I'm not making any assumptions about what the players are "allowed" to choose. But I am making an assumption about how stakes and consequences are established, namely, by the GM in authoring the setting/situation and then extrapolating from it.
To me, it is telling that not far upthread
@Pedantic said that if you rule out what I am calling railroading, there's nothing left!
Whereas to me, when I rule out what I am calling railroading, I see vast quantities of RPGing remaining: 4e D&D (with player-authored quests), AD&D (the way I played it from around 1986 to 1989 - I wouldn't use it anymore, because I've discovered better systems, but I know from experience it an be done), AW, DW, Burning Wheel, Agon, Prince Valiant, Classic Traveller (which I discovered a few years ago can be played as a type of PbtA precursor), etc, etc.
That's really not what I said, but I see your point. I'm trying to find the pivot point on which that definition sits in terms that are relevant to TTRPGs as I interact with them. A lot of your concerns are really esoteric from where I'm sitting, so it's not obvious from what you're saying.
Yes, I'm aware of that. I hold a different view.
I haven't said that the player would author that bit of fiction.
The player authors the action declaration, "I search the upper floor of Evard's tower for spellbooks". Now, this happened in BW, so the rule for the GM is "say 'yes' or roll the dice". The GM is expected to say "yes" if nothing is at stake (where what is at stake is relative to the players' evinced concerns for their PCs). In this case, there clearly was something at stake: Aramina, Thurgon's travelling companion, had brought them to the tower to find spellbooks, and that was why Thurgon was searching for them. So the GM called for a check (my guess would be Scavenging, though I can't recall for certain anymore).
You're focusing on a different part of the why we get here, but I think you actually are supporting my earlier proposition that the primary difference is the player's ability to dictate elements outside of their direct PC actions. Or rather, I would say that's the primary difference, and you're very concerned with how that's achieved precisely in a way that I don't think is particularly relevant. In the same way I could present two scenarios you'd both conclude are railroads under your terms, but I could point to one as providing ludic agency and one as not doing so, a concern I don't think you'd share.
If the check succeeds, then intent and task are realised: Thurgon finds spellbooks for Aramina.
So, this is player authored fiction. The player's action declaration, through a game mechanism, caused a change in the setting outside of their character's control. This is the direct translation of intent->result I was talking about earlier, skipping over intent as determining action declaration, to intent
as action declaration.
If the check fails - which it did - then Thurgon's intent is not realised. What he actually found were letters, that appeared to reveal that his beloved mother Xanthippe is, in fact, the daughter of the evil wizard Evard.
The player (me) did not author the fiction, but plah was not a railroad: the stakes and consequences are not being established solely by the GM. They are being authored having regard to my (the players') evinced concerns for my PC - his Beliefs (about Aramina and Xanthippe), his Relationships (to Aramina and Xanthippe), etc. To use the language of AW/DW, this is an example of the GM being a fan of the characters.
Same here, though more abstractly, in that the player's choices are a constrain on the GM, and there's a separate mechanism that requires the GM to say something. Essentially, it's just speeding up the plot hook process to happen on every given roll, with specific constraints on what kind of hooks can be offered.
Now, when you (
@Micah Sweet) say that play should not revolve around the PCs, I take you to mean that play should not play out in the fashion I've just described, and that if the GM has made a decision about what is in the tower (spellbooks, letters, whatever) then that's that. The players can learn about what the GM has decided is there; and the players can choose which "there" to poke around in; but the GM will not author fiction about what is there in response to the players' evinced concerns for their PCs.
Ultimately, this comes down to "does the player have control of the setting beyond their own action declarations?" as the differentiator between what you're calling railroad, vs. non-railroad play, and once that's established, you're adding rules to ensure the resulting narrative is interesting, both by constraining the player's control and prompting the GM to specific kinds of declarations. Which, bubbling outside enmity aside, is probably why there's so much friction when these things get presented as GMing advice. They aren't, really. They're rules that enable an interesting narrative to form under a different basic set of assumptions about what the activity at hand is.
The sort of play that I have described in the previous paragraph is what I regard as a railroad. (Again, I repeat this caveat: if essentially we're playing a wargame, like Isle of Dread or White Plume Mountain, then the whole logic of things is different, and the characters are just player pawns. That's not a railroad, but it's not really a game with characters at all in any meaningful sense.
Well that's just offensive. I routinely go out of my way not to say what you're doing isn't a game, something I would do as a matter of normal jargon in other contexts, because it's obviously not appropriate or helpful here, and "game" has a more expansive definition in this context than it would elsewhere. You could try a little harder with "character." If you're setting your own goals for the pawn, repeatedly, I don't know that it actually needs anything else to be a character, except a retelling of events.
With that in mind, you're doing some lumping that's problematic:
That's because, as best I can tell, you count it as not railroad if the players are allowed to choose what actions they declare in relation to the GM's material, even if the consequences of those actions, and what's at stake in them, is all a function of the GM's material (and the GM's "logical" extrapolations from it). This could be anything from "What's behind the door" to "What will this NPC say if we offer a bribe?"
You're conflating the GM's role in creating a setting and the results of action declaration. One is a process of authorship, and the other is mechanistic. Perhaps a better way to put this, is that I don't actually think you need any more agency than that afforded by non-TTRPG games to achieve a state where events happen outside the GM's control. Would you say that Isaac Vega authored the events of a game of Dead of Winter? He wrote all the cards, and laid out all the available actions, but there are still choices made by the players over the course of a game that produce different outcomes. He hasn't written every possible game state, just the mechanisms that allow them to happen, and a TTRPG already has a leg up on that, in that it's unbounded in time, has a much wider set of action declarations available, and uses a flexible human brain instead of a fixed board.
It is, however, certainly possible (and sadly not at all unheard of) for a GM to constrain action selection down to a preferred set, and to simply dictate possible outcomes. I can conceive of a TTRPG play that offers less agency than Dead of Winter, and it seems unreasonable to say the two things are more similar than different.
It shares a basic similarity of form with the sort of RPGing I enjoy, but in its details is a completely different activity.)
Are we doing the same hobby? Because I'm increasingly not sure we're actually doing the same hobby.
