Why we built COMPASS ID
The world’s research is searchable by keyword, but not by what it covers or leaves uncovered. We built COMPASS to make the shape of knowledge — and its blank spaces — measurable.
Search engines answer the question you already know to ask. They can’t tell you what an entire field has no indexed research on.
The problem: knowledge has a shape, and we couldn’t see it
Billions are spent every year synthesizing research for policy and funding decisions. Yet the basic questions — which species, regions, and problems are well-evidenced, and which are dark — have no measured answer. Keyword search returns what exists; it is silent on what’s missing.
The thesis: measure the whole field, then measure the gaps
COMPASS takes the opposite approach to a filter. Instead of ranking a few “important” papers, it ingests the complete literature of a field and tags all of it. Completeness is the moat: only a corpus that leaves nothing out can tell you, honestly, what has been left out.
We hold ourselves to one rule above all others — measure, don’t claim. Every gap we report is backed by a tagged, auditable corpus, and every accuracy figure is measured on gold data.
What it revealed
The first atlases surfaced a pattern that keeps recurring: research is often anti-correlated with need. The rarest species are the least studied; the most degraded drylands attract the least fieldwork. That misallocation is invisible to keyword search — and visible, and quantifiable, in COMPASS.
