From 8054d064e861f8bc5dac9ce23df815557abdc3a6 Mon Sep 17 00:00:00 2001 From: Jean-Pierre Martineau Date: Sun, 19 Jul 2026 21:49:29 -0400 Subject: [PATCH] Atlas: document that "N hops out" is the ring, not the walked trail length The focus line's hop distance is the shortest observed path from us to the node (a BFS over the evidence graph), which legitimately reads shorter than the breadcrumb trail when the walk wandered or doubled back. Record why the two can differ so it isn't mistaken for a miscount. Co-Authored-By: Claude Opus 4.8 --- meshterm/ui/atlas_screen.py | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/meshterm/ui/atlas_screen.py b/meshterm/ui/atlas_screen.py index 3c85954..c83634e 100644 --- a/meshterm/ui/atlas_screen.py +++ b/meshterm/ui/atlas_screen.py @@ -487,6 +487,15 @@ class AtlasScreen(Screen): The leading glyph already carries the node type, so the line no longer spells it out; the name reads ``name (hash)`` — the addressed path-hash in parentheses, its digits lit in the node's hue — rather than a bare slice of the key. + + The ``N hops out`` distance is the *shortest* observed path from us to this node + (a BFS over the whole evidence graph, :meth:`_hops_out`) — the node's ring in the + mesh. It is deliberately **not** the length of the breadcrumb trail above, which is + the route you happened to *walk* to reach the focus: wander out a long way and + double back, or step to a node also reachable by a shorter link, and the walk is + longer than the ring. The two answer different questions — "how near is this node?" + versus "how did I get here?" — so a shorter "hops out" than the trail is correct, + not a miscount. """ node = self._focus glyph, color = self._glyph(node)