diff --git a/03-DESIGN/01-to-be/28-building-the-bus.md b/03-DESIGN/01-to-be/28-building-the-bus.md
index c6ffb33..1421c79 100644
--- a/03-DESIGN/01-to-be/28-building-the-bus.md
+++ b/03-DESIGN/01-to-be/28-building-the-bus.md
@@ -150,9 +150,55 @@ paper is wrong until there is a second mesh to find out.
       and names no broker module anywhere in its source. What remains is naming `nats` instead of
       the deprecated broker where a genesis module set is declared, which is scenario and installer
       configuration — carried with 1.6 rather than before it.
-- [ ] 1.7 **the composition, delivered** — the controller gathering its principals, composing the
+- [~] 1.7 **the composition, delivered** — the controller gathering its principals, composing the
       file, and asserting the streams and consumers on start.
 
+      **In**: the user list is derived from the mesh's records and the credentials are kept.
+
+      A bus user's bcrypt hash is now recorded, keyed by the username the file needs, and the
+      plaintext is returned exactly once. That state is new and the reason is worth stating: on the
+      bus the mesh runs on today an account is a management call — mint, hand over, seal to the
+      holder, keep nothing — and that works because the broker remembers. Here the users are one
+      file rewritten whenever any of it changes, so keeping nothing would mean **the first person's
+      access change silently blanking every module's password**.
+
+      **Permissions are not kept, only credentials.** Authority is derived from what each module
+      declares every time the file is written ([ADR 0043](../../02-DECISIONS/0043-a-module-broker-account-is-scoped-by-emits-and-consumes.md));
+      a stored permission list would be a second account of a user's authority, able to disagree
+      with the records it came from while both looked internally consistent.
+
+      The derivation refuses two things where they can still be named: two users with one name — the
+      server reads the file as one of them and which one depends on the order — and a module
+      assigned but absent from the catalogue, which would compose a user with no authority and fail
+      on its first publish with an authorisation error that says nothing about a missing manifest.
+      A user the mesh has minted no password for is *named* rather than dropped or written as a user
+      anybody is: an ordinary situation with an obvious remedy, and the caller decides whether a
+      partial file is worth writing. A seat's protocol is gathered across the whole catalogue, not
+      from one manifest, because a seat is declared by one module and held by another.
+
+      **Out, and what each needs.**
+
+      *Delivery.* Resolution already prepends `file` resources whose content comes from the
+      rendering — a certificate does exactly this, and refuses when a module asks for one and none
+      was issued. The composed configuration is the same shape, and the open question is what the
+      module *declares* in order to receive it, which is design 29's ground rather than this
+      document's: a field naming where it wants the file, or nothing at all because the module
+      holding `mesh-broker` is the one that gets it. The server's own values — ports, TLS paths,
+      store directory — are constants of the module's own resources today and would have to be read
+      from one place rather than two.
+
+      *Minting, and it is transport-coupled.* A password is minted at enrolment and at assignment,
+      and an enrolment reply carries exactly one. **A node on the old bus must not be handed a
+      credential for the new one**, so which bus a node is joining has to be a fact the controller
+      holds before it can mint for both — the same switch the host's `Transport` is, from the other
+      end.
+
+      *Asserting on start.* The streams, the controller's consumers and every node's are defined
+      and idempotent; nothing calls them from a start path yet.
+
+      **People are not in the list**, deliberately: the account model is built and `operator issue`
+      is not (4.4), so there is nobody to derive. Left empty rather than guessed at.
+
       > **This corrects a tick, not a decision.** Tasks 1.3 and 1.4 are ticked and they are honest
       > about what they built — the composer, the derivation, the permission model, the stream and
       > consumer definitions, the asserter, all pure and held by unit tests and a golden