4 ms·
Your Phone Could Make You Into a Thunderstorm Predicting Machine
- cryptoz 14y agoThis article is about pressureNET, which is my open source Android app that collects atmospheric pressure readings from sensors in phones and tablets. I'm happy to answer questions as always :). Edit: You can browse a visualization of our collected data from last year at http://pndv.cumulonimbus.ca http://pndv.cumulonimbus.ca to get an immediate overview of our quantity and quality of data.
- sudont 14y agoWhat do you think of adding iOS devices into the network once BLE sensors become more common? http://mosoro.com/labs-hardware-enviro.php http://mosoro.com/labs-hardware-enviro.php
- cryptoz 14y agoAbsolutely! I hadn't heard of Mosoro before; that Enviro product looks very, very cool. We actually plan on building an iOS client very soon anyway, in order to gain greater visibility (and also to learn how to build iOS apps). So far, there is little to do inside pressureNET if you don't have a pressure sensor, so we've held off so far.
- cameldrv 14y agoCan you tell us more about how you post-process the data? Given that you're looking for small differences from what the nearest ASOS is reporting, compensating for altitude and sensor calibration seems like a big challenge.
- cryptoz 14y agoAt the moment we are not doing any post-processing. Our current approach is to work with researchers at universities who specialize in the calibration of pressure sensors like ours. We're talking with Cliff Mass at the University of Washington, and as soon as we put out a new update to pressureNET with privacy controls for our users, we will be streaming our data live to Cliff (and other researchers who are interested). This is our primary short-term goal, as it's the major hurdle we must overcome before pressureNET is a truly useful product.
- aw3c2 14y agoplease please if you give users the control, please let them also allow completely free and open sharing. that way random interested hackers can play with the data for random weekend projects and the like.
- cryptoz 14y agoDefinitely! This project has the potentially to be immensely beneficial to humanity, but only if it's open. So our code is open. But so far we've struggled to make the data open out of privacy concerns, so we're not totally open yet. But open sharing of data is absolutely critical. So yes, it'll be done, and done ASAP as well. We're hoping to have that update out next week.
- kposehn 14y agoCompletely awesome. Sent you an email - we have some tools that might help :)
- miahi 14y agoI had your application installed for a while on my Galaxy Note, but it drained the battery so I had to uninstall it. Just downloaded it again to check if it still has issues.
- cryptoz 14y agoWe had battery issues until December, and everything should be fixed now. Thanks for downloading again and I hope it's better this time.
- preinheimer 14y agoA friend had a barometer on his watch in Highschool. We were out at the Ex (large carnival/exposition in Toronto) and he watched the pressure plummet over the course of an hour. We quickly made our way to the exit, and hopped on the last train leaving before the downpour started. It was a moderately silly looking watch, but I never mocked it again.
- miahi 14y agoI had one of these for years. It doesn't work every time (not every time the pressure drops it will start to rain) but it helps you prepare. I also have a barometer on the Galaxy Note phone. I tried some barometer applications, but I could not find one that doesn't drain the battery even when they are not showing data, so I stopped using them. The idea is great, but the implementation of the applications and phone not so much - especially when you compare it with a solar powered watch.
- pavel_lishin 14y agoEvery time an exterior door in our office building opens, I can feel a strong gust. Presumably this means that the air pressure inside a building is higher than air pressure outside, right? How would this affect the data collected?
- cryptoz 14y agoThe pressure indoors is usually very similar to the pressure outdoors, though there are cases where it's different. A much bigger related issue for us is altitude. The pressure by your feet is measurably different than the pressure by your head due to the altitude difference. So you can imagine it's a lot of noise as people move around and go upstairs, hills, etc. We're working with researchers who are writing papers on the calibration of networks of pressure sensor data (Cliff Mass' lab, specifically) in order to help us solve this problem.