Hacker News Reader

Comment by JoeAltmaier | Hacker News Reader

Reading the article, its clear that won't ever happen. The data sources are expensive.

I've been thinking recently about crowdsourcing our own dataset of airplane data. When you get on a plane, tap a button that says 'we're delayed 5 mins'. When you land, tap a button that says you've landed. In fact, you probably don't even need users to tap those buttons. The better way would be to use the accelerometer and barometer to automatically detect takeoff and landing.

Airplane data sources are ALWAYS wrong in my experience. I have been on a flight that was 2 hours late but all the flight data sources/apps would say 'On time'. Bullshit. And very often there are delays that the flight data does not account for, but are real delays. User's phones could provide a much more real-time view on where airplanes actually are than any existing flight data source. I hate sitting on a plane and knowing that I am more up-to-date on how late we are than the official data source is. That's super annoying and it seems like there's an 'easy' fix: just build a new kind of crowdsourced, real-time flight data service.

Users will complain loudly about battery life. Using the GPS, accelerometer, barometer isn't free.

It should be possible to piggyback on the user texting or performing some other common action after landing. But probably isn't easy.

> use the accelerometer and barometer to automatically detect takeoff and landing.

Or check location with a given interval and see when they no longer travel at mach 1?

Would that even be possible? I know I've never been able to get GPS to work reliably on any phone I've owned while on an airplane (even holding it up to the window), and presumably cell tower location wouldn't be available either above a certain altitude.

Edit: I am utterly wrong about my first point, the export controls don't kick in til 1,200mph or 59,000 feet; Both much faster and higher than standard commercial jetliners. [1]

GPS on your phone can't, by US law, work when you're in a plane. It is required that consumer GPS devices stop sending data when it is traveling over a certain speed, to prevent them from being repurposed as guidance systems for cruise missiles. Seriously.

You don't really need it, though. If you've got two or three people on the plane, the signal from when they all turn off airplane mode is a fairly good landing indicator. Takeoff is less steady; You're theoretically supposed to turn off your phone when they close the cabin door, but most people I know don't until they are actually taxiing down the jetway.

[1]https://en.wikipedia.org/wiki/CoCom#Legacy

Out of curiosity, do the GPS chips stop working above a given speed or it's the smartphone sw that must stop reading from them to pass import controls? And what about GPS chips manufactured outside the US, put in phones sold outside the US by non american companies (I'd say the vast majority of the market), are they still subject to that limitation?

I dont really know how it works, but its possible to circumvent, atleast some non-profit guys managed it: http://copenhagensuborbitals.com/gps-without-limits/

Deleted comment

Waze realtime traffic data was entirely crowdsourced, and probably still is largely so.

This is somewhat impaired by "airplane mode".

I've seen enough cases of people creating crappy free "replacements" for premium apps or services, and people flocking to them, that I don't know if you'd actually need to fully replicate the data sources. Just find something "good enough" and people will come. Or pay for a little while, planning to "figure out monetization later" and then fail in six months or a year.

That just means that the free alternatives won't stick around, but you could still get a steady stream of people who see your prices, assume they can undercut you and still make a profit and take away a few months of revenue each before they run out of money and give up.

They're welcome to try, but you're correct.

Reply on news.ycombinator.com