Yes, turning Background App Refresh off goes a long way toward achieving this (and saving battery). Even with Background App Refresh turned off though, some activities may continue, particularly Location, audio, voice, and the like [https://developer.apple.com/library/content/documentation/iP...]. This is why it's Location "While Using" is important. It also looks from OP's comment [https://news.ycombinator.com/item?id=14502423] that (some of?) these features will be more prominently displayed to users. This is good, users should have complete knowledge and control of their devices, particularly as it affects privacy.
In iOS 10 Settings > Privacy > Location, under the "Location Services & Privacy link, it says: "Your iPhone will keep track of places you have recently been, as well as how often and when you visited them, in order to learn places that are significant to you. This data is kept solely on your device and will not be sent to Apple without your consent. It will be used to provide you with personalized services, such as predictive traffic routing." This is consistent with Apple's philosophy of keeping personal data processing on-device as much as possible to protect user privacy. You can always keep the frequent locations feature turned off, or use "Clear History..."
It's worth mentioning though that there's often not a 100% feature parity between iOS versions on different devices, typically because of hardware changes. These even applies to devices of the same generation, e.g. iPhone 7 and iPhone 7 Plus have different camera features since the latter has two cameras and the former just has one.
The differences between feature sets on different generations is less and less it seems though. If I recall correctly, the differences between the iPhone 3G and iPhone 4 was bigger than for instance iPhone 6 and iPhone 7. If I weren't such a sucker for new and shiny things, I'd probably buy an iPhone SE today.
And my Galaxy Nexus stopped receiving updates 2-3 years ago despite it being a Google device (mine was also straight from the Google store, so it wasn't the carrier's fault).
Cyanogenmod supported the Galaxy Nexus until 13.0 which is Android 6.0.1. So Google just dropped support after 4.3 (coincidentally the last unsupported version) even though it was absolutely possible to update to newer versions of Android. As a result the Galaxy Nexus doesn't receive security updates from Google and is thus not usable as an online device.
While iOS upgrades are usually free these days, it hasn't always been like this. According to wikipedia, iOS 1.1.3 was a paid upgrade on iPod Touches (while being a free upgrade for iPhones).
No they are right. Due to some sort of accounting thing once upon a time on some of the first iPod touches there would be a small fee to upgrade the OS.
Next battle: getting Apple to give users more granular Location controls. Almost NO App really needs precise GPS. "within a mile" or city or zip code or other fuzzier locations should be fine for most apps. Users should have a better trade-off than exact or nothing.
Screw any app that really thinks it needs my precise coordinates in the foreground, let alone the background. Turn-by-turn directions in maps is the only exception I've found in 10 years of iPhone use.
Off the top of my head, any app involving directions – public transport, say – can reasonably use precise coordinates. Uber is also a reasonable use-case (when you want it to use that data) given the density of some city streets. Then there are location-sharing apps generally, or apps where you want to check in to locations (maybe a bit fuzzier, but there could be a lot of taggable locations within say 500m), or camera applications, or geocaching, or…
It's great to have control, and it would be even better if we were able to specify the accuracy an app is able to achieve. But there are obviously some reasonable use-cases for both foreground and background tracking.
For an app like that, I'd like to see a "approve every request" option. I don't always need my transit app requesting my position (even if only when in the foreground) when I really only need to look up the nearest stop occasionally.
I want to be tracked by Google. I went out of my way to turn it on in iPhone. My location is leaked to cell towers and by what wifi network I'm connecting through anyway, and I'd rather be able to figure out which cool restaurant I went to months after my visit to Vilnius is over.
Plus my memory of my personal life is so unreliable, especially if I don't have any context, so if I'm ever in court I'd like to be able to share exactly where I was when I was there.
I've stopped thinking that real privacy is attainable. That being said, I refuse to share it with scummy companies like Uber.
A car dealership who botched a warranty repair on my vehicle is stonewalling me - once they realized the error was likely to cost them $2k - $3k, their service department deleted my file and began claiming no evidence exists that I had ever been to their location. I got the scheduling department to confirm my appointment and just found my location history for that day showing that I was at the dealership right on time during the appointment. Going to send this to the manufacturer's corporate complaints departments. Thanks for the heads up.
Google's location history is excellent and I'd gladly pay for an equivalent application which I could self-host and set up with the minimum of hassle, but having them do it creeps me out too much. I've not managed to work out how to get OwnTracks to substitute fully for it.
Awesome, thanks for the link. I had been using Moves by ProtoGeo
https://appsto.re/us/PNDwE.i and it's not very accurate. Still good for going back and remembering that awesome random place you found on vacation after the fact so you can recommend it.
I also wish the NSA had something like a personal data explore, but I know that will never happen.
Unfortunately I've not found WHIB to be enormously accurate either, but it does well enough. The main nuisance is interpreting having paused whilst walking past an establishment has having gone in it.
Even Google's version couldn't get this right, at least in my case.
Explicitly decided to keep Google Maps' location history activated on my device for precisely the same reasons. I'm yet to use it in court or something like that, but the argument about human memory resonates with me strongly - I've used location history many times to review e.g. on which day I was in some place or other few years ago. (Or once, years ago, to check how exactly did I get home after one party.)
RE people saying "just take a photo and use EXIF", that won't work because I don't know in advance what locations I will be interested in in the future.
Apple's API don't discriminate with regards to precision. An app that shows nearby stores with a 1km radius precision requires the same level of permissions as an app that's trying to do 1m precision.
Apple's API allow very fine-grained control of both the accuracy and frequency of location updates, depending on, among other things, running in the foreground/background, the user's location etc.: https://developer.apple.com/library/content/documentation/Pe...
There's currently no explicit penalty for requesting too much information, but the battery power indication could probably get users to uninstall apps wasting energy, and it can even get you app terminated.
Unlike Android which have precise and coarse location. If Google wanted, they could limit precise to GPS and similar apps and force everyone else to use coarse location or geo fencing.
Unfortunately Google doesn't care about your privacy. In fact, it's only a matter of time before they remove coarse locations and/or give every app full access to location data (the same way they did with internet access).
That's a good point. Maybe app developers should be able to request the level of access they want, like precise location, neighborhood (postal code), city, state or country.
If an app never needs more precise information than the country (say Spotify trying to decide whether they operate in your country), they don't need to request it. That also guards against bugs or security breaches in apps that try to do the right thing.
Right. An application can ask for what it wants (low accuracy is much faster and burns less battery since it doesn't start GPS) but there's no way for the user to say that an app can ONLY get low accuracy data.
That's your personal choice and desired level of privacy. I like that my Lyft app knows where to set pick up and that my transit app knows the nearest street car location.
Agree with the OP that Uber can burn in hell for "always on"; I guess just in case I need an emergency Uber ride :-/
But feel free to just uninstall all those other apps that I love using.
Geotagging in photos it pretty useful. And find my iPhone. Then there's Pokemon Go - maybe not your cup of tea, but a lot of people love it. It's pretty useful to be able to share your location with someone in iMessage, I've done that a few times. Your mileage may vary.
I had a completely different reading. I thought the fable portrayed Alice as thoughtful and bright. The punchline is that this talented student was right to leave programming for cooking because the lack of sequential consistency in high-level languages is absurd.
I saw that a few times. Brilliant college minds failing at programming because of legacy and ad-hoc features. They preferred maths and physics, not devoid of arbitrary but, probably tinier and more stable over time.
But is the argument for higher consistency based on the existence of "saner" architecture as in x86, and wasn't argument insufficient in the time the article was written (1998) as well as is now, given that ARM predates the article and its popularity significantly increased since?
Then ignoring the existence of the weaker memory model of ARM (versus the stronger x86/x64) is even stranger. And even if x86/x64 models are stronger, they still need some fences and using them all the time would be too slow. So I still don't really understand the arguments of the article.
The article does not ignore architectures like ARM. You do need fences, but not all the time - the compiler can avoid fences in places where there is no danger of violating sequential consistency (e.g. on accesses to provably local objects).
Yeah, that's my understanding too. It's not "math is hard let's go cooking" it "fuck this nonsense". It's not math problems that are the problem here, it's inconsistent execution in the name of performance.
To be precise, Alice quits programming because modern languages lack proper concurrency specification, and thus lack actual "Math" in this area. I can sympathize with that.
The concurrency behavior of Java/C++/etc is specified, it's just that the specification doesn't match the model given in the article. It follows a different mathematical model—that doesn't mean that it "lacks actual math".
What don't you like about that? If you heat the vegetables and oil together, they can slowly lose their water and essentially boil, and taste boring; if you let the oil get to the right temperature and then add the vegetables, then they'll cook up nicely, with plenty of Maillard reactions (browning), and taste delicious.
