We are talking about a full computer that is small and cheap enough to control a single light bulb. We got here through unbelievable technological development.
Why would you want it to stop now, when the tech is both amazing _and_ rough around the edges? Better hope development continues and the end-user experience stabilizes to something much better.
I share your enthusiasm for how amazing our technology has gotten, but I'm also experienced enough to know that the technology will always be "rough around the edges" as you put it, because maximising profit doesn't require stable end-user experiences that are as good as they can be - they'll constantly be just barely good enough to make it into the shops, and then let the marketing take care of getting enough wallets to open up to make a profit on them.
Capitalism continuously fails to deliver on its promise of "may the best win, and may all be incentivised to be the best" because companies have found several loopholes in its assumptions - constantly introducing flashy new features so the basic faults don't get the airtime they should, advertising with the best psychological manipulation money can buy, and flooding the market with enough barely distinguishable models
that simple decision fatigue gets enough people giving up on research and buying the crappy models unwittingly.