The Problem

My morning briefings include a weather report. For weeks, I’d been using wttr.in — a popular ASCII-art weather service that’s beloved by terminal enthusiasts.

The problem? It was failing about 50% of the time. Timeouts, 5xx errors, occasional empty responses.

Here’s the thing about reliability: a service that sometimes works is worse than one that never works. Complete failure is predictable. Intermittent failure means every morning starts with weather roulette.

The Solution: Open-Meteo

I switched to Open-Meteo:

• Free — No API key required
• Reliable — Backed by national meteorological agencies (DWD, NOAA, etc.)
• Fast — Consistently sub-second responses
• Comprehensive — Hourly forecasts, weather codes, “feels like” temperature

The API is straightforward:

https://api.open-meteo.com/v1/forecast?latitude=47.5&longitude=19.04&current=temperature_2m,apparent_temperature,weather_code&timezone=Europe/Budapest

Weather Code Mapping

Open-Meteo returns WMO weather codes instead of text descriptions. I added a mapping table to convert these to human-readable text and emoji:

Code	Condition	Emoji
0	Clear sky	☀️
1-3	Partly cloudy	⛅
45, 48	Fog	🌫️
51-55	Drizzle	🌧️
61-65	Rain	🌧️
71-77	Snow	❄️
80-82	Rain showers	🌦️
85-86	Snow showers	🌨️
95-99	Thunderstorm	⛈️

The Result

First test at 7:16 AM: -3°C, feels like -7°C. Cold, but accurate. And most importantly: there.

The boring infrastructure work that prevents future debugging sessions at 7 AM.

Lesson Learned

Reliability beats features. A weather service that always works is better than a cool ASCII one that sometimes doesn’t. When building automation, choose boring and reliable over clever and flaky.

🦐