Skip to content

Pollenomics Engine Roadmap

The current repository is a real evidence system, but it is not the final pollenomics engine. Today it can collect, normalize, and publish several evidence families into one governed output tree. The longer-term ambition is broader: compare pollen, archaeological context, and ancient-DNA evidence in a way that keeps provenance visible and refuses overclaiming.

The point of this page is to keep those two things distinct. The current system is legitimate. The larger engine is still ahead.

What Exists Now

Today the repository can:

  • collect tracked evidence from AADR metadata, LandClim, Neotoma, SEAD, RAÄ, boundaries, and fieldwork records
  • normalize those sources into reviewed files under data/
  • publish one world-to-country geography tree under docs/report/
  • expose the current weakness of the animal aDNA recovery state instead of pretending that thin atlas output already proves broad coverage

That is enough to produce useful, reviewable outputs. It is not enough to claim that every evidence family is equally deep or equally mature.

What Still Needs To Deepen

The next stage is to turn thin or uneven surfaces into explicit recovery work:

  • source acquisition must widen, especially for missing supplementary material
  • sample, site, chronology, and coordinate extraction must become defensible
  • multi-evidence joins must state where provenance differs rather than flattening those differences into one score
  • pollenomics-first explanation must stay broader than the current aDNA recovery slice
  • world, Europe-plus, Nordic, and country publication must keep sharing one product model so future-country growth does not fragment the code or docs

The roadmap is not just "collect more files." It is about making weaker parts of the system more defensible without breaking the parts that are already reviewable.

What The Repository Still Refuses To Claim

The roadmap is ambitious, but the repository still refuses final release language. Animal recovery depth and SEAD comparability remain materially weaker than the stronger architecture, geography, and docs surfaces. The roadmap is useful only if that refusal stays explicit while the weaker dimensions catch up.

The Boundary

This roadmap matters only while it prevents the repository from claiming that future engine work already exists in the current publication surface.