1) Brandon, multi-volume epic fantasy presents unique storytelling challenges and unique demands upon a reader. You said in your essay that with The Stormlight Archive, "I didn't want to intentionally build a story where I relied upon reader expectations." But I assume you meant that in a specific rather than a global way: you do intend that subplots will get wrapped up eventually, that there is a main plot, that characters have arcs, and that the story has an ending...right?
2) If that's a valid assumption, then as a storyteller chunking a story out in ten volumes, how much do you worry about imposing the traditional limits of a novel on each volume? (i.e. Chekhov's Gun: "If you say in the first chapter that there is a rifle hanging on the wall, in the second or third chapter it must absolutely go off. If it's not going to be fired, it shouldn't be hanging there.")
For example, I wrote a scene for The Black Prism which was interesting in its own right and introduced a cool monster and a setting that I plan to use later in the trilogy—but it didn't accomplish anything necessary for book 1. It slowed the headlong rush to the end of the book; it looked like gratuitous worldbuilding. It wasn't, but a critic wouldn't know that until they read book 3—which I haven't yet written. So I cut it.
Would you have? Would you have cut an analogous scene in Mistborn 1, but not from TSA 1?
Are you writing these books so that each volume has that rousing, bang-up finish, or are you fine with a cliffhanger, content that the series must be judged as a whole? In a ten-volume epic, do you conceive of them as telling one story or ten stories? Or both? Or more?
The short answer to your first comment is a yes, you are right. The realization I came to while working on The Way of Kings was that I was so accustomed to writing self-aware fantasy in the Mistborn books that I was searching to do the same with Kings. While anyone can enjoy Mistborn (I hope) it works best as a series for those who are familiar with (and expecting) tropes of epic fantasy to come their direction. That allows me to play with conventions and use reader expectations in a delightful way. But it also means that if you don't know those conventions, the story loses a little of its impact.
But this is an interesting discussion as to the larger form of a novel. Is it okay, in an epic fantasy, to hang a gun on the mantle, then not fire it until book ten of the series written fifteen years later? Will people wait that long? Will it even be meaningful? My general instincts as a writer so far have been to make sure those guns are there, but to obscure them—or at least downplay them. People say this is so that I can be more surprising. But it's partially so that those weapons are there when I need them.
It often seems to me that so much in a book is about effective foreshadowing. This deserves more attention than we give it credit. When readers have problems with characters being inconsistent, you could say this is a foreshadowing problem—the changes, or potential for change, within the character has not been presented in the right way. When you have a deus ex machina ending, you could argue that the problem was not in the ending, but the lack of proper framework at the start. Some of the biggest problems in books that are otherwise technically sound come from the lack of proper groundwork.
In the case you mentioned, however, I think I would have cut the creature. Because you said it was slowing things down. There's an old rule of thumb in screenwriting that I've heard expressed in several ways, and think it works well applied to fiction. Don't save your best storytelling for the sequel. If your best storytelling isn't up front, you won't get a sequel. Of course, once you're done, you do need to come up with something as good or better for the sequel, otherwise it might not be worth writing.
For The Way of Kings, I've had to walk a very careful balance. I do have ten books planned, but I had to make sure I was putting my best foot forward for the first book. I had to hang guns for the later novels, but not make this story about them—otherwise readers would be unsatisfied to only get part of a story.