Appendix D: Glossary

🤖 AI Drafted (Not reviewed)

Library. A .fichero package on disk holding your documents and everything Fichero knows about them. One window works with one library.

Document. Anything imported into the library: a PDF, a scan, a recording, a folder. Documents form a tree.

Page. One page of a document. Most extracted facts are anchored to a page.

Rendition. An alternative image of the same page — the archival original, an enhanced version. Flip with ↑/↓.

Reader. The center pane that shows the page, with tabs for the page itself, its extracted knowledge, and your notes.

Inspector. The right-hand panel: Source, Notes, Knowledge, and Artifacts sections for the selected document.

Annotation. A marking tied to a page, text span, or region. Can be revealed on the page and promoted into a claim.

Note. Free-text writing linked to a document as a whole, kept separate from the source text.

Workflow. A step-by-step processing pipeline (transcribe, extract, catalogue) run over selected documents.

Preset. A ready-made workflow that ships with Fichero. There are about fifty.

Artifact. Anything a workflow produces — a transcription, a catalogue entry, a dataset — attached to its source document.

Entity. A named thing extracted from your sources: a person, place, organization, event, concept, or date.

Claim. A statement extracted from a source — who did what, where, when — tied to the page it came from.

Knowledge graph. The web of entities and claims across the whole library, browsable and searchable.

Curation. Your review of extracted material: approve, reject, merge, or suppress entities and claims.

Provenance. The recorded origin of every extracted item: which document, page, workflow, and model produced it.

Semantic search. Search by meaning: results can match your query’s sense even when the exact words differ.

Engine. The backend server inside Fichero that stores, searches, and processes your library.