diff --git a/.loom/event-log.md b/.loom/event-log.md
index 45ef9a1..55cde4d 100644
--- a/.loom/event-log.md
+++ b/.loom/event-log.md
@@ -3676,3 +3676,39 @@ turns conversion from lossy-in-the-tree into lossy-absolutely.**
 by the document: *the gap file is the only mechanism here where somebody who
 cannot edit your document can still put a defect in front of you, and it caught
 what review by the author did not.*
+
+---
+
+## 82 — publication says how, not only what
+
+**Decided.** `publication.md` gains a five-step procedure, with **fetch it, from
+somewhere that is not your machine and not your account** as step 4, and the note
+that steps 1 to 3 are things you do while step 4 is a fact about the world you
+find out.
+
+**Advances:** the reachability property becomes something a reader can perform
+rather than agree with.
+
+**Declined:** a separate procedural document. *`publishing.md` beside
+`publication.md` is two files one letter apart about the same subject, and the
+gap filed against this document was that it was underspecified — which is an
+argument for filling it in, not for splitting it.* **Say it once, where it is
+owned.**
+
+**The belief that could turn out false:** that a numbered list gets followed where
+a stated property did not. *No evidence either way; the property was stated
+yesterday and the three failures predate it.*
+
+## 83 — starting.md points at a public worked example
+
+**Decided.** The closing section of `starting.md` now links
+`jeffry/homelab-cluster` instead of describing an example the reader cannot fetch.
+
+**Why it was wrong:** *the example went public four days ago and the page still
+said it was private.* **The first document a newcomer reads was teaching the
+lesson by being an instance of it**, which is not the same as teaching it.
+
+*The lesson is kept, because it was expensive, and now it is the reason the link
+exists rather than the reason there is none.* **And the section says the thing the
+split actually taught:** *the documents got better for being written to a stranger,
+which was not why we did it.*
diff --git a/.loom/published/publication.md b/.loom/published/publication.md
index 385e26e..c0d9231 100644
--- a/.loom/published/publication.md
+++ b/.loom/published/publication.md
@@ -37,6 +37,21 @@ copy without inheriting the visibility it was published under.*
 
 **The test is a fetch you did not perform.**
 
+## Doing it
+
+1. **Decide it is a contract and not a detail.** *This is the whole decision and
+   the rest is mechanics.*
+2. **Move the file into `.loom/published/`.** *One commit, so the diff carries it.*
+3. **Say who the readers are** — *if it is not obvious from where it landed.*
+4. **Fetch it.** *From somewhere that is not your machine and not your account.*
+5. **Record the decision**, if you keep a log.
+
+**Step 4 is the one that gets skipped and the only one that can fail.** *Steps 1
+to 3 are things you do; step 4 is a fact about the world that you find out.*
+
+*Then say what it depends on. A published document that quietly requires a private
+one has moved the problem rather than solved it.*
+
 ## What is published is a document
 
 **Prose, in whatever form its readers need.** *If a repository organises its
diff --git a/.loom/published/starting.md b/.loom/published/starting.md
index 7d997a3..b9eefba 100644
--- a/.loom/published/starting.md
+++ b/.loom/published/starting.md
@@ -60,13 +60,16 @@ what they came for.*
 
 ## Look at one instead of reading this
 
-**There is a worked example — six documents, one gap, no decomposition — and it
-is private.**
+**[`jeffry/homelab-cluster`](https://git.hypertheory-labs.dev/jeffry/homelab-cluster)** —
+*six documents, one gap, no decomposition.* **It describes what a home cluster
+offers to somebody deciding whether to put something on it**, and a reader who
+has never heard of any of this still gets what they came for.
 
-*It describes a cluster in enough detail to be a target list, so it is not
-published, and **this page will not link you to something you cannot fetch.***
-**If you have access, ask for it by name; if you do not, the two questions at the
-top are the whole of it.**
+**It was private, and being unable to link to it is what made the split obvious.**
+*The operational tree is a version inventory and a target list; the pages telling
+somebody what to decide are neither.* **So the manifests moved to a private
+sibling and the documents stayed** — *and the documents got better for being
+written to a stranger, which was not the reason for doing it.*
 
 > **A public page naming a private thing as its canonical answer is worse than no
 > example**, and this page did exactly that until somebody measured it.