nx_research_ocean.nx

nx_research_ocean.nx -- THE OCEAN. Renders /wiki/research.html from the `resbrief-` information plane, so every research river (/code/research, /product/research, and any area added later) flows into ONE surface. Operator 2026-08-01: "both should like rivers to the ocean go into nishifamily.com/wiki/research under sections for product and code and other research areas". The AREA LIST IS NOT HARDCODED: sections are the DISTINCT values of the area column, in first-seen order, and a brief's URL is derived as /<area>/<slug>. A new research area therefore appears on this page by adding ONE plane row -- no page surgery, no organ edit. That is the difference between this and nx_wiki_products_page, whose families are string literals while its header claims a file source it never reads (debt 1785625149). Reads with sts_load_honest, NOT sts_load: a stale/low q:n silently truncates the plain loader, and an index that under-reports its own library is indistinguishable from a complete one. Declared vs loaded vs reachable-beyond are printed on stdout AND rendered on the page, so partial coverage is visible to a reader of the page, not just to whoever ran the organ. Plane row (8 col): id area slug title date status forks_off scope license_tier: ORIGINAL No hw writes (Rule 26). expect_exit: 0