Geyserich / Institutional architecture brief · EN
Historical intelligence infrastructure, with a source trail.
Geyserich is developing a connected historical knowledge layer where archival sources remain traceable and relationships, context and uncertainty become structured.
Discussion draft · Phase 3 · Revision 0.1 · Not a report of completed infrastructure or investment terms.
Use your browser’s print dialog to save as PDF. The print layout retains development status and limitations.
The problem and intended approach
Historical knowledge is distributed across records, languages and collections. Geyserich’s infrastructure direction extends beyond a single application: a shared data model would connect source-linked historical datasets with institutional workflows and researcher review. The intended workflow is archival input → AI-assisted reading → entity mentions and evidence → evidence graph → researcher-reviewed output. Genealogical research is one application domain.
Platform responsibilities and current boundaries
Archive Network / Institutional source acquisition & collaboration
Partnership framework. Archive Network provides the proposed framework for institutional source acquisition and collaboration. Its intended role is to bring collection context, custody and agreed permissions into the knowledge workflow; participation would be agreed with each holder.
No connected archive network or institutional partnership is claimed here.
StirpeAI / Application and interpretation layer
Current application focus. StirpeAI’s current focus is AI-assisted reading, transcription and translation for genealogical research. Its intended role in the infrastructure is to turn readings into source-bound mentions and interpretations for researcher review, with extraction kept separate from identity matching.
The wider extraction, evidence-graph and review workflow is a development target. The public demonstration is authored and illustrative; it does not run the live product or an extraction model.
Hesperides / Historical Knowledge Graph Layer
Development target. Hesperides — Historical Knowledge Graph — is the development target for connecting entities, evidence and historical context. Its intended model structures claims, people, events, places and sources while preserving time, uncertain geography and competing identity hypotheses.
This site does not expose a production graph database, search API or cross-archive identity service.
Ladon / Integrity, provenance & access principles
Planned architecture layer. Ladon’s purpose is to carry integrity, provenance and access principles across the architecture. Its intended role is to retain source lineage and decision history, with collection access conditions applying to original records and derived data.
These are design requirements, not a security certification or a claim of implemented controls.
Evidence and review principles
- Preserve the source. Keep the original reference, text and image region alongside every interpretation.
- Separate observation from inference. An observation is not an inference; extraction is not identity. A proposed graph connection is not confirmation or historical truth.
- Retain context through time. Event dates, place interpretations and later review decisions must remain distinguishable.
- Make review reversible. A corrected reading should not erase the record or the reason for an earlier decision.
A bounded institutional conversation
Institutional workflows and specialized archival knowledge shape the intended data architecture. A proposed pilot starts with collection scope and research questions, then agrees rights, review responsibilities and a handover. Source-linked readings, proposed entity links and documented limitations are intended deliverables; volume, format and access conditions would be agreed with the holder. No completed pilot or partnership is asserted.
Long-term data architecture and open questions
The long-term value of the intended data model lies in retaining source references, collection context and review history as datasets grow. Provenance can help make that knowledge foundation defensible through inspectable evidence and accumulated domain understanding; it is a design rationale, not a demonstrated competitive advantage. Graph persistence, entity-resolution evaluation, permission enforcement and revision history still require implementation and validation. The public visualization uses authored fictional passages and demonstrates no live end-to-end platform.