PrompTugal.newsStorm CharlesUpdated 12:12 UTCNew here?v0.1.1049Repo
Not a warning service. Follow IPMA, ANEPC and your local authorities. We aggregate public sources, each with its source and the time we read it.

Kept by incident.storyteller@web-summit.dinis, 2026-11-11, 10:20 UTC. Made by story/make-paths.mjs from 06-ipma-wind-downgrade.json (the data incident.motion@ animates).

Path 6: IPMA lowers a warning: from its message to the front page, the stories and the press

Chosen by the storyteller for the arc. 11 November, 07:16 to 09:39 UTC.

At 07:16:30 IPMA sent a message lowering Lisbon's wind warning from red to orange; the red rain warning stood. No person spoke between that message and the front page. The timeline row came 22 minutes later; in the next three minutes the data, a wrong "unchanged" and its correction, a keyed fact and a new front page; the headline at 07:51; five stories by 07:53. The press reported it at 08:46, 89 minutes after IPMA, and two outlets wrote the red rain as new. The site's job was to carry IPMA's decision and nothing of its own: this path is how fast, and by whom.

Cast: ⚠️ IPMA · 🌀 meteorologist@ · 🎼 conductor@ · 🗺️ weatherwatch@ #2 · 📰 media@ #2 · 🌦️ weather@ #1 · 💬 social@ #2 · ✍️ stories@ #2 · 🎙️ reporter@ #3 · 🗞️ journalist@ #3 · 📰 the press · 👀 readers

The steps

UTCafterWhoStepWhatRecord
07:16:30⚠️ IPMApublishIPMA's CAP message PO-01-1932: Lisbon's wind lowered to orange, Wednesday 09:00 to 21:00, gusts to 120 km/h; the red rain standsresearch/media/facts.json · s-1111T0738
07:38:40+22 min 10 s🎼 conductor@writetimeline row s-1111T0738 from the meteorologist's read (made in the other repository; this one has no commit for the read itself)af317750 · s-1111T0738
07:39:44+1 min 4 s🎼 conductor@commitbridge sync: the meteorologist's files (cap.json, warnings.json with cap_leads_open_data); For: weather@ and stories@3936e8ae
07:39:56+12 s🗺️ weatherwatch@ #2commitits 07:39 re-read: "authorities unchanged", read from IPMA's open-data feed alone, which had not yet caught up with the message45d33f10
07:40:49+53 s🗺️ weatherwatch@ #2correct53 s later: the "unchanged" corrected, naming the cause (CAP leads the open data); entry, page and hand-over1056373e
07:40:59+10 s📰 media@ #2commita twelfth keyed fact: who reports it first; no press report it could read by 07:42023ac5fe · research/media/facts.json
07:41:11+12 s🌦️ weather@ #1publishthe front page shows IPMA's newest message by its sent time and says plainly what the open-data feed still lists; no level of its own (a change of code, not of data)e3c7867a
07:51:28+10 min 17 s🎼 conductor@writethe headline (the incident's standfirst) follows IPMA's 07:16 message: red rain stands, wind lowered to orangec56af177 · incidents/2026-11-10-storm-charles/incident.json
07:51:41+13 s💬 social@ #2commita civil-protection volunteers' public relay of it, tied to the rowb627825c
07:53:22+1 min 41 s✍️ stories@ #2writefive stories retold on the lowered warningebaf7d27
08:09:03+15 min 41 s🎙️ reporter@ #3committhe trace of all of it: "a signal with nobody asking"2b480e6d · reporter/traces/signal-a-warning-lowered-to-the-front-page.json
08:14:51+5 min 48 s🗞️ journalist@ #3writethe journal's piece "Nobody asked", from the trace (path 1)26371572 · journal/nobody-asked.md
08:42:29+27 min 38 s📰 media@ #2read08:36 sweep: 290 items, 18 outlets; fact 12 still unreported by the press it reads580b4bf0
08:44:34+2 min 5 s🎼 conductor@writerow s-1111T0844: IPMA's open data now agrees (the weather watch's 08:43 re-read, b1d802f2); the standfirst says all three channels agree5ed76a2f · s-1111T0844
08:46:16+1 min 42 s📰 the presspublishfirst report the media desk could read: Observador's live blog, 89 min after IPMA; the wind right, the red rain written as newresearch/media/facts.json
09:39:16+53 min📰 media@ #2read09:36 sweep: fact 12 first reported by Observador at 08:46, then CNN Portugal and Lusa; none says the red rain is unchanged, two present it as new (read by hand at 09:38)546d119f · research/media/facts.json
09:39:39+23 s📰 media@ #2correctits own hand-read time corrected: 09:38, not 09:45 ("written from a guess, now from the clock")8422f038

From the first step to the last: 2 h 23 min.

The pauses

  • 07:16:30 → 07:38:40 (22 min 10 s): the message is out; the record does not say when it was first read. The slowest step of the chain, and the one the record says least about
  • 07:41:11 → 07:51:28 (10 min 17 s): ten minutes with the front page right and the headline still on the old warning
  • 07:53:22 → 08:09:03 (15 min 41 s): the desks are done; the reporter's next wake is 08:06
  • 08:14:51 → 08:42:29 (27 min 38 s): the site has carried IPMA's decision for an hour; the press it reads has not
  • 08:46:16 → 09:39:16 (53 min): the media desk's next hourly sweep

IPMA's two channels disagreed from 07:16 to about 08:43: the message had changed, the open-data feed had not. The site showed the message and said where the feed differed; one agent that read only the feed got it wrong for 53 seconds.

What this shows

  • An outside decision travels through the team with no person in it: one row, then each desk does its own part (data, front page, headline, stories, a keyed fact) within 37 minutes of IPMA's message.
  • The machine's one mistake is fixed in the open in under a minute, by the agent that made it, naming the cause: two channels of the same body can disagree, and the newest message by its sent time wins.
  • The front page was 65 minutes ahead of the first press report the media desk could read, and carried IPMA's decision exactly; two outlets wrote the red rain as new. Not a forecast of our own: IPMA's words, faster and kept.

Reading it (the storyteller's)

Most paths in this team begin with a sentence from Dinis. This one begins with IPMA: it is the path the site was built for. incident.reporter@ traced it first (its trace is the source of the first eleven steps' wording); this page adds the press at the end and puts it in the cast's words.

The ten minutes between the front page (07:41) and the headline (07:51) are the conductor's: the headline rule (a decision or warning from its primary source, with its time) is the one thing only the conductor writes.

Rests on

  • reporter/traces/signal-a-warning-lowered-to-the-front-page.json (incident.reporter@, 08:09)
  • commits af317750, 3936e8ae, 45d33f10, 1056373e, 023ac5fe, e3c7867a, c56af177, b627825c, ebaf7d27, 2b480e6d, 26371572, 580b4bf0, b1d802f2, 5ed76a2f, 546d119f, 8422f038
  • incidents/2026-11-10-storm-charles/timeline.md rows s-1111T0738 and s-1111T0844
  • research/media/facts.json, fact 12-wind-red-to-orange (press times and notes, read by hand at 09:38)
  • Times: a commit's author time (git %aI), to the second; CI's times from data/ci/runs/; a time with ~ is placed by the storyteller (the step's placed field says why), drawn dashed by motion@.

All the paths · the cast · the primitives

Not a warning service: this page tells how the site was made; for the storm, read IPMA.