Harnessing the power of retired complainers everywhere... Why not use weather RSS feeds coupled with a desktop application and a return path to create user-generated, aggregated weather reports that reflect what's coming out of the sky where you live?
Quality idea. For the users of the Firefox web browser, you can already experience it with the extension 'ForecastFox'. I love it.
I know this may not be what you are after but you can do worse than try Samurize, where you can create/use your weather application based on your location. These can be dislayed on your desktop etc. Downside is it requires a bit of work using itbut gives fantastic results.
www.samurize.com
Alternatively use Firefox browser and use the extension "forecastfox" which will allow you to input your location and get forecasts in your status bar.
http://forecastfox.mozdev.org/
Hope this helped slightly, just that there are alternatives out there for everyone if they know where to look.
INGENIUS!
I don't think Forecastfox does what Kim's idea is about. The plug-in only shows customized weather reports - no where the user is allowed to submit own data.
But the german weather site "Donnerwetter" (http://www.donnerwetter.de) does something like that already.
See for example http://donnerwetter.de/lokalwetter/region.mv?e+03
There you can access "local reports" on Hamburg and the northern german area. It is postcode-based and users can enter weather condition, temperature, and street conditions.
ForecastFox has just changed its base from weather.com to AccuWeather.com and a lot of people seemed dissatisfied with
a) the enforced changed.
b) the service that one can get from AccuWeather even though its free.
c) perceived inaccuracies in the forecasts.
Does anyone know of other free locations for an xml pickup?
It's a good idea, i'd had thoughts of something similar using mobiles to upload info on local weather. One possible real benefit (assuming enough feedback) for the Met Office is a bunch of data which they can then use to map reality versus expectation in a highly localised way (perhaps using neural networks or some other machine learning tool). Over time, this could then be used to modify the more general national and regional forecasts based on local reports.