Right, I understand that. So your model fits closely enough to what most of us would acknowledge as the basic structure of the game. As I said, I don't have a problem with the approach....more with the implications you've made along the way in the discussion about other methods being more railroady.
I think, with the bowl issue, the difference pemerton (and others?) is trying to illustrate is the difference in how the DM approaches it. So, first the player states something like, "I look for a bowl or container to catch the blood in." Then we split. [P]emerton's preference is that this is now a potential new element to the fiction -- he doesn't know if there's a bowl, but he'd like to find out. The player is asked for a check, and the success or failure of that check establishes if a bowl or container exists or does not. In this case, the existence of the bowl is an element the player is trying to establish, and the check is to see if this is the case.
On the other side, the side pemerton is calling railroading, it starts the same way with the player declaration, but instead of the check seeing if a bowl exists, the DM decides whether or not a bowl exists and how easy it is to locate the bowl. For you, the existence of the bowl is determined as 'does' and the likelihood of finding it is '100%' so you just say, 'sure, there's a bowl on the nightstand with some random coinage and a bubblegum wrapper in it.' But another DM using this method might determine there is no bowl. Another might think there is a bowl, but it's under the bed and not easily noticed, so the check is to see if you find it in time. But all of these start with the DM determining the answer to the question 'is there a bowl?' and move forward. What I gather from pemerton is that this act is the railroad (again, I strongly disagree with this use of the term) because it's a function of the DM forcing the fiction instead of allowing it to be a collaborative event. With pemerton, it's more important to acknowledge the player's contribution to the story, and the DM's job isn't to say yes or no or determine the answer, but to provide a challenge in the form of a die roll that will determine the success of the player's authoring of new fiction (in this case, a bowl; not all fiction is exciting).
That's the gist that I pick up from manbearcat, pemertion, and Campbell. While it looks superficially like the standard presentation of play (present, declare, narrate), it differs fundamentally in how the narration is authored. In the case of the bowl, pemerton's method is that the player has authored the bowl, so he's just narrating what the player established. With the other (railroady, as permerton says) method, the DM determines the fiction and only narrates the outcome of things he (the DM) is uncertain about.
I'm not sure I'm explaining that last part very well. I see it as a fundamentally different approach to the game, though, and one that fits with the presentation of other games and how they operationalize creation of the story. And, again, it's a fine playstyle, a fine theory of game, one I have no issues with (although I occasionally fail to recognize it), but I don't think the DM authoring style is a railroad, per se, it just approaches the creation of fiction differently. You can use the DM as primary author method just fine and maintain player agency in the world. It's just the domain of player agency that shifts. Railroading is the removal of player agency -- nothing they choose matters, they go the same place no matter what.
