[627467]

Lately I've been thinking of the ethics of buying Apple iOS devices compared to more opened devices - most, yet, not all Android devices. In the long term, iOS device are doomed to become unserviceable my users/enthusiasts dues to its closed nature, while many Android devices seem to have it's usefulness extended in time by efforts such as postmarketOS.

Is there an inherent quality of iOS devices that make them unsupportable.

Any given gen of iOS device has, arguably, more devices than any given flavor of Android devices (more fragmentation in Android platform) so in theory it would be more fruitful and impactful to extend the use of iOS devices (if at all possible) yet, this is not what we see. Leading me to think that Apple deliberately makes iOS devices unthinkerable.

This, to me, leads to the conclusion that buying iOS devices is unethical as the only long term alternative for iOS device is to have it recycled while android devices may retain the chance for reuse into the future.

[teruakohatu]

> Is there an inherent quality of iOS devices that make them unsupportable.

When you say inherit do you mean the Apple Axx CPU can only only execute the iOS kernel and nothing else? Then the answer is no.

https://fossbytes.com/linux-based-mobile-postmarketos-on-iph...

But for all practical purposes it cannot run anything but iOS and any discussions about running an aftermarket OS is purely academic.

[arghwhat]

> But for all practical purposes it cannot run anything but iOS and any discussions about running an aftermarket OS is purely academic.

The project you referred to had a follow-up in late april: https://blog.project-insanity.org/2020/04/22/linux-with-wayl....

Calling the project (or any project, really) "purely academic" with the implication that it will never evolve is nonsense. Like many other projects it's of course done for the hell of it, but so is most hobby/open-source projects, and yet many evolve into something much greater. There's plenty of reverse-engineering projects out there that have brought open life into closed hardware.

The question is just whether or not there's enough people interested for it to happen.

[teruakohatu]

"purely academic" is not an insult. I think it a statement of fact of where running Linux on iPhone is right now. Aluminium–air battery may change the world but right now they are also "purely academic", as Lithium-Ion once was.

Do you honestly believe the average consumer, or even the average enthusiastic is going to be running Linux on the latest iPhones anytime soon? Apple is dedicated to locking iOS, and probably MacOS, more every hardware iteration. And by running anytime soon I mean usable LTE, GPS, Bluetooth and Wifi.

[arghwhat]

The opposite of "purely academic" is not being in the hands of average consumers and/or enthusiasts, nor is it that everything works and is usable.

Case in point: postmarketOS and UBPorts, as well-established and very serious projects with plenty of traction, are not in the hands of the average enthusiast, and still has ways to go to handle even the Pinephone properly.

[627467]

Both pmOS and UBports are 2 of the examples I have in mind demonstrating the conceptual resilience of non-iOS ecosystem: these projects have small and fluid teams and yet manage to support dozens of devices with different HW requirements. iOS devices are so homogenous, and yet, so hard to them to be the center of these aftermarket efforts.

[627467]

Yes, I became aware of that effort. And I applaud and look forward to them.

The problem I see in long term is: if there's no inherent property of iPhones that make them unhackable by aftermarket efforts I don't understand why there aren't more efforts or more success. Any given model of iPhone has many more units out in the market than almost any given model of android device that is officially supported by aftermarket efforts (ie. LineageOS, postmarketOS, etc).

[sergeykish]

Drivers.

I have a netbook with GMA 500, it has closed source driver for XOrg 1.9 (ten years old) and corresponding kernel [1]. I bought it because Intel graphic cards had open source drivers, not this one. Later I thought as you - someone would hack it, yes, kind of - there is a version without hardware acceleration. Eventually I was living with outdated XOrg server and patching driver to support updated kernel [2]. That I think is about as good as LineageOS - much better start than iPhone.

Hacking out of necessity is a bad idea. If not PinePhone I would definitely check open source drivers support in Replicant [3]. iPhone is too far away (and its users like iOS).

1: https://en.wikipedia.org/wiki/System_Controller_Hub#GMA_500_...

2: http://sergeykish.com/linux-poulsbo-emgd

3: https://redmine.replicant.us/projects/replicant/wiki/DeviceS...