Have Weatherflow considered adding a feature to the apps (and/or website) to allow users to flag incorrect data? I realise that they are investigating/testing the CL/AI system, but I wondered if such a feedback process would help even further.
I am thinking of the following types of scenario:
I’m sitting at home in blazing sunshine, when I get a Rain Started alert on my phone. I look out the window and see a bird sitting on top of the Sky. Once it flies off, I can select the range in the app where the false rain is showing, and choose “Report Bad Data”. Couple more options pop-up - Type: Rain, Cause: Bird, Verified: Yes. With those verified reports, it could help WF build up more fingerprints of bird events.
I’m away from home and get a Rain Start alert. Every forecast was for dry, but very windy weather. The event is relatively short, so again I select it, choose “Report Bad Data”, then Type: Rain, Cause: Wind, Verified: No.
In a situation like #2 where I am able to verify it, WF could then start to build fingerprints for the vibrations caused for my specific installation.
WF reports heavy rain, but a separate rain gauge is showing significantly less, again a report could be sumbmitted with Type: Rain, Cause: Over Reading, Verified: Yes.
You get the idea.
I realise such a system would be open to abuse, but whilst the install base is relatively small and us early adopters (I would like to think) are more likely to want good, robust readings so are less likely to submit false reports.
Awesome feedback and this is something the team is working toward in some fashion. Both examples you have provided hit home with me as this would allow more timely submissions and fine grain details.
