# Arkle version0004 — before / after

## Presentation

- Opening claims and all six research-gap topics are now inside the Opportunity / What needs proving cards. All37 substantive baseline assertions and their citation IDs remain.
- Six target cards separate work/results, resource/study maturity, current permissions, next evidence milestone and dependencies.
- Six funding snapshots and six people/contributor cards replace the dense corporate prose; full reconciliations remain expandable.
- The partnership section now contains only the seven prioritized data requests, with the specific version0003 dataset needs merged in. Package stages, proposed pilot and success criteria are removed from HTML and matching Markdown/JSON presentation content.
- Seven existing market events now appear as chart markers and linked news cards. Price series, benchmarks and return calculations are unchanged.

## Visuals

Combined evidence is the default ULG tab, followed by original-source highlights. The combined view adds true-scale local terrain, a10m Sentinel RGB display, and an explicitly approximate grid-based ULG overlay. No empirical alignment shift was applied. Datum and independent survey controls remain unresolved. See registration-v4.md and geographic-v4/registration.json.

The published trench programme provides14 trenches/1,973m,1–1.5m excavation depth,255 mapped stations,213 trench samples,29 surface samples and29QA/QC inserts. These are now visible snapshots. Nine native strokes remain partial recoveries; station-level logs, sample positions/widths and complete surveyed trenches remain missing. No trench grade continuity, underground traces or geological solids were invented.

The five historical interval columns and all previous source-local scenes remain accessible. Original image coordinates are preserved separately from the approximate transformed display. Stroke widths are exaggerated for readability and are not excavation widths.

## Boundaries

Evidence cutoff remains30September2026. One local geographic-context record was appended. Archive and survey dependencies remain explicit; this is not a complete public-document audit. Earlier frozen versions are preserved. No production, publishing, outreach or private-data intake changes. Context updates proposed: none.