Wayfinder: notes on boop — signed, publishable, verifiable

[brianorwhatever]

Destination

boop grows a notes surface where a note is a list — same lists table, same assetDid/listEnvelopes identity, same publish path — carrying prose instead of items. Cloud-native: notes live in Convex, not as files on disk.

Shipped means: a stranger lands on a published note and can confirm who wrote it and that it has not been altered since. Provenance is the product, not the editor.

Notes

Domain: boop — Convex (realtime DB + HTTP actions), React 19 + Vite + TS + Tailwind v4, Originals SDK for DIDs (did:webvh, did:cel) and VCs. Relevant existing machinery: lists (assetDid, vcProof), listEnvelopes (signed CEL event log), items, publications / sites / siteDidLogEntries (signed publishing), subscriptions + list cap, presence / comments / activities.

Skills every session should consult: /grilling and /domain-modeling by default; /prototype for the prototype tickets; /research for research tickets.

Standing preferences for this effort:

• Plan, don't do. These tickets resolve decisions; the build is a separate handoff.
• A note is a list. Reuse the primitive — do not introduce a parallel data model. If a ticket's answer requires a new primitive, that's a signal to re-examine the destination, not to quietly add one.
• Provenance is the wedge. When a decision trades editor polish against verifiability, verifiability wins.
• This audience is not Obsidian's local-first users. They were consciously given up when cloud-native was chosen; do not re-litigate it per ticket.

Decisions so far

None yet.

Not yet specified

• Editor choice, and whether markdown is the storage format. Hangs on where prose lives.
• Concurrent / multiplayer prose editing. presence and comments exist for lists, but concurrent prose is a different problem than concurrent item rows. Revisit once sealing granularity is settled.
• Import from an existing Obsidian vault. Whether a migration path exists at all, and what it costs, once the note shape is known.
• Mobile note-writing. A Capacitor app exists; unknown whether notes need it at ship.
• Discovery of published notes. SEO, indexing, whether published notes are listed anywhere.

Out of scope

• Local files / local-first sync. Cloud-native was chosen deliberately; notes live in Convex.
• Notes as a separate primitive — own table, own DID lifecycle, own editor stack.
• Graph view, backlinks, wiki-links. This is where "Obsidian replacement" becomes literally true and the surface becomes its own product.
• Plugin ecosystem.
• Winning Obsidian's local-first users. Different audience, consciously not pursued.