It's funny that when discussing the topic of "updates" and obsolescence, users focus on the vendor as if they have exclusive control over the situation. Authors of applications may also play a role. For example, when they "update" their applications to only work with newer versions of Android. Depending on the user's application needs, that can shorten the life of a an Android device. Some applications will continue to work with both older and newer Android versions, some will not. For example, F-Droid has numerous programs that will work on older, "obsolete" Android versions. This allows older hardware to be re-purposed and to continue to be useful for some uses. Not sure that Apple has anything like this; consider how many programs in the Apple App Store work with older iOS versions.

Both Android vendor and Apple hardware continue to work long after the software has become "outdated". That hardware does not die when the software becomes "obsolete". The vendor may choose to ignore this fact in the interest of sales but it does not mean that authors of applications must ignore it as well.

The third factor besides the vendor and the authors of applications are the operating system authors. With older PC-like hardware, I can run the latest versions of NetBSD. Forever. I update when and if I decide it is time. x86 has its benefits. It is sad that these pocket-sized computers called "smartphones" are so inflexible.

A non-HN reader recently told me that the "tech" industry has turned us all into "beta testers". The entire "updates" concept needs a serious examination. Updates are not a substitute for quality control.