In theory, markets should regulate and competitors should be able to rise and provide products over never ending services and be successful. That doesn't seem to happen though. Is it because service designs expose real business costs of continued investment in R&D sort of deal, is it because such reliable products are bad for business (you provide something so good, reliable, and cheap you kill your own revenue stream) or what?

Similar to the concept in medicine that it's bad for business to cure vs good for business to treat something. From the consumer side, clearly you want cures but markets don't seem to function that way. Treating kidney disease is far less lucrative than providing a cure (if possible). Your only option as a business to find this appealing is to somehow look at the treatment option, roll all expected profit and revenue up, and tuck that into a "cure" option. If altruism in medicine isn't there, I don't know why we'd hope for it in something as artificial as electronics and software. I don't know what the solution is here if we want to continue to pretend markets self regulate.

I mean ... the market has spoken on this. It just isn't something that enough people value enough for it to have a positive return on investment. Like pretty much everyone else here, I think that sucks, but I also struggle to begrudge the preferences of people who just don't care about this.

So the question is just, is this a market failure in the sense that the market outcome is undesirable to society such that there's a strong case for regulation? Or is it a market success in that most people are getting what they prefer (cheaper products)?

If we were to get regulation passed on this, and prices for digital products and services were to rise, would people broadly cheer or jeer us? I think it's pretty clear that most people would be unhappy with this, at least above some level of price increase (which may be zero).

So to me the question is (like with all regulation) whether this can be regulated without increasing costs to people beyond the value of the regulation to society. Maybe so, but unfortunately I'm pretty skeptical that's the case.

If companies released just the hardware without the services, then we could have open source jumping in to provide those services, and things would be cheaper.

Unless of course you want to argue that without financial incentive nobody would ever do such a thing, and open source is a fairy tale that can't really exist ...

There is a reason that open source thrives in the niche of tooling for other software developers, and ... thrives less outside that niche. The target audience for software development tooling ascribes huge value to being able to read and modify the source of the tools we use. And for good reason, it's not just a principled stand; open source directly helps us do our jobs better.

But outside that niche, nobody really values this capability, and it's not a mystery why not. What good does open source do an accountant? Any given work-a-day accountant is never going to read or modify the source of their software. They just have no interest in that. They may have a computer enthusiast friend who has made a compelling case to them for why this stuff matters, so they may agree in theory, but they aren't going to do extra work or pay extra for this. It just isn't a direct enough impact for them.

In my experience markets dont regulate anything. Governments do.

If something is a service then there should not be a "buy" button, but a "lease" button.