Running Unsupported iOS on Deprecated Devices
nyansatan.github.io
nyansatan.github.io
Better they allowed installing linux on old devices. But even then it wouldn't move a needle - it's such a niche case.
Maybe some would end up in extremely poor countries, but even there people can afford $30-50 for a brand new computer and Apple rather get those old devices recycled properly.
This comment gave me whiplash
Huh?
Which is not how I spend my money -- I have always purchased unlocked phones when they are on sale -- but there are too few of us.
No, at least for Apple devices, the overwhelming majority are replaced before they reach EOL. According to https://telemetrydeck.com/survey/apple/iPhone/models/, only around 25% of people are using iPhones that were released more than 3 years ago.
Maybe more people aren't running older hardware because it's too difficult, rather than because they don't want to. The basic idea is here is taht if a device can still hold a charge and the user is OK with limited features, they should be able to keep using it as long as they feel like it.
Citing large absolute numbers for rhetorical effect is dishonest because multiplying a huge number with any percentage will result in a shockingly large number. The original claim is that "people who keep computers from 2009 to play with and wish they could get more use out of their 12 year old iPad Air [...] it's simply not a thing for most people", which is true even, if there are millions in absolute terms.
Unfortunately, battery technology doesn't - and even if we had long lasting batteries, we'd also need fall-resistant screens. And no matter what, even if you have a device held together by screws and allowing easy repair instead of messing around with glue and click-tabs... screens still are really expensive, making it often enough more worthwhile to take the opportunity and upgrade the whole device rather than to repair the screen.
Not in most phones. You always have to mess with glue and unless you take extreme care and caution to remove _all_ pieces of it you will end up with compromised water-tightness, not to mention risking the screen cracking or being exposed to air (ruining OLEDs).
> my main desktop monitor
We were talking about phones. Phones get dropped, scratched by keys, ... the list why phone LCDs/OLEDs can get broken is loooong.
In addition, they don't want to spend money on it. They'd rather spend money on things they actually care about. Festivals, clubs, vacations, a new TV, a car, restaurants, whatever. Your average non-tech person is happy if they don't have to spend anything on gadgets for 10 years.
Ssssshhhh ..... Microsoft does not want people to hear this .....
That's not a high bar to clear. Who's realistically going to use a laptop/desktop with a Core 2 Duo (2006), for instance?
But the thing is.....old PCs are really not that unreliable. If they survived the last 5-10 years then they are probably still chugging along just fine and for a small business there is literally nothing wrong with using them.
Well... Outlook is already a web app, the rest of the Office suite will follow rather sooner than later, and inventory - it's either web apps or SAP, both memory hogs.
I was literally still using a Core2Duo Macbook Pro as a kitchen laptop just for looking up recipes and watching youtube videos etc until last year. Worked absolutely fine until Chrome decided that it's not going to update itself anymore and since I'm on an old version of chrome I can't use google sync. That's what killed it for me - the hardware itself was still perfectly functional.
Used by a tiny percentage only because Apple has made it as difficult as possible to not upgrade, which is especially egregious precisely because their devices are long-lasting.
(This comment brought to you via a perfectly functioning iPhone 8 running the latest possible iOS that supports it.)
How does apple do this ?
Compare that to the number of 1st gen iPads or older iPhones and the demand might not be to the same level to justify the endless reverse engineering effort.
The solution is for the companies to have to open up the device once it's officially not supported.
This would prevent e-waste and put this old hardware to better use. A community OS could then be built on top of this common SW-layer and be maintained for a wider range of devices.
I would e.g. LOVE a "Browser on everything" OS which just provides a Browser OS for outdated hardware, but the only way this could work on scale would be if the device-vendor would be mandated to provide and document the lower layer...
Someone would have to make the economic case for such a regulation as well, i.e. demonstrate the benefit for society and reduction of e-waste if such a law is in place. But the chances for this are razor-thin, especially in today's public/political climate.
If the App Store still doesn't work, you can always jailbreak and install apps on your own
Obviously not everything works, but someone did try.
Go to the App Store on an iPhone 7 and every app will refuse to install because it requires the latest and greatest OS even though they used to work just fine on the old OS. They deliberately drop support even when they don’t have to. Total shitty behavior.
I gotta throw away my phone because you, Mr. Developer, can’t be bothered to keep the old code around for backward compatibility.
If you’re writing an app that targets the newest hardware features, say because you’re making a camera app that uses the latest updates, it’s not going to run on iOS 5. You can’t hold that against app authors, or even against Apple, really. There’s not a lot of return on investment for sinking thousands of dev hours into supporting ancient phones that almost no one uses, and which by definition are more likely to be used by people who won’t spend a dime on apps or services.
In some cases (such as with Google Play), app stores will even unpublish apps that were built with old versions of the SDK.
In other words, if you want to update your app at all, you have to stick to the rules that they provide regarding SDK versions.
[0] https://developer.apple.com/news/upcoming-requirements/?id=0...
I'm sorry for being unclear.
And additionally, even if there once was a compatible version, Apple only lets end users download it if they have previously purchased/downloaded the app.
In my opinion, this is almost fully Apple's fault
TL;DW
1. you must use current Xcode to submit to the App Store
2. Current Xcode only supports a short list of the most recent iOS SDKs as targets
3. Therefore you can’t make a build with an older Xcode to submit, and you can’t make a build that supports say, iOS 15, with current Xcode.
The video highlights a complete hack which can for now be used to make builds that target older SDKs, but Apple could change their systems to break that at any time, and in fact the latest iOSs don’t use these device support files that he shows are the key to the hack. So while you can do this now for old iOS, when 26 is too outdated for Apple to want you to target, this hack won’t work to bring it back.
https://developer.apple.com/support/xcode/ the iOS 26 SDK can target iOS 15 or later
There aren’t requirements to update the minimum. They’re independent. There might be limitations on what APIs are still available though in Xcode if they’re removed after a deprecation period however.
You can use the latest SDK, and if you don’t use any new APIs, your app will continue to deploy to whatever minimum you have set.
If you want to use the new APIs, you can put their use behind availability checks.
It’s just that at some point, as a dev, you want to stop testing on older devices and you want to stop branching your code for OS versions that are a minuscule fraction of the active user base. The ROI changes quickly.
The App store requires you to link with a minimum SDK version, but the minimum supported version is set independently by the developer in Xcode. The latest version of Xcode + the latest SDK can still support devices running iOS 15. Developers are deliberately choosing to drop such support. Or maybe they just don't know how, as evidenced in this thread.
Yes, it is I, Mr Developer, that decided that every year the minimum XCode version / SDK Version must be raised
In that era, Safari added new features, including adding support for web standards, in the major update and in a .1 update about 6 months later, Safari 12.1 came out in March 2019.
It cannot be a good deal to buy an expensive device that is guaranteed complete deprecation in a 7 year window (and very likely to suffer some major annoyance at the 5 years mark).
If you buy an equally expensive Android device (or Windows PC), because the model is different, you'll fare much better. And the reality is that at equal characteristics/performance the Android/PC version is generally quite cheaper.
Apple aficionados will rave about the build quality and ecosystem but I think they are irrelevant (most of the ecosystem advantages are to be found in the alternate world) and the devices feel more premium than they are actually qualitative (in fact they are very fragile in many ways).
They are surfing the Apple Silicon wave but as their efficiency lead is shrinking it will lose relevance/power.
You call it “old but capable” but it’s really more close to just “old.”
We really want tech to be less disposable but the problem is that tech is still progressing fast.
Automobiles have barely changed in the past few decades as a fundamental concept and in general capability and road-worthiness and that’s why you see a lot of 20-30 year old examples on the road.
But imagine that buying a 2000 Volkswagen Jetta meant a car that had a 20 horsepower engine that gets 20mpg. That’s what old tech devices are often like. Sure, you can use it as a glorified gas-guzzling golf cart, but not many people need that and they’d rather just get a golf cart if they do.
A 2013 iPad Air is not going to be a very usable experience as the device was originally intended even if you get suitably lean software on it.
This is a dual core device with 1GB of RAM. At this point it can barely browse the web in an acceptably performant way. We can lament our inefficient web apps and fuss and moan but if grandma can’t get her slippers from Amazon without waiting half a century for the page to load it’s not a useful device anymore.
Sure, you can use it as a server or something, or maybe some kind of smart display, it would work fine. But let’s go back to the golf cart analogy: presumably the original XX million units that were sold can’t all be web servers or smart home screens. The quantity of people who originally bought them for the original mainstream purpose have moved on to something newer and aren’t looking for a niche secondary use case. You have to be a very specific person to try to fit that square peg into a round hole.
I have been a user of the OpenCore Legacy patcher. I bought a 2012 Mac mini, excited that I could use it as a Mac server with the latest OS. The experience was sluggish at best even with a brand new SSD installed and RAM maxed out. I also had random kernel panics that I couldn’t resolve. So I ditched the Mac server idea and installed Linux. I went with that for a while but it turned out to have insufficient single core performance for my applications. The architecture also couldn’t accept more RAM even if I found higher capacity sticks, the technology was completely at the limits. I ended up selling it and built a server with much newer (but still mostly used) parts instead.
The calculus for reuse gets even worse if you start thinking about performance per watt and energy efficiency. There are some devices especially in the desktop category where you get to 15 years old and you have a real legitimate “I’ll save more energy and cost on my power bill” argument.
E.g., Let’s say I own an AMD FX 9590 (2013) with its 220W TDP, I can replace that with an Intel N355 15W embedded class chip and it’ll be faster by a double digit percentage with the same number of threads. I know this is an extreme example but it’s still a demonstrator in the amount the technology has changed.
This would be like if my 2013 Toyota got 2mpg.
I think we need to be more pragmatic and accept these devices for what they are: a temporarily owned thing, almost like a lease. They aren’t all that different than buying some fast food that comes in disposable packaging, the difference is the time scale. Once you’ve eaten it, you’ve consumed it, and it’s over. There’s maybe a bit of material you can recover at the end, let’s call it 10%. And we have to grapple with that reality rather than pretending we can fully change it.
Don’t get me wrong, I’m not trying to argue against right to repair. Yes, they should have things like replaceable components and a requirement to become more open as they age. They should be designed to be as usable as possible when they become older. At the same time, we should be pragmatic and accept that the most likely scenario is that a device like a 2013 iPad Air will still only see 5-10% of buyers reusing the device in this way rather than sending it to the bin even in the most ideal scenarios. Of course, that number is a lot better than some smaller number like 1-5%, with many iPads being thrown out as soon as the battery swells.
These devices are are small, snappy and powerful enough in 2025.
† I don't have the skills to reverse the process, though :)
Take me back to the days where things were governed by UX and not revenue.
This happens every time, a redesign comes out and people hate it. Same when iOS flat design came out. I dug into the archives, because HN has been around for a while now so we can do that, yay!
https://news.ycombinator.com/item?id=5869121
(hand picking / editing comments freely because I can)
> I read article after article from historically pro-Apple bloggers/authors explaining that no, flat design was fundamentally a bad move: the strongest metaphor is that of the phone as a tool -- that we needed skeumorphism, we need hints for interactivity, we needed polish.
> I think iOS 7, on the whole, looks worse than iOS 6. The stock icons look outright ugly; interfaces like the call-answer screen and the calculator look poorly designed, and everything has the sense that it just needs another run or two through the review process.
> Look at his iMessage screen comparison [1]: yes, the old screen looks a bit geocities, but you can actually read text very well; the new screen is almost unreadable. The prime aim of iMessage is to make people read text, not to look cool.
> [1] https://www.flickr.com/photos/mattgemmell/9023510971/
> Hopefully I'm not the only one that thinks this is going to kill usability. The reason old people can figure their way around <=iOS6 is that everything that can be tapped looks like a button. A more "mature" audience isn't what apple's good at appealing to.
> I'm volunteering at a center that teaches senior citizens various computer skills. One of the courses we teach is on how to use their iPhones. I'm dreading the moment that iOS7 is released: all of these people are going to have to start right back at the beginning in their understanding.
https://news.ycombinator.com/item?id=5856398
---
I'm fairly sure I read the thread I'm quoting from at the time because all of these arguments are still fairly fresh in my head.
Just a browser is all I want.
The code or schematics doesn’t even have to be freely licensed and distribution could still be restricted, patents still apply and what not, but programmers could develop their own ”patch sets” on top of the proprietary code that can only be used by people who already have the code, extending the lifetime and adding features.
Obviously it’s not perfect but I think it’s a pretty reasonable compromise between companies IP interests and the rights of consumers and fighting e-waste, because the way this is set up today is in my opinion quite absurd. There’s no natural law forcing us to accept planned obsolescence.
$(cat n18.10A403.kextlist | sed 's/^/--bundle-id /') - this weird expression appends --bundle-id to every line from the file at n18.10A403.kextlist.
The oldest iPhone OS that natively boots on my particular one is 1.1.4, 1.1.1 (which is the highest version number where you can trivially escape the OOBE via the emergency dialer) fails to initialise the FTL (flash translation layer), probably because the chip is sufficiently different from that used in the older phones.
It would bring me great joy to be able to relive emergency dialer hacktivation again, but I have lost that particular iPhone 2G, and only have this 16GB one left.