nx_research_multilang.nx

nx_research_multilang.nx -- ALL-LANGUAGE research coverage (operator: "handle all languages ... so we TRULY get to physics optimal"). The deepest reason multilingual matters: a HIGHER physics optimum can be published in ANY language -- a Russian physics preprint, a Chinese materials paper, a Japanese patent, a declassified non-English report. If the researcher only reads English (~half of indexed output, and a SMALLER fraction of frontier/grey/declassified work), it can MISS the true optimum and wrongly conclude it is at the limit. So physics-optimal-GUARANTEED requires covering ~100% of the global research frontier across languages. This measures the Nishi researcher's current language coverage vs the global research- output distribution, finds the biggest uncovered gaps, and emits the coverage research program that closes them -- the multilingual drive that lets the autonomous loop (nx_research_loop) truly reach physics-optimal. The Nishi search engine searches a query across all COVERED languages; the provability gate (try-to-reproduce) applies per source regardless of language. Live cross-language FETCH = the delegated step (the web browser); the coverage measurement + gap + drive + uniform per-language processing are sovereign. HONEST: language output-weights are REPRESENTATIVE (the SHAPE is real -- English-dominant indexed + a long multilingual tail incl under-indexed frontier work -- not exact scientometrics). No hw writes (Rule 26). expect_exit: 0 license_tier: ORIGINAL