoogleCloudPlatform/knowledge-catalog.git
cd knowledge-catalog/okf

# Install the package
python3 -m venv .venv
.venv/bin/pip install -e .

# Regenerate viz.html for this bundle (adjust path to wherever you cloned this repo)
.venv/bin/python -m enrichment_agent visualize \
  --bundle /path/to/sarah-durkin-okf/bundles/sarah_durkin \
  --name "Sarah Durkin OKF"

Notes on link style

The OKF spec (§5.1) recommends absolute, bundle-relative links (/people/sarah_durkin.md) because they're stable when documents move within a subdirectory. However, the reference visualizer in knowledge-catalog currently only follows relative links when building its graph (./other.md, ../sibling.md). To make the checked-in viz.html show every relationship, this bundle uses relative links throughout — which the spec also permits (§5.2).

A spec-conformant viewer should accept both forms. Filed as a thing to notice while learning the format.

License

The bundle content is © Sarah Durkin and is shared for demonstration and educational use. The OKF format itself is open and unowned.

Issues