I’d give it a B-, although I don’t think it’s Apple that did it, I just think it’s the direction of the industry.
The main thing that has unfolded since then is that SmartThings is now going in an entirely new direction for the things that made it special, particularly the use of custom code. There’s a whole thread discussing that, so I’ll leave the details for that one:
But SmartThings as it was in the fall of 2016 will be gone within another year.
Meanwhile, plug and play systems with simple rule systems, like HomeKit, will exist and probably be in more homes then SmartThings was at the end of 2016.
And the new SmartThings in 2019 is going to look a lot more like HomeKit 2019 then it does like SmartThings 2016.
I suspect it will still support a lot more devices than HomeKit does and I hope it will continue to have a more complex rules engine option, but we won’t know for sure until the end of this year.
There’s a lot of uncertainty yet, on both sides, but market forces are demanding more reliability and simpler installation. And Samsung has made the decision that, like many other brands, the way to get to those is to limit the number of choices, or at least “offshore them” by requiring individual developers to take the responsibility of hosting custom code.
HomeKit runs completely locally except for voice command processing. Samsung is Not taking SmartThings in that direction ( another reason I only give it a B minus), but it is greatly simplifying how its cloud will be used, and limiting what will run there.
So I think the OP recognized some important industry trends, but maybe not the instigators of those trends. Apple was just riding the right wave. 