<?xml version="1.0" encoding="UTF-8"?><oembed><type>video</type><version>1.0</version><html>&lt;iframe src=&quot;https://www.loom.com/embed/df8155a2b15446e7b650574591b6a89c&quot; frameborder=&quot;0&quot; width=&quot;1662&quot; height=&quot;1246&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1246</height><width>1662</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1246</thumbnail_height><thumbnail_width>1662</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/df8155a2b15446e7b650574591b6a89c-517bea2cd0432309.gif</thumbnail_url><duration>352.809</duration><title>Zed - discourse-graph, comments Aaditya&apos;s paper — discoverSharedRelations.ts - 19 August 2026</title><description>This Loom describes the first step of importing relations by gathering all the nodes, relations, schemas, and triples needed for the eventual data transfer. It calculates numeric local IDs to RIDs for nodes and relations, checks which node RIDs are already present from prior imports, and filters relations based on access (published nodes) or prior imports. It distinguishes triple schemas and relation type schemas, then performs a complex loop to resolve an impedance mismatch between Obsidian relation pointers and relation types by matching triples to the actual node types. Finally, it ensures implicated node schemas are available and fetches author IDs for mapping before returning the prepared bulk data used in the next PR.</description></oembed>