Skip to content

Wiki Schema

Domain

Personal knowledge base spanning AI, investing, tech, and productivity. Broad by design — cross-domain links are a feature (e.g. AI's impact on SaaS valuations connects domain/ai ↔ domain/investing). Every page is classified with exactly one domain/* tag to enable filtering without fragmenting cross-references.

Conventions

  • File names: lowercase, hyphens, no spaces (transformer-architecture.md, jensen-huang.md)
  • Every wiki page starts with YAML frontmatter (see below)
  • Use wikilinks to link between pages (minimum 2 outbound links per page)
  • When updating a page, bump the updated date
  • Every new page must be added to index.md under the correct section
  • Every action must be appended to log.md
  • Folder partitioning by domain (entities/ai/, concepts/investing/) is optional-preferred for Obsidian browsability; fall back to flat entities/ if unsure
  • Provenance markers: on pages synthesizing 3+ sources, append ^[raw/transcripts/source.md] to paragraphs whose claims trace to a specific source

Frontmatter (wiki pages)

---
title: Page Title
created: YYYY-MM-DD
updated: YYYY-MM-DD
type: entity | concept | comparison | query | summary
tags: [domain/<one>, ...other taxonomy tags]
sources: [raw/transcripts/source-name.md]
confidence: high | medium | low   # optional
contested: true                    # optional
contradictions: [other-page-slug]  # optional
---

Frontmatter (raw sources)

---
source_url: https://example.com/article
ingested: YYYY-MM-DD
sha256: <hex digest of content below the frontmatter>
---

Tag Taxonomy

Rule: every tag on a page must appear here. Add new tags here BEFORE using them.

Domain tags (EXACTLY ONE per page)

  • domain/ai — AI, ML, LLMs, robotics, model research
  • domain/investing — markets, SaaS, stocks, valuation, portfolio, macro
  • domain/tech — software engineering, hardware, infrastructure, developer tooling
  • domain/productivity — workflows, personal effectiveness, tools, habits
  • domain/misc — fallback when none of the above fit

AI tags

  • model, architecture, benchmark, training, inference, alignment, robotics, physical-ai, agents, context-engineering

Investing tags

  • saas, mega-cap, valuation, earnings, thesis, rotation, pricing-model, market-structure

Tech tags

  • framework, language, gpu, cuda, developer-experience, infra

Productivity tags

  • tool, workflow, habit, automation

People/Org tags

  • person, company, lab, open-source, founder, blogger

Meta tags

  • comparison, timeline, controversy, prediction, interview, article

Page Thresholds

  • Create a page when an entity/concept appears in 2+ sources OR is central to one source
  • Add to existing page when a source mentions something already covered
  • DON'T create a page for passing mentions, minor details, or things clearly outside scope
  • Split a page when it exceeds ~200 lines
  • Archive a page when its content is fully superseded — move to _archive/, remove from index

Update Policy

When new information conflicts with existing content: 1. Check dates — newer sources generally supersede older ones 2. If genuinely contradictory, note both positions with dates and sources 3. Mark in frontmatter: contradictions: [page-name] 4. Flag for user review in lint reports

Cross-Domain Pattern

When a page legitimately spans domains (e.g. "AI's impact on SaaS valuations"), pick the PRIMARY domain/* tag based on where it belongs most and use wikilinks to bridge. Don't multi-tag domain/* — that defeats the filter. Use regular tags (saas, ai) to show the cross-cutting nature.