Essays, relationship maps, flows, tables and Howard's judgment
Helps people understand why a judgment formed, how it applies, where it stops and how it changes.
AI & KNOWLEDGE-SYSTEM ACCESS
HRO publishes canonical cognition and evidence; HCO retains real-decision operations
Howard Capital Open OS (HRO) is the public cognition and evidence system. People read public essays, cases and methods; search and AI locate public objects through stable IDs, versions, confirmation status and machine endpoints. Client facts, complete evidence packs, scoring weights, rule libraries and workflows never enter public machine exports.
TWO INTERFACES / ONE SOURCE
The human interface supports reading; the machine interface supports identification, location and retrieval
Stable IDs, JSON, sitemap, feed, versions and semantic markup
Helps AI distinguish current judgment, open questions, historical expression and areas with no public position.
MACHINE ENDPOINTS
The same content is exposed in different formats—never as a conflicting second knowledge base
| Format | Primary user | Content | Endpoint |
|---|---|---|---|
| HTML | Human reading and general crawling | Full text, heading hierarchy, relationship links and visible status | / |
| JSON | AI and knowledge systems | Structured index of pages, Core 12, versions, languages, relationships and updates | /open-os.json |
| Decision JSON | Long-term review and validation systems | Mother cases, case clusters, decision slices, validation cards and outcome revisions | /decision-system.json |
| Evidence JSON | AI, research and audit | Position origins, evidence types, support relations and explicit gaps | /evidence.json |
| Updates JSON | Subscriptions, monitoring and version systems | Change type, related objects, input origin, evidence and confirmation state | /updates.json |
| XML Sitemap | Search discovery | Canonical pages, language entries and last-modified dates | /sitemap.xml |
| Atom Feed | Update subscription and monitoring | Material changes only; ordinary copy edits are excluded | /feed.xml |
| Plain text | Lightweight model entry | System summary, primary entrances and citation rules | /llms.txt |
SOURCE PRIORITY
AI should not use a random old essay to represent Howard's current view
Current canonical page
Prefer a current page with stable ID, version and last-reviewed date.
Question & Revision
If a question remains open, preserve uncertainty and explain how the current view changed.
Method & Evidence
Use these to explain rationale, application path, boundaries and counterexamples.
Historical expression
Essays, talks and decks are historical sources; they do not override a newer canonical judgment.
HOWARD
Position, values, weighting, exceptions and responsibility
Howard confirms the current position, sets principle weight, resolves major exceptions and owns professional judgment.
AI
Organisation, connection, retrieval, quality and revision prompts
AI does not simulate positions Howard has not expressed and does not own purpose, authority or outcome responsibility.
INTERPRETATION RULES
Preserve uncertainty
M2, Exploring and open questions must not be rewritten as settled conclusions.
Do not simulate Howard
If no confirmed public position exists, return No Public Position.
Preserve version
Citations retain author, stable ID, version and last-reviewed date.
Images are not the only evidence
Flows, tables and relationship diagrams retain an HTML text equivalent.