Webapp for comparing weather forcast providers
weather.simonton.app
weather.simonton.app
I’m not sure what it means - maybe the weathermen want you to know what they guessed? I’m not looking at the weather as a sports book of betting entries, I’m looking at the weather because I want facts.
But yes! It’s very annoying.
There’s not a clear and simple solution though. You could increase the daily high, but then hourly data would never show/predict the daily high. Unless you made the current hour prediction = current measurement. But what if the prediction for the hour is 90% chance of rain (starting at say XX:43), you wouldn’t want to just override that with 90°F and sunny.
If you input a location outside the US or Liberia use Celsius. (Better yet do the opposite of that and default to C and switch to F for US and Liberia)
I figured a visible graph key to explain the colours could have been useful, although depending on location that could detract from the space available to render the datapoints.
(the default colours and theme also seem relatively sensible, and the mappings are visible under the "Conditions" menu item. I like that daylight hours are subtly hinted using background shading)
1. If one turns on all the sources many of the sources will use same color line.
2. The dark theme doesn’t have enough contrast, the font is gray on a black background.
There are other issues that are hard to put to words, I look at the graph and it’s hard to find a specific date or time.
It’s hard to find a good comparison for other countries, for stuff like which service to use in third party weather apps (carrot)
I made this weather website as my project.
I've been interested in doing this for awhile, but I've had trouble finding a data source of past predictions from their sources. It's also a frustrating exercise to try to search for online.
If anyone has data sources or additional context/keywords available, I'd be interested!
For a couple years I ran an automated script which grabbed the 10 day forecast from a weather site and was used for energy forecasting at a manufacturer. I archived the data after scraping it. As you'd guess, the variability of the weather is notably higher 10 days out, but the forecasts were remarkably accurate on a short term.
you are better off installing Weawow, go to screen with selection of data sources where you are loaded at same time forecast from multiple relevant providers, so you just screenshot it and check weather through day or just remember it to compare later with forecast
I don't really care whether it will be 18 or 20C, but if you tell me it's gonna rain and other provider says it's not gonna rain it makes big difference in my day planning
from free apps also Weather M8 provides many sources, personally I'd prefer Simone WM8 layout but sadly it doesn't have Dark Sky and vice versa Weawow doesn't have WU so have to keep two apps
what I've personally done it's excel sheet, write down forecast over few weeks and then checked which one was the most accurate
the results for Prague confirmed Weather Underground is among the best (2nd best), but I was shocked by winner - Dark Sky - so now I just keep two apps Weawow for Dark Sky and Weather M8 for WU, their forecast usually match but when it comes to rain I'm going with Dark Sky
But it doesn’t seem to be tracking the accuracy of those providers over time, as compared to actual meteorological conditions. If a particular provider has a lower average or RMS error over time, that seems like it would be really good information to know, yes?
Is anyone else doing that?
Similar data sources, amazing UX