I thought of this immediately when watching the keynote, but then they mentioned they used GPS technology, and I figured they would be passively monitoring to confirm that for the past X minutes the phone has likely been in a car (non-walking speed + on roads).
Clearly they aren’t doing enough of that if you can trigger this after waiting in line in a roller coaster park. I suspect they may end up geo-fencing known problematic areas as their algorithm doesn’t seem as great as they claim.
Or a simple debounce mechanism that detects continuous falls, maybe training on momentum data of a fall vs continuous fall actions like slamming hammer etc.
A friend and I are wagering on which would trigger first: Apple's Crash Detect, or Ford SYNC's 911 Assist feature. SYNC 911 Assist triggers the millisecond that the RCM (restraint control module) signals the airbag deployment and is "about 10 seconds". We THINK that the car might beat iOS itself to command a 911 call, but we're not sure.
These things are quite sensitive and their heuristics are sometimes wrong. I've been driving at the track for quite a few years, and now and then, you see a car deploy airbags for no good reason when cornering really hard or hitting a berm or cresting a hill. OnStar calling 911 is pretty common, because it decides to do that based on driving style, and on the track, it's pretty near the limit.
I don't know how many seconds passed (of course it felt like a half of eternity), but by the time I re-learned how to breathe the car was already calling the emergency services.
I wonder if new Apple Watches have the same problem.
The iPhone wouldn’t be a problem for me, since I wouldn’t take it on a rollercoaster anyway, but I’d probably forget to take my watch off before every ride.
Off topic but something I find quite neat is that your watch syncs wifi networks with your phone.
I was at a friends house the other day and had forgotten my phone, but was still getting texts/notifications come through on the watch because I had previously connected to the wifi network on my phone (which was sitting in my house).
Lot of highly sophisticated software thinkers here are doing the equivalent of a 'meh' to something the 'richest, most sophisticated, most caring, tech company in the world' has proclaimed. I did a 'meh' at the presentation of crash detect and I am not of that software programmer ilk. I do think it is important to remember you are being sold something, even when it emanates from ones own camp. I work with sound so, as you might imagine, have had decades of practice.
I am very curious if they did assessment of other situations which could trigger similar sensor reactions that they had to rule out. Theme parks could actually be geofenced off without not that much work - there are under 1000 roller coasters in the US.
Did they have to consider people dropping phones out of car windows (or off a boat ?) What about ski crashes or skateboard crashes?
There are traveling carnivals to things like state/county fairs, as well as just big empty parking lots of malls/etc for a few weeks, with significant g-force rides. So well over a thousand locations in the US.
> Theme parks could actually be geofenced off without not that much work - there are under 1000 roller coasters in the US.
I'm cognizant of the 80/20 rule, and "perfect is the enemy of good," but if a run-of-the-mill coaster is setting off that detection, then I'd wager the traveling fair rides will be even more violent, based on my experiences growing up
I have no idea where this falls on the privacy spectrum, but I'd guess if there are 5 or 10 triggering events within some timespan and some geo boundary, that's an indicator of a themepark-ish setup
Clicking "report false positive" on the phone would also likely go a long way
To be fair it didn’t fall off a motorcycle. It was thrown (due to a hard bump) from a moving motorcycle. If it had been in the person’s pocket at the time because they had fallen off too, calling 911 would have been very fitting.
I guess the question to me is what the final false positive rate on this feature is. Its obviously a huge net win if the proportion is small - but if it is high enough I'd imagine that dispatchers would be less likely to want to send officers to the automated calls.
Let me try playing devil's advocate (with zero inside info). Isn't it possible that more lives are saved such that it's worth the false alarms?
It is obvious to me that these decisions will become more and more difficult. I am certain that at some point (not sure if 10 years from now or 50) driverless tech will be safer than people--by a margin of say 99%. I am equally certain that this will be accompanied by a set of accidents unique to the driverless tech. I think of it as similar to compilers. When I started programming PCs assembly was always king. C got better and better. 10 years into my career C was mostly better. Now C/C++/Rust etc. almost never need code rewritten in ASM.
> "Crash Detection is on by default." [iPhone use guide]
You're kidding? After 20 seconds? Ridiculous. Most people would be getting out of the car to inspect damage and any injuries, not attending to their iPhone's urgent needs.
And if you do need to make a call, the emergency services would appreciate more information that what a recorded message blurts out. So most would call a second time after the initial time-waster bot call. I'm amazed this is allowed.
When the phone is counting down the last 10s, it’s playing a loud alarm tone. No one is going to be surprised it’s up to something, it’s actively drawing attention.
Also the fact it calls is incredibly useful if the driver can’t. They could be pinned, unconscious, or something else. But the phone has no way of knowing that. So either it calls and tries to help people (what they do), or the phone could do nothing leaving those who couldn’t call themselves helpless.
If you're busy clearing debris from the road and talking to other drivers about your fender bender, your phone is likely back in the car out of ear shot.
I'm not sure why you think questioning the default-on status of this feature is somehow an opposition to its potential usefulness in the extremely rare times you're unconscious or pinned AND nobody else is around to call. I mean, this article is about the phone calling emergency services from roller-coaster rides, which says it all. It should default to OFF. You have purchased an i Phone not an i Emergency Beacon.
I’m up there with the most libertarian people on phone alert annoyances and I think this is a non-issue. If you’re able to get out of the car, you’re taking the phone with you. You need to take pictures for insurance purposes — and of the other driver’s license (if there are others involved). If there are other people involved, you may need to call emergency services for them. If you’re too dazed to remember the phone, it probably _should_ notify emergency services.
In your view, how many false, unnecessary auto-calls from iOS to emergency are acceptable per year nationally, 50? 1000? 10000?
It's common to exit your car instinctively after an accident, then go back for phone. Many people keep their phones in their bag while driving.
After the initial call-triggering event, the hazard may not be over. You may need to move your car off the road, or any number of other safety-related post-accident actions.
There are more important things to take care of in the 20 seconds following an accident than tending to your phone apps.
What next? "child coughing detected, please take your medicine in 10,9,8...".
