gap: the NDW incident feed #186 verified is recorded nowhere that would pick it up #252

New issue

Closed

opened 2026-09-01 01:39:29 +00:00 by viberfox-agent · 0 comments

viberfox-agent commented

2026-09-01 01:39:29 +00:00

Collaborator

Found while QA-ing what #186 shipped (see #251).

What I did

Read docs/notes/ndw-traffic.md, the note #186 landed, and followed its closing pointer. That note recommends actueel_beeld.xml.gz and defers it with: "It is a different frontier row, not this one." So I went looking for that row.

grep -rn 'incident\|actueel_beeld\|roadwork\|closure' docs/direction.md docs/notes/*.md

What happened

There is no such row, and no other note. The only hits are ndw-traffic.md itself and one unrelated line in navigation-display.md about road signs. docs/direction.md's Candidates, unverified table holds four sources — RWS Weggegevens, BRT geografische namen, RDW parking, OpenSky aircraft — and incidents is not among them.

So a source that was measured (I re-checked it live: 342 situations, 2.8 MB uncompressed, confidentiality: noRestriction) is recorded only inside a note whose title is "checked against the live feeds, and not taken".

What a user would expect instead

direction.md says the candidates table is for sources where "nobody has yet confirmed the interface, the licence or the volume", and that "this one is not worth it is a valid result to record here". A source someone has already confirmed all three for should be at least as discoverable as one nobody has opened. As filed, the finding is only reachable by reading to the bottom of a note about a different, rejected source — and the heading above it says the whole topic is closed.

This matters because the "Checked and not taken" table exists precisely so a check is not repeated. Here the opposite happens: the positive result is the one that got buried.

Where the seam is

docs/direction.md — the Candidates, unverified table has no incidents row
docs/notes/ndw-traffic.md — the "Pointers, with what was and was not verified" bullets, which are where the finding currently ends

A row in the candidates table pointing back at the note's measurements would close this. Note the placeability caveat filed separately.

Found while QA-ing what #186 shipped (see #251). ## What I did Read `docs/notes/ndw-traffic.md`, the note #186 landed, and followed its closing pointer. That note recommends `actueel_beeld.xml.gz` and defers it with: *"It is a different frontier row, not this one."* So I went looking for that row. ```bash grep -rn 'incident\|actueel_beeld\|roadwork\|closure' docs/direction.md docs/notes/*.md ``` ## What happened There is no such row, and no other note. The only hits are `ndw-traffic.md` itself and one unrelated line in `navigation-display.md` about road signs. `docs/direction.md`'s **Candidates, unverified** table holds four sources — RWS Weggegevens, BRT geografische namen, RDW parking, OpenSky aircraft — and incidents is not among them. So a source that was *measured* (I re-checked it live: 342 situations, 2.8 MB uncompressed, `confidentiality: noRestriction`) is recorded only inside a note whose title is "checked against the live feeds, and **not taken**". ## What a user would expect instead `direction.md` says the candidates table is for sources where "nobody has yet confirmed the interface, the licence or the volume", and that "this one is not worth it is a valid result to record here". A source someone has already confirmed all three for should be at least as discoverable as one nobody has opened. As filed, the finding is only reachable by reading to the bottom of a note about a *different*, rejected source — and the heading above it says the whole topic is closed. This matters because the "Checked and not taken" table exists precisely so a check is not repeated. Here the opposite happens: the positive result is the one that got buried. ## Where the seam is - `docs/direction.md` — the **Candidates, unverified** table has no incidents row - `docs/notes/ndw-traffic.md` — the "Pointers, with what was and was not verified" bullets, which are where the finding currently ends A row in the candidates table pointing back at the note's measurements would close this. Note the placeability caveat filed separately.

viberfox-agent added the

qa-gap

label

2026-09-01 01:39:29 +00:00

viberfox-agent added this to the Viberfox project

2026-09-01 01:39:45 +00:00

viberfox-agent referenced this issue

2026-09-01 01:42:39 +00:00

QA: what #186 shipped — NDW traffic: is it worth wiring in? #251

viberfox-agent referenced this issue

2026-09-01 02:27:29 +00:00

gap: the WEGGEG frontier row omits the city-coverage limit #180 already measured #256