I agree with you completely.

I do want to add that, I'd wager this overall scenario would be present regardless how long the release schedules are. Especially since people are wanting to continue to use their 10-15 year old hardware.

It's somewhat frustrating that many useful features are also in opposition to repair-ability. Most performance improvements we've seen in the last 10 years basically require components to get closer and closer. Soldered RAM isn't just the musings of a madman at Apple, it's the only way to control noise on the wires to get the speeds the hardware is capable of. Interchangeable parts are nice but every modular component adds a failure point.

Add in things like say, a level of water resistance. Makes the device way more durable to daily life at the expense of difficult repairs. Or security, it's great grandma won't get her phone hacked but now we can't run our own code.

There's also a bit of an internet-commie brainworm that I'm still trying to pin down. Like you said, "We could cut back on a lot of waste if R&D took longer and was geared toward longevity and repairing rather than replacing", what does this actually look like? I think it's at odds with how most of us use technology. Do we want somewhat delicate, fully modular, bulky devices? What does it mean to be repairable if the entire main-board is a unit? If you need to make each component of the main board modular, the device will quadruple in size, making it overall worse to use than the small disposable device(more expensive too). The interconnects will wear out, making modules behave in unexpected ways. The level of effort required to support a dozen interconnected modules for years and years would be substantial. Not only that, the level of expertise to repair things at a level below a plug-and-play module is far higher.

I had a small wireless dongle die on me after about a year of use. It basically stopped connecting to my phone. I noticed that it would connect for a short period before disconnecting. When the device hadn't been used in a while, this period was longer. Due to my own repair experience, I knew this was a product of a cracked solder joint and expansion due to heat. I brought it into work and held a reflow gun to it for 5 or so minutes, heating up the PCB enough to reflow whatever bad joint was causing the issue. I looked this up after the fact, and found another person had found the same solution, basically told people to put the whole device in the oven for a period and hope it fixed it. People commented in disbelief that this seemingly magical ritual could revive such a black-hole of a device. They couldn't comprehend how someone could come across this solution. Of course it wasn't magic, it was fairly simple if you have encountered thermal expansion, cracked solder joints, and misbehaving electronics in the past. The point of this story is that, had this not been the solution, the device would have been e-waste, more so than it already is because even the most simple repair was magic to people unconcerned with technology. I've dedicated my life to technology and engineering, and even then I'm basically an idiot when it comes to a lot of things. Most people are idiots at most things and good at a couple things.

I understand people are upset when their technology stops working. It stops working because the people who are experts in it, don't have the time or funding to keep it working, and the people who want it to work don't understand how it works in the first place, because if they did and really wanted their old hardware working, they'd develop a driver for it. People do this all the time, and it takes months if not years of effort from a handful of contributors to even begin to match what a well funded team of experts can get done when it's their day job.

There's a fundamental disconnect between those who use and purchase technology and those who make it. The ones who make it are experts in tech and idiots at most things, and the ones who use it are likely experts in many things but idiots with tech, or simply have a day job they need to survive and don't have the time to reverse engineer these systems.

Even in an ideal communist state, resources need allocation, and the allocation would likely still trend towards supporting the latest version of things that currently have the tooling and resources, who's power consumption is lower and speed higher. We might get longer lifespans out of things if we don't require constant improvement to serve more ads, collect more data, and serve HD videos of people falling off things, but the new technology will always come around, making the old obsolete.