You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
id: design: land/parcel capability domains — per-parcel policy granularity without tokenized land #6015
Part of the worlds design program: #5955. Wave-2 follow-up from the prior-art research (Lane B item 47); parent topic issue: #5975. Research source: workspace/scratch/worlds-capabilities-brief-2026-11-04/prior-art-research.md.
Precedent
Second Life land remains the fullest implementation of virtual real estate with in-world permission granularity: parcel permissions, group-deeded land, per-parcel ban lists, per-parcel script permissions — land as a capability domain. LambdaMOO/MUSH used room ownership + quota instead of land markets. Decentraland-style tokenized land is a dead end for id's purposes: it financialized scarcity without producing better worlds; its governance/UX lessons are mostly negative.
Scope
#5989 surveys the land/ownership line; this follow-up designs the id adaptation under the self-hosting stance (#5975):
A parcel/region inside a world as a capability domain: its own script-permission policy, access/ban list, and asset rules, nested under the world's ceilings (a parcel can only narrow, never widen — attenuation, per the program's delegation semantics).
Ownership/delegation: parcel control as a delegable, revocable grant (writer-set/delegation substrate), not a token.
What a self-hoster's server enforces vs what is world-author policy.
Done looks like
A design note: parcel domain model, nesting rule, grant shape — or a recorded decision that worlds don't subdivide this way, with reasons.
Negative precedent is already settled in the body: Decentraland-style tokenized land is ruled out; do not re-open it in the design note.
Next action (named): a design-authoring pass on this issue producing the note in the body's shape (domain model, nesting/attenuation rule, grant shape, server-vs-author split), finalised after id: research: prior art — capability systems, OCAP, Spritely Goblins/OCapN, CRDT shared worlds #5989 lands. This is design work, not a small_fix — no fix packet was enqueued from this triage, because a small_fix packet for an unwritten design would be a packet defect. The issue stays open as the design's home.
Review: Codex reviewed 4 rounds on PR doc: land/parcel capability-domain design note (#6015) #7601; five P2 findings (parcel-aware hosting enforcement, Rescind semantics, owner-only membership management, bearer-secret sharing limit, affected-extent checks) were each fixed in the PR with a disposition reply on its thread; the final round on head c6647e8 completed with no findings.
Activity
developing-today commentedon Nov 5, 2026
CODE-1681
dezren39 commentedon Nov 5, 2026
🤖 Agent note · triaged — task 10381 (queue repo-health-code-pilot, lane code-f)
15 remaining items
doc: land/parcel capability-domain design note (#6015) (#7601)
developing-today commentedon Nov 8, 2026
🤖 Agent note · done