FROM PUBLISHED DRAWING TO EXPLICIT EVIDENCE
Explore what a drawing records, what its author reconstructs, and what a new diagram can responsibly show.
01 / CHOOSE AN EVIDENCE FRAME
Functional groups are interpretations, not chronological phases.
Drawing coordinates remain local. The export is deliberately not geographic GeoJSON.
02 / CHECK THE DOCUMENT BEFORE THE GEOMETRY
Seven published plan figures from five publication, plus one photograph of a model, were inspected. PDF object counts describe complete pages: table rules and page borders are not archaeological vectors.
03 / RENDERING IS A METHOD, NOT AN UPGRADE IN CERTAINTY
A drawing made from stated measurements can expose a publication’s argument. The frigidarium example keeps the reported basement length separate from the conditional upper-floor dimensions. Changing the width explores an estimate; it does not select the ancient building’s true dimensions.
The Sidi Ghrib diagram, which redraws seven generalized room areas, is held back from this public copy until reuse permission for the source plan is settled. Source numbers connect each shape to the author’s description. It omits wall fabric, door details, mosaics and undigitized spaces. Curves and polygons simplify the printed geometry; no independent error estimate is available.
A GIS layer needs a known coordinate system, identifiable controls, independent checks, a scale appropriate to the question and explicit drawing conventions. The composite Carthage coast and Utica palaeogeographical panels need their underlying evidence and model histories before they can support measured overlays.
A vector version remains dependent on its source. Redrawing does not itself grant permission to republish a plate. These pilots remain in the local study build, with the publication and redrawing lineage retained. Public reuse of source-derived geometry requires a separate rights decision.
