The toolset is robust. The DAW by itself is insane. This "game", Dreams, has a fully functioning, fully scriptable DAW built into it and it's not even the main event. The potential this platform has as far as games go is very real... But this presents a bit of a contradiction. A mixed message. Here's a full dev system for making games, but it's something no one is really going to do. It's a complete contradiction.
I suppose we have a fundamental disagreement here. I don't think there's a contradiction here - having a ridiculously powerful toolbox doesn't inherently have anything to do with what should or needs to be achieved with it. Just because it's possible to create a 10-hour long nigh on AA quality game with it, it's not the end game goal here. That's not to mention that even though Dreams is a game itself, it doesn't mean the creation has to be a game, or that "full potential" means "full games" in my opinion.
The reality is, no matter how intuitive or accessible your tools are, making a game is *very very hard* and takes *many, many hours*. The only thing in reach for anyone who doesn't commit to a large project are scattered, incomplete thoughts, tech demos, and, perhaps very occasionally, a completed, very small game. Something akin to a small flash game on kongregate (not a large flash game, but one of the smaller ones. The vast majority of people making flash games were doing it because they could make money through kongregate/armor games sponsorships).
While you're obviously right about game development being *very very hard* and taking *many, many hours* (I unfortunately also have some experience there too, before I realized what a fucking pain game development really is), I think a crucial difference here is collaboration and sharing - something I shortly elaborated on in the post you quoted, but unfortunately edited it in too late for you to see it.
I believe the goal here is to cultivate a community that shares everything and anything. Not like you would share open-source libraries on github or sell full asset sets on the Unity asset store, but everything from the smallest objects to the biggest system templates you created would by default be available to everyone (I don't have the game yet and it's been a while since the beta so my memory is a bit fuzzy on the actual logistics here, though), with the important note that everything can just be dragged-and-dropped into your Dream. Everything is easy. The idealistic future here is that if you wanted to create, say, a small GTA clone, you wouldn't have to make everything from scratch yourself. From scratch it would be a ridiculous undertaking, but after a year or two the Dream library (or whatever it's actually called in the game) could very well have great implementations for third-person movement and combat, driving, various AI routines and endless amount of pedestrian, car, building etc. models, all of which you could essentially just drag-and-drop into your Dream in an evening or two.
At the risk of sounding dramatic, I definitely think introducing monetization - especially a harder type like an "asset store" sort where people would lock their content behind paywalls - would thoroughly corrupt the culture and the idealistic future would become an impossible future.
I know some of you may insist otherwise, but from the player perspective, scattered, incomplete thoughts and tech demos aren't going to be compelling for very long. The novelty of that wears off quickly.
To be completely frank, I don't understand why the novelty of playing "full" games that were created in a game would last any longer. Because that's the thing, here - the interesting part isn't playing a full game, the interesting part is playing a full game that was created with another game. After you've played a few of those, what's the reason to really play more? Because if you're willing to pay money for full games and you're looking to play some, why would you boot up Dreams and not... the PS Store or Steam or something, with their vastly bigger libraries and explore/discover functions made to actually find new games to play, and not just new cool things, whatever they may be.
And, of course, this leads to an important clarification on my part - I'm not against every type of monetization. I think being able to sell your games on PSN as completely separate packages is a great idea, and a.. Let's say light Patreon integration or something similar would be good, too. But I'm vehemently against any type of monetization that actively discourages sharing and collaborating, and I don't think money should be involved this early at all. Maybe in a year or two, depending on how the community and culture fare.