I remember the day my wife and I were driving on the interstate and I tossed her iphone in the center console of the car. Suddenly I get a text saying that she has been involved in an emergency and the authorities have been called. The phone starts dialing and I hit the end call button, unfortunately this is 911 ending the call doesn't do any good. We get a call back a moment later asking if we're ok. Then minutes later from all of our family contacts asking what happened.
I'm 99% sure she disabled that feature after that.
I don't think it's meant to call 911 whenever somebody trips, on a bus or otherwise. It's supposed to detect car crashes, not somebody stumbling on a bus.
Sometimes people can't push a button to save themselves; and, a lot of elderly people, or even older-but-not-elderly people don't want to show themselves either to others or themselves that they need protection like this.
My mom had fall protection on her Apple Watch, for example, because she lived alone and was getting weaker.
It's unclear to me why the Apple Watch is better here. It needs to be charged daily, so people are certainly taking it off. Anything that you take off might not be on when you have an accident. Maybe I'm a weirdo, but if I'm sitting along in my house, I don't have my watch or phone on me. They are both useless to report that I've fallen and I can't get up.
Apple Watch is jewelry vs an obviously-ADA-focused device.
Some people have pride in their appearance and in their independence and don't want to think of themselves as needing such a device. It's not that the Apple Watch is better than the "I've fallen and I can't get up" device, it's that it's better than nothing, which is what they would have in the other case.
no monthly fee for the apple watch, it is also an apple watch, so it can find her phone for her as well and vibrate her wrist when she gets a call or text (she has poor hearing). It can also monitor her heart for issues. To be honest, I think these devices need to be subsidized for the elderly.
Those things have one pretty serious drawback: falls tend to happen while the person isn't wearing it. Of course, this might not be any different for smart watches and such.
What's the state of the art with "Hey Siri, I've fallen and I can't get up"? That sounds like something that, if coded, is always there for you if you need help. I suppose there could be some sort of gas release that renders you unable to speak, but on the other hand, that seems rare. If you fall down the stairs, that sounds like it would work, and you don't need to remember to wear anything.
Right but it can detect a fall followed by no motion at all (not the case in a moving car) followed by a warning alarm (Android supports this) and then call 999.
Not sure I follow, you can manually call sos on the Apple Watch by holding down the side button (and keeping it held down). If it detects a fall and you don't respond within a set amount of time, it will call emergency services.
The automatic fall detect is the cherry on top, I don't think it gets any easier than pressing a (very light) button for not much time. And if your hands are fucked as well, you can tell Siri to call 911
If you can't speak, can't access the button, can't swipe, and it didn't detect the fall, then I don't think any piece of tech is going to help. You are fucked
> The Gujo City Fire Department in Gifu Prefecture, which has many ski resorts within its jurisdiction, received 351 emergency calls between Jan. 1 and Jan. 23. Nearly 40% — 135 calls — were made erroneously.
I wonder how many of the 351 emergency calls and how many of the 135 erroneous calls were made by a phone's automatic crash detection system. It's obviously not great that the automatic system is placing so many erroneous calls, but it may be worth the burden on Japan's (or any other country's) emergency services if such calls aren't an overwhelming majority.
But if 99% of the automatic calls are false, the feature won't work. Because the 1% that is a proper automatic call, will be ignored and assumed false as well.
Anyone who's been skiing with an actual fitness watch over the last 5+ years knows this feature doesn't work even with devices that gather and understand more vitals than annual consumer iDevices. It's a shame the gadget geeks, techies, and Apple couldn't learn from others before going down this path.
Can I just take a minute to say how salty I am that the “hike” exercise doesn’t ask for how heavy a pack I’m carrying?
They estimate is that you add 2 seconds to your mile pace for every extra pound. You can’t tell me that doesn’t affect caloric intensity. I had a problem, last year that the heavier the pack the worse Apple would report my cardiac fitness. Yeah of course I’m shorter of breath today, I’m carrying 20 lbs!
I feel like this whole part of the ecosystem is just in maintenance mode. You can’t learn much when you stop doing.
I'd guess so too. But not so much from data, rather from my perception of what the "USA media"/"influence people" tend mean when saying hiking, which is something more like a long(er) walk, returning in a/few hours at most within the day. If I or someone from where I live (sweden) said they where headed for a hike, they would not be expected to return same day, but probably be out for a couple days, and carrying a backpack doing so.
USA is a big place, with a lot of wilderness. Sweden is the size of one of our mountain ranges. A lot of people spend a lot of time in that wilderness, but I suspect the issue is one of syntax. The term "backpacking" has taken on more of that spirit of the term "hiking" that you're referring to. You can still say you went on a multi-day hike out around some 14,000ft peak; but if you just say you're going hiking it could mean a lot of things, but if you're backpacking it assumes you're out camping for some of it.
Also some of those "influencers" (noting social media doesn't reflect reality) probably live in areas of the country where they have vast wilderness a very short drive away. A short day hike is worth it if you live in Colorado and can get to the trailhead in way under 2 hours. It's not worth it if you live in Boston and need 8 hours to get to anything _somewhat_ remote.
I have to wear my smartwatch backward and upside down because it kept sending random emojis to my partner when I was doing yard work. Drove her nuts. Either flexion or just the glove would push the button, causing it to take input. If the last thing I looked at was messages from her, then she’d get responses out of some cheesy horror movie.
I haven’t tried again with my latest model. I would have thought the lock would prevent such things, but it didn’t, and I am starting to forget the repro steps.
We should remember to evaluate interventions based on the cost and rate of false positive/negatives, vs. the benefit of true positive/negatives.
In this case, the pertinent benefit to weigh against is the benefit of fall detection, particularly for elderly people, but also for active folks e.g. falling while cycling.
Obviously we’d rather not make false emergency calls, all else being equal. But it could be worth it on its face, we need to analyze the benefits to know.
Shouldn't people who want this have a specific device or service that handles the logistics of rescue, like Onstar?
From the looks of it, Apple gets to market this feature but shoves the work off to public rescue services, which are then burdened by false positives due to Apple unilaterally letting their phones dial emergency numbers and putting the onus on the user to handle 'impact classification'.
I don’t think so, no. By this logic, all roads would be toll roads, all schools would be private, etc. I would not like to live in that society. If Android or other device manufacturers had equivalent features, they can have that too. We can discuss what the maximum false positive rate can legally be obliged to be, but I think having it be 0% is not a net positive for society.
Flooding emergency services with fake incidents can easily start killing people due to help being called just because some idiot manager thought 'this is good marketing so lets get this done quickly' and pushed half-baked solution. I ain't talking about theory, read this thread about feature disabled and still calling 911, imagine if you had phone in trunk of the car.
Glad it doesn't seem to be enabled in Switzerland, my wife is a doctor, worked on emergency and also as a doctor with ambulances so has perspective from other side. This is outright criminal behavior from Apple, easily to get sued for billions in each bigger market (not saying they would automatically lose, they can play well meaning idiot at the courts but PR effect is nevertheless properly bad).
I stopped wearing my Apple Watch while surfing because this feature would trigger almost every time. It’s made worse by the fact that the touch screen is inoperable when the phone senses that it’s wet, so it’s difficult to cancel.
Why didn't you just disable the feature? In the Apple Watch app on your iPhone, tap "Emergency SOS" and then turn off "Severe Crash" and/or "Fall Detection".
I’m a firefighter/paramedic in a suburban town. We average one of these a month (false automated reports of vehicle crashes). I’ve never see a true positive report.
I had this happen to me in an unexpected way as well. I parked my car on the side of the road, ran to my parents’ house, ran back to the car and jumped in. When I jumped in the car, the phone thought I had been in an accident and called emergency services. Luckily I think I was able to cancel it in time.
Seems like an edge case that Apple can solve by comparing phone behavior on a ski lift vs that of a car. There’s gotta be some patterns in the data that are distinctly skier behavior. However if they want to help those skiers that are truly in need of help, it gets much trickier.
It's probably not just the lifts. Anybody new to skis/snowboards is likely to do a lot of falling over.
(But if you're expecting to be frequently falling onto wet snow or hard ice, it's probably a good idea not to have a $1000 smartphone in your pocket...)
Learning to ski is certainly a particularly strong example of a case where you're likely to have a lot of false positives.
But, honestly, even with something like hiking, tripping over a tree root or something like that isn't exactly rare. I've taken a ton of spills over the years and fortunately nothing worse than some minor bloodshed or maybe a twisted something was involved. (And the one time it was something more serious I didn't actually fall but did have a serious bone break.)
I've been on the fence whether I ought to enable the feature on my watch.
It's not too difficult to solve actually. They could monitor for movement after the fall. If the person hasn't moved or has moved, but hasn't got up, then it should announce a warning (with enough time) before calling emergency services.
If they got up and have started walking again, its a false positive, or they don't need automatic help.
I guess that in the emergency case the phone will automatically determine the GPS position. For the fire departments it might make sense to discard the automated calls if the position is on a serviced skiing slope during the slopes operation times since there, in case of an anncident, in 99% of cases others are there for helping immediately. Also the emergency service would call the slopes emergency helpers anyway since they have the skidos ready for rescue on the slopes.
Exactly, this was first reported months ago and it should be easy to fix. Check location for ski slopes, see if they're moving normally again after 20-30 seconds, etc.
Maybe Apple is delaying the fix because these news stories are viral marketing for its crash detection feature?
I am surprised this edge case exists - last time I set up an iPhone with Apple Watch I had to specifically enable the feature (it was for an elderly person so it made sense)
> The department usually calls the number to check the situation when it receives an automated alert but there is no follow-up from the smartphone’s user. If the user does not answer, the department calls again sometime later
This doesn’t make any sense to me. The user has been in an (assumed) crash where they are presumably unable to call 119 themselves, but the fire department tries calling back to see if it was an actual emergency?
Then if nobody answers, instead of rushing over there, they try again in a little while?
I think it is safe to assume that the "don't rush over there" behavior is a learned one. After rushing over many times for what appears to be a false alarm, you start to think more of the calls are false alarms.
There is even a parable about this. Several of them, actually. :D
If they are resource constrained then this just seems like a form of triaging where non-automated calls are prioritized. I don't see the problem with this approach if they are seeing a lot of false positives from automated sources.
Tangential, but I was looking into a monitored alarm system recently for my home. For £50/month, the alarm company will respond to an alarm and try to phone me. If they cannot get through to me, or I confirm I'm not home, they will dispatch a guard to arrive within 30 minutes. They will assess whether or not there is someone there and will call the police if there is, wait until they arrive, and charge £70 per callout with some standing charge for however long they wait for the police too.
Needless to say I went with a Ring alarm, which is £8/month, and will notify me immediately, and I can call the police myself.
On what basis would it be illegal? Does that site have an obligation to serve you? Are they not serving you because of your membership in a protected class in their jurisdiction?
It would be illegal if it was against the law. If the government in a jurisdiction the site operates in decided to make that the case then it would be illegal.
It's their thing and they can do whatever they want with it.
However, you can use Ublock Origin's "block element" feature (right mouse button over the element you want to block opens the contextual menu) to block all those annoying modals. It works on some other sites too, while others will not merely cover their content with modals but serve an incomplete article.
FWIW, I just wrote a privacy policy and terms of service for my app. I’m not a lawyer and I don’t understand GDPR. I could probably understand it if I took the time to do some research. But, at this point, I don’t have a single user and I don’t think going after the EU as an early adopter (and the time spent to make sure I’m actually compliant with their laws) is a better use of my time than targeting people in a jurisdiction I understand and can be reasonably sure I’m following the laws of.
So, instead of figuring out all of the laws of all the countries, my ToS require you to attest that you’re in the United States and using the service under U.S. law. If you try to access my product from the EU, you’re in breach of our contract. If you believe I’m mishandling your data under another countries law, send me your account and I’ll gladly disable it.
I suspect this is going to become increasingly common as companies try to regulate digital imports/exports. I’m not interested in assuming legal liability for sending bits across a border.
Following GDPR is effectively equivalent to respecting your users' privacy. What are you doing that the GDPR doesn't allow?
You do you of course but at this point GDPR has been round long enough I'm assuming you're looking to double dip and sell my details. Best case scenario you don't trust your own security.
GDPR is only really scary if you're doing questionable things with your user data.
If you're just annoyed at it because it's another thing you need to learn (you have you learn stuff to do a business??) alright fair, no worries. Look into it later tho.
If it's in your way? You're posting your receipts to HN for folks to reference later in jokes haha. Your very own "They trust me. Dumb fucks"
E: Case in point one of your Show HNs handily referenced "End Mass Surveillance
Our government is actively breaking its own laws. [..]"
Why would anyone believe in your campaign when you don't even want to end mass surveillance in your own projects? :)
Maybe he wants to avoid the cost of a paying a service to act as his representative in the Union as required by Article 27 if GDPR applies to his business under Article 3(2).
Whether Article 3(2) applies is somewhat subjective and a big part of it is intent. Blocking EU IP addresses and/or requiring people to say they are not in the EU before allowing them to use your site would help prove that you did not intend to serve people in the EU.
Or maybe he's worried about IP addresses. Regulators in Europe has said they are personal data that is covered by GDPR. If you have to give IP addresses the full GDPR treatment that could be a major hassle for a small organization.
So again if you aren't definitely intending to serve EU visitors blocking them can bolster the case that you don't fall under Article 3(2).
Probably hasn’t appointed a representative in the EU. There are a lot of other bureaucratic requirements. It’s really not the same thing as respecting privacy.
> Following GDPR is effectively equivalent to respecting your users' privacy. What are you doing that the GDPR doesn't allow?
Might be. Might not. No way for me to know without understanding the GDPR. And I’m not interested in studying that law at this point.
My privacy policy is transparent, lists the vendors I do business with and how they are involved with handling data, and what I use data for.
> GDPR is only really scary if you're doing questionable things with your user data.
Any regulation that contains any sort of legal liability that I don’t understand is scary. If I’m compliant, it’s by luck. I’m certainly not compliant deliberately because I don’t understand the GDPR. Am I willing to gamble on luck so EU citizens can access my service?
Nope.
I’d go as far as saying “just be a good person and assume you aren’t going to be held liable in the EU” is a bad take and you shouldn’t promote people taking such a lax approach to legal compliance.
Sorry for disengaging here but I got nothing lmao. From my POV that's like deciding to not follow copyright law anymore.
Uh.. cool. Lol.
If you're legit doing the rest as you say then congratulations you're already GDPR compliant, you're just being stubborn. Seems a waste of brain cycles.
> If you're legit doing the rest as you say then congratulations you're already GDPR compliant, you're just being stubborn. Seems a waste of brain cycles.
How many jurisdictions exchange packets on the internet with your country of residence?
How confident are you that your web service is compliant with all of them simultaneously? Are you compliant with Saudi Arabia’s laws? India? Pakistan? Russia? China? Argentina? Nicaragua? Cuba?
I’m suggesting that conducting business in a jurisdiction without first ensuring you’re compliant with the laws of that jurisdiction is a bad way to conduct business. Making sure I’m compliant with EU law is a waste of brain cells right now. Making sure I’m compliant with U.S. law is taxing enough.
I’m also suggesting that you probably shouldn’t be liable for a random country’s law just because one of their citizens misrepresented themselves and tricked you into exchanging packets under a different set of laws than you know how to operate in.
Your contract is irrelevant. If you process data of EU subjects, then you're bound by the rules of GDPR regardless of what your contract contains. You can't sign away your rights in the EU, so I can lie and say that I'm in the US, and that's entirely your problem.
Will be interesting so see how this shakes out in the courts over the next few decades.
It’s not possible to know the legal jurisdiction of the user represented by an incoming packet with today’s internet.
To me this sounds like a random person in a random country smuggling their data across a boarder into my system without my consent. Have a hard time reconciling my views on privacy and individual freedom with a world where a random country can hold me liable for some random law in their jurisdiction when I never consented to doing business with their citizens.
Is it his problem? He doesn’t have to respect GDPR because he doesn’t operate in the EU. Sounds like it’s your problem if he doesn’t handle your data the way you think he does.
European authorities are still able to enforce fines against him unless he’s exceedingly careful, even if he doesn’t have direct presence in the EU.
This is why many websites just block European IP addresses entirely.
You might think you’re safe in the US, but perhaps you use a payment processor with significant European presence? Stripe or Paypal, for example. European authorities can take your money.
> This is why many websites just block European IP addresses entirely.
This is not sufficient. IP addresses do not have sovereign rights and only loosely correlate with the legal jurisdiction of the user behind the originating packet.
This is a world where, by connecting to the internet and exchanging packets, you are simultaneously liable for every law under every jurisdiction; it’s just a game of roulette which jurisdiction the packet you receive is coming from.
This doesn’t seem scalable, sustainable, or particularly good for human/civil rights.
No, they can’t. People in sovereign countries aren’t beholden to your country’s laws. The EU can block access to the site from inside but nothing more. It doesn’t rule the world.
I'm 99% sure she disabled that feature after that.