Failing Forward

[Umbran]

When I talk about procedural play involving the adequate deployment of skills, I'm not talking (just) about the mechanics of skill checks. The sort of play that GUMSHOE emphasises - visiting the right places, declaring the right action/skill use, collecting the information and solving the puzzle - I would count as an example of procedural play.

Fair enough. The game is certainly procedural, in the sense that TV shows are procedural.

To note, however, GUMSHOE doesn't require that the players actually figure out what the *real* procedure would be - the player doesn't have to know police evidence gathering procedures to succeed. The game calls for the characters to be in the right place, and have the right skill to use. In, for example, old school D&D play, we relied on player expertise, and if the player didn't explicitly state the character was searching, say, the legs of the bedstead, a secret compartment in one of them would never be found, but that's not required here.

In GUMSHOE, if the character with Forensic Anthropology says, "I examine the body," that's enough. They don't even really have to specify which skill they are using - so long as they have an applicable skill, and what they narrate is reasonable to include a particular skill, we can assume it is applied, and they get the appropriate information.

The question of what we do once the puzzle is solved has potentially more dramatic/character-driven dimensions to it, but if the GM has determined the answer to the puzzle in advance of play, then the dramatic elements are not likely to be adapted to and expressive of the dynamics of actual play in the sort of manner that I prefer.

Again, fair enough. A person's preferences are what they are. I wonder, though if much of what you are labeling as character-driven is really player-driven, which is not the same thing.

When I hear the phrase "character driven", I think of the literary sense, in which the piece is focused on in internal changes and conflicts in the character. It is *not* character driven in the sense that the events are dominated and controlled by the character's choices, nor that the characters are in charge of their own destinies - you can have a character-driven story in which the characters are tossed upon the seas of fate, so long as the focus is on the internal life of the character, and how they feel and react to events.

This is in contrast to the plot-driven story, in which the details of events are the focus of the piece.

You can thus have a player-driven game, that focuses on the directions the player wants things to do, that is overall plot-driven... if the player is interested in making things external to his or her character happen. Meanwhile, I can have a significantly pre-authored piece, that as fiction is character-driven, in that the pre-authoring is designed to yank the character's emotional chains, and produce conundrums the player reacts to.

This latter, I find (definitely in my own, anecdotal experience) is true for those players who want to reach emotional immersion in characters. The meta-level of thought required for player-driven play tends to keep them non-immersed. So, the GM has to know where the PCs in coming from, and put things into play that resonate with their known issues. A great deal of human drama and emotion arises from the fact that the universe is often unyielding and intractable, and that is hard to model with something that is highly player-driven, and thus yielding and tractable.

I find the example of the spectrum of playstyles seen in this discussion interesting.

A great many folks come down on the side that things like Fail Forward, or dice fudging, or other narrative influencing techniques are badwrongfun, for a variety of reasons generally amounting to, "a form of objectiveness to the game reality must exist down to the individual task resolution." I often find myself in alignment with you in these discussions - play of the moment, to me, is generally pretty open to manipulation.

But here, we find we part ways - you seem to take it to the other extreme. While you don't call it badwrongfun for others, even having objective reality exist on the larger, longer scale is detrimental to your play experience, while I find the presence of a significant number of pre-determined facts to be useful.

I think I know of an analogy I see in character creation...

I know a great many people who, in character creation, want to be given a "blue sky" situation, where they GM says, "Make anything you want, the sky's the limit!" without any direction on where the campaign will go, themes, or the like. I, personally, don't work well in blue sky mode. When i am force to make a character in such situations, the results are... uninspired. However, give me a framework, a few restrictions, and it is an entirely different story. If creating a character were writing poetry, some folks are best in free verse (which I often really don't understand), but I am a master of the more structured haiku and sonnet forms that most folks find too restraining.

 

[Balesir]

One major benefit of pre-authorship is pre-design. Sometimes, you actually want to make sure what you are presenting is really thought through before players encounter it.

This is (collectively) a good point, and I think it has some degree of merit, but I don't think it is particularly clear cut. For example...

Games that are tactically deep typically need fairly carefully considered design of the tactical challenges, resource depletion rates, and the like, to keep them challenging, but not overwhelming. But, you can't pre-design the tactical challenges without knowing what's going to come up - that means you need pre-authorship of much of the material. IMHO, nobody should be "winging it" for mid to upper level D&D 3.x play focusing on combat, for example. Meanwhile, a game like FATE, or Cortex+, that isn't so tactically detailed, can be authored on the fly easily, because the number of tactical bits needed to make a worthy challenge are far fewer.

While really full-on tactical challenges might need some careful design, I think more modern tactical rule sets, both RPG and wargaming, can be made pretty tactically rich with very little setup or using setup tools designed to work quickly with the rules they are attached to. Several folk have said they can run encounters on the fly with 4E, 13th Age is certainly pretty well set up for it and, on the wargames side, the "De Bellis Antiquitatis" series can be played with very short setup times and quick action. So, a qualified "pre-authoring/pre-design useful" but not, I think, necessary - at least not for the whole game.

Cogent mysteries also need a fair bit of pre-authoring. If the goal is for the players to use reasoning to figure out what amounts to a big logic puzzle, you have to pre-author that puzzle, or you are likely to become either inconsistent, or not put in enough information for the players to work out the puzzle, both of which will lead to highly frustrating play experiences.

I can see that if you want to use "clever" clues as beloved by mystery writers that pre-design of the "truth" behind the mystery could be useful, but, again, I don't think that mystery and discovery are impossible with a "no myth" approach. I think Dungeon World stories and Universalis show this most clearly, to me. The approach is actually similar to the scientific method, except that repeats of the same test are generally assumed (rather than "happen to have") the same outcome.

In other words, a "rule" has to be that, once some "observation" is made, that observation is assumed to be "the truth", and any repeat of the same essential experiment will yield the same result. Investigation proceeds by a process of identifying untested possibilities, working out an in-game way to test them, and then resolving the (randomised) outcome of those tests and considering what the outcome of the test means for the revealed "truth". When you have eliminated that which the dice say is untrue, whatever remains must be the truth.

Note that this method is not without the benefits (and I agree that there are some) of constraint and limitation. It's just that, instead of being set at the beginning (and not revealed to all participants), the constraints are built up gradually as "facts" are established in the fiction.

Note how both of these are situations where play has a significant center around use of logic.

Yes, I see that, but I don't think that "discover as you play" style play need be in the slightest bit removed from the application of logic.

Further, I wonder whether an analogous case might be made that more effective emotional or aesthetic impact can be achieved through the use of prepared material than via pure improvisation? I guess the question I would narrow down on is "at what point does an increasing focus on individual design and preparation make the whole thing an entertainment rather than a collective creative activity?" The "in the moment" end of the spectrum has, if nothing else, the benefit of being more of the latter than the former.

As almost an aside, I have generally found that the use of "clever clues" in RPGs has a nasty tendency to fall flat. I think the trick is that mystery writers have the singular advantage of both controlling the main investigators and creating the "clever" clues; they can thus be assured that the "truth" revealed by the "clever clue" will, at some point, become evident to the stalwart detective...

 

[Balesir]

Gandalf had no say in whether or not Bilbo found the one ring. When he showed up at the beginning of Fellowship and threw the ring into the fire, he was inside a story controlled by facts outside of himself. If he had suddenly starting talking to the reader and announced that he decided it was not the one ring, that would be a very unsatisfying development.

I think you set up a false comparison, here. Let me draw another picture of the "DM of the Rings" as it might play out:

- Either following previously played-out investigations or because the GM agreed with the player of Gandalf as part of character creation that Gandalf had strong suspicions that the "ring of invisibility" that Bilbo had found was, in fact, the One, Gandalf has such strong suspicions. Strong enough suspicions, in fact, that to risk damaging a run of the mill "invisibility ring" to find out the truth once and for all.

- Gandalf throws the ring in the fire. Dice are rolled to see what happens. Because of the evidence gained beforehand and/or because of the steer given to Gandalf's player at game start, the odds are stacked pretty well to the "One Ring" side, but it's not a foregone conclusion.

- Either the ring is the One, and the players must decide where to go from there, or it isn't - which raises more questions about why the Great Darkness is once more stirring. Sauron is still seeking the One with all his power, after all...

Not only does this strike me as a perfectly good way to go with the play, it also seems to me to be much, much closer in immersive feel to the fictional scene that we might imagine in Bag End that evening. After all, Gandalf, though he had strong suspicions and good reasons to think that the ring Bilbo found was the One Ring, he didn't know it for certain. Without the real risk of an contrary result, the tension that we imagine being felt by the characters on that night evaporates. We already know, where they did not, what the outcome will be.

 

[Umbran]

I can see that if you want to use "clever" clues as beloved by mystery writers that pre-design of the "truth" behind the mystery could be useful, but, again, I don't think that mystery and discovery are impossible with a "no myth" approach. I think Dungeon World stories and Universalis show this most clearly, to me. The approach is actually similar to the scientific method, except that repeats of the same test are generally assumed (rather than "happen to have") the same outcome.

I think you miss my meaning. The issue at hand isn't about it being "clever". Merely that the clues are logically consistent and actually eliminate all but one possibility.

In Atomic Robo, a FATE-based game, there's an excellent procedure for the PCs to figure out what is causing a particular weird science phenomenon, and in the process generate an Aspect they may tag to help deal with it. The players determine a series of "facts" (usually, but not necessarily, based on things they've witnessed about the phenomenon), and then come to a "conclusion" that is consistent with those facts. The GM does not need to decide how the phenomenon works, as the players will generate the reason it works for the GM.

So, for example, in a recent arc of that game I ran, I put forth a swarm of giant bees the PCs had to deal with. They used the fact that they had apparently escaped during a lightning storm, a completely made up piece BS about how bees navigate that is patently not true in the real world, and something else I cannot recall, to come to the conclusion that the bees were vulnerable to strong electromagnetic fields. They then sacrificed an electric car to kit-bash it into an electromagnetic beacon to attract and capture the bees. As weird science creativity, it worked just great, and generated a solution to the problem I would not have considered. It even told me how the bees had escaped their confinement, which I hadn't determined that before play began.

This functions just so long as the final result can be *anything*, unconstrained. For a murder mystery, for example, so long as you can generate a suspect the PCs have never actually met to be the ultimate bad guy, this will function - if you can always claim there's a Mr. Whithers that has not been seen before the PCs unmask him, you are fine. However, if they try to pin it on a specific person they've already seen, unless they are *very* careful (as in, are taking care at a level I simply don't expect is practical) the facts they may generate may not be consistent - the butler will have been in the study, and the kitchen at the same time, with the candlestick and the pipe wrench, respectively. Or, those "facts" will not actually converge and narrow the field to any particular conclusion at all!

Moreover, this sort of procedure does *not* give the players the satisfaction of solving a logic puzzle! The mental act of creation is not the same as the mental act of deduction, and the players can tell the difference!

Further, I wonder whether an analogous case might be made that more effective emotional or aesthetic impact can be achieved through the use of prepared material than via pure improvisation?

I don't think we can generalize. I think some people will find the emotional or aesthetic impact of prepared materials greater, and others will find improvisation more effective. I don't even think we can state it clearly for a single person, as we cannot really separate the impact of the source of the material from the impact of the content of the material, from the impact of the presentation of the material - it could be more a question of the GM's theatrical skill and/pr understanding of the players than the source of the material.

I might go further to say that while Pemerton has found what works supremely well for himself and his group, his group's tastes are likely in the minority. Likewise the folks who argue against his positions from the polar opposite point similarly have groups that have been selected for those particular tastes, but are likewise in the minority. The "truth", such as it may be, may thus lie in the middle - the most generally effective approaches are apt to be hybrids, not purebreds. Individuals may like something specific, but mutts are pretty broadly beloved

Or, we might posit that generation of material is like my generation of characters - maybe one GM is most inspired by improvisation, another GM is inspired by the process of authoring beforehand, and that is what generates the impact.

As almost an aside, I have generally found that the use of "clever clues" in RPGs has a nasty tendency to fall flat.

If I am getting what you mean by "clever clues", that's not what Ashen Stars operates on. They do not recommend you find one key point that's arcane or obscure in order to solve the case. In the adventure I'm currently running, there are no fewer than four different ways to discover the source of the trouble the PCs are contending with. It is by no means hidden in some particularly clever minor tidbit.

 

[pemerton]

a "rule" has to be that, once some "observation" is made, that observation is assumed to be "the truth", and any repeat of the same essential experiment will yield the same result. Investigation proceeds by a process of identifying untested possibilities, working out an in-game way to test them, and then resolving the (randomised) outcome of those tests and considering what the outcome of the test means for the revealed "truth". When you have eliminated that which the dice say is untrue, whatever remains must be the truth.

Note that this method is not without the benefits (and I agree that there are some) of constraint and limitation. It's just that, instead of being set at the beginning (and not revealed to all participants), the constraints are built up gradually as "facts" are established in the fiction.

<snip>

I don't think that "discover as you play" style play need be in the slightest bit removed from the application of logic.

I see this as analogous to the point I made upthread about extrapolating to fictional facts that have never been authored, by working within the context and constraints provided by what has been authored.

I wonder, though if much of what you are labeling as character-driven is really player-driven, which is not the same thing.

When I hear the phrase "character driven", I think of the literary sense, in which the piece is focused on in internal changes and conflicts in the character. It is *not* character driven in the sense that the events are dominated and controlled by the character's choices, nor that the characters are in charge of their own destinies - you can have a character-driven story in which the characters are tossed upon the seas of fate, so long as the focus is on the internal life of the character, and how they feel and react to events.

This is in contrast to the plot-driven story, in which the details of events are the focus of the piece.

You can thus have a player-driven game, that focuses on the directions the player wants things to do, that is overall plot-driven... if the player is interested in making things external to his or her character happen. Meanwhile, I can have a significantly pre-authored piece, that as fiction is character-driven, in that the pre-authoring is designed to yank the character's emotional chains, and produce conundrums the player reacts to.

This latter, I find (definitely in my own, anecdotal experience) is true for those players who want to reach emotional immersion in characters. The meta-level of thought required for player-driven play tends to keep them non-immersed. So, the GM has to know where the PCs in coming from, and put things into play that resonate with their known issues. A great deal of human drama and emotion arises from the fact that the universe is often unyielding and intractable, and that is hard to model with something that is highly player-driven, and thus yielding and tractable.

In your description of the pre-authored character-driven game you still talk about the player reacting to the challenges into which the GM frames his/her PC.

Unless those reactions are what I would call mere colour - the player emotes his/her PC, but doesn't actually declare action resolutions that both express that emoting (and so would be different were the emotions different) and shape the direction of ingame events - then we have something that is, to at least some extent, player driven.

In the "pure emoting"/"mere colour" variant, I see two issues. One is a matter of taste, and so maybe not of any consequence to anyone but me: generally I don't find play where the players' emoting of their PCs makes no difference to events very satisfying. An exception is a well-run Cthulhu one-shot, where the emoting is fun (as you play out your PC's baltherings to Nyarlathotep, or whatever it is) and the play of the game is significantly about generating that colour. And even in these, often one's final choice for a PC might change the way in which the world comes crashing down as a finale.

The second issue is more practical: if the player's choices for his/her PC don't shape events in a significant way, then unless the GM is an incredibly engaging author there areincreasing prospects of a gap opening up between what the GM has in store for the PC, and what will actually move the player who wants to be immersed in his/her PC. I have seen this happen more than once in highly GM-driven, pre-authored campaigns.

Turning these concerns into a positive statement about RPGing: the scope for the player to choose (by way of immersion or inhabitation of the character) how a protagonist responds to the challenges s/he is confronted with, and then have those choices both reinforced and put under the microscope by the framing of new challenges, is something that RPGs have that seems fairly unique to me. Turning the RPG into just another device for the players to hear someone else's story seems not to engage with that distinctive feature of RPGing.

I know the above has been framed in strongly contrasting/absolute terms, and as you say actual play in many if not most cases will involve mixes of approaches and techniques. That's even true for my BW game: I am using my GH maps (I think I have all of them from the early 80s through to the 3E Gazetteer), without worrying too much about minor variations across eras. So when one of the PCs has, as part of his backstory, knowledge of a ruined tower in the desert foothills we (the player and I) are committed to locating that tower somewhere on the Abor-Alz map.

So I'm not intending to advocate for a technique in any sort of purist or absolutist fashion. It's more that I'm trying to explain how a certain approach actually works in practice, why it has some merit, and (most importantly) why the frequent identification of a "real, living, breathing campaign world" with a very heavy dose of GM pre-authorship is a fallacious one. You can have the former without the latter.

 

[Balesir]

I think you miss my meaning. The issue at hand isn't about it being "clever". Merely that the clues are logically consistent and actually eliminate all but one possibility.

I don't think I miss your meaning, but I need perhaps to explain my own a little more completely to make that clearer (one way or the other).

When I talk about "clever" clues what I have in mind are clues that, when looked at a particular way or together with other available information, are supposed to point unambiguously to one specific perpetrator. I almost never come across this in real life - the usual sort of "clue" is one that eliminates one suspect from the list of possible suspects or adds a suspect to the said list. As a result I think of specifically incriminating clues that are found in combination as rather contrived or "clever"; I don't really think they feature in real investigations at all often, but that does not mean that they could or should not be explored in the context of an RPG. Their usual abode is logic puzzles and murder mystery books, so the desire to achieve their extension to RPGs seems quite natural, in fact.

Now, the reason I think this sort of "clue" structure tends to fall flat is that players either don't notice the clues (because they are not as evident as their author imagines them to be), or that they are not as unambiguous as their author believes them to be (or, at least, are not seen as unambiguous by the players). This can either lead to baffled and frustrated players (and characters!) in the first case - avoided by mystery writers by simply having the "genius" investigator notice the clues which they themselves have planted - or by the players/characters reaching the wrong conclusion in the second case - avoided by the mystery writer by having their primary character just as blind to the ambiguity as they are, which is kind of hard to avoid doing...

Getting the right answer is satisfying, of course, but getting a "wrong" answer can be not only "un-fun" but also intensely frustrating if you can see ambiguity that the author is apparently blind to*. I think it is also unrealistic, in the sense that what is generally left at the end of an investigation is some element of ambiguity. You might be wrong, but you will probably never know - although some investigations are "proved" wrong years or even decades after the fact.

So, do I think there is a type of game here that can only be done with pre-authoring? Yes - but it's only a subset of the "investigation" genre, and I think it's chancy at best to get "right". For some, the effort required to get it "right" may very well be worthwhile - good luck to them. To me, it seems like too much effort/risk for too little gain.

*: Edit to add - I take it as given that one key feature of this sort of pre-authored mystery is that you can know absolutely at the end of play whether or not you were "right". If this is not a desired feature, then I see much less benefit from the pre-authoring.

 

[Umbran]

In your description of the pre-authored character-driven game you still talk about the player reacting to the challenges into which the GM frames his/her PC.

Unless those reactions are what I would call mere colour - the player emotes his/her PC, but doesn't actually declare action resolutions that both express that emoting (and so would be different were the emotions different) and shape the direction of ingame events - then we have something that is, to at least some extent, player driven.

This is part of why I noted definitions of "character driven" vs "plot driven". These terms come from media - there is no "player driven" novel or movie out there.

But, the fact that we have a "player driven" option, in which the character drives the action, and the player has an avatar of a character, that "character driven" means "character drives events". It is a drift in language use I was trying to elucidate here. "Character driven" has little to do with how events unfold. Character driven stories are rather the opposite - they are less about how events unfold, and more about how *character* unfolds.

Which is to say that, in the terms I'm using, "character driven" and "player driven" are more orthogonal than we might at first think.

The second issue is more practical: if the player's choices for his/her PC don't shape events in a significant way, then unless the GM is an incredibly engaging author there areincreasing prospects of a gap opening up between what the GM has in store for the PC, and what will actually move the player who wants to be immersed in his/her PC. I have seen this happen more than once in highly GM-driven, pre-authored campaigns.

Yes, that is a possibility. But nothing is perfect. It is also a possibility that the players choices shape events, and the result is still unsatisfying, however. When you were a kid did you ever think, "Gee, I'd love to have that toy!" only to find that the toy, once gained, really wasn't all that fun? Or, it is also a possibility that you have a group of players who choose things that wind up in great conflict, and they aren't good at negotiating among themselves how to go, leading to an unsatisfying experience, and so on.

Ultimately, "I get to author things," is not a guarantee of anything, including player engagement. Even Stephen King writes stinkers, kinda frequently, even! Robert Jordan, given his head with little editing (I suspect because he *married* his editor, leaving her in a position with major conflict of interests) ended up with the quality of his work degrading rapidly, with lots of meaningless filler and very little action or development of character - entire novels of mostly stasis.

So, there are pitfalls on all sides. Nothing guaranteed.

Turning these concerns into a positive statement about RPGing: the scope for the player to choose (by way of immersion or inhabitation of the character) how a protagonist responds to the challenges s/he is confronted with, and then have those choices both reinforced and put under the microscope by the framing of new challenges, is something that RPGs have that seems fairly unique to me. Turning the RPG into just another device for the players to hear someone else's story seems not to engage with that distinctive feature of RPGing.

Yes. But you know what? I find the concern to be largely a boogeyman. Rare, indeed (IME) is the GM who, while pre-authoring, isn't also taking player thoughts into account. Having, for example, a pre-authored mystery (past events in which the PCs took no part are fixed, but resolution of the remaining conflicts of the situation not fixed) doesn't seem like it is at great risk of missing out on this particular distinctive feature. It is only the past which is pre-authored, after all, not the future. Other than having some expectation that the players may well figure out what's going on, there's not much preventing having the player choices given spotlight. And the pre-authoring in this example is only on a case-by-case basis, not pre-authored for an entire campaign.

So I'm not intending to advocate for a technique in any sort of purist or absolutist fashion. It's more that I'm trying to explain how a certain approach actually works in practice, why it has some merit, and (most importantly) why the frequent identification of a "real, living, breathing campaign world" with a very heavy dose of GM pre-authorship is a fallacious one. You can have the former without the latter.

Being someone advocating mixed-approaches, you don't really need to convince me. All things are useful, when applied thoughtfully.

 

[BryonD]

I think you set up a false comparison, here. Let me draw another picture of the "DM of the Rings" as it might play out:

- Either following previously played-out investigations or because the GM agreed with the player of Gandalf as part of character creation that Gandalf had strong suspicions that the "ring of invisibility" that Bilbo had found was, in fact, the One, Gandalf has such strong suspicions. Strong enough suspicions, in fact, that to risk damaging a run of the mill "invisibility ring" to find out the truth once and for all.

- Gandalf throws the ring in the fire. Dice are rolled to see what happens. Because of the evidence gained beforehand and/or because of the steer given to Gandalf's player at game start, the odds are stacked pretty well to the "One Ring" side, but it's not a foregone conclusion.

- Either the ring is the One, and the players must decide where to go from there, or it isn't - which raises more questions about why the Great Darkness is once more stirring. Sauron is still seeking the One with all his power, after all...

Not only does this strike me as a perfectly good way to go with the play, it also seems to me to be much, much closer in immersive feel to the fictional scene that we might imagine in Bag End that evening. After all, Gandalf, though he had strong suspicions and good reasons to think that the ring Bilbo found was the One Ring, he didn't know it for certain. Without the real risk of an contrary result, the tension that we imagine being felt by the characters on that night evaporates. We already know, where they did not, what the outcome will be.

This is all cool.

I would hate it.

There is zero perception on my part that this Schrodinger's aspect of whether it was or was not the one ring was ever in play. I have never discussed the books or movies with anyone and received the slightest indication that they felt that a character not knowing a truth within the fiction made that truth in doubt to the larger story.
I want the experience of being in the story that way.

Again, importantly different goals and tastes with massive implications.

I'm not saying yours is in any way flawed. I never have.
But I think rejecting alternatives is a bad approach for discussing ideas and finding common ground.

 

[BryonD]

This strikes me as confused about the techniques under discussion.

The players in the examples discussed in this thread (eg the mace episode from my BW game) are not authoring anything. The GM is determining the relevant backstory, including the presence of the black arrows in the ruins of the brother's private workshop.

No, I understand quite well and you seem to be simply obfuscating the point.

Whether the DM or players make the changes is completely beside the point.

Again, to me this jarring inability to get what I'm saying points to the preconceived notions and inability to look at it differently, which results in this "talking past other people".

 

[BryonD]

First of all, any opportunity to bring in DM of the Rings to the conversation should be celebrated heartily.

The question of pre-authoring versus Story Now / Just-in-time GM-ing / mutable fiction is obviously not a binary. In fact, I'm a strong believer that diligent, coherent pre-authoring is a necessary precursor to running a successful campaign. To me it's much easier for the GM to later break that pre-authoring when needed if they have a strong grasp on how a given "break" will spill out into following frames.

The shift to "Just-in-time GM-ing" happens more directly in play. It's a reaction on my part to trying to be more open and flexible to player desires. And I know for me it has worked wonders in building the types of campaigns that I enjoy. My 14-month long Savage Worlds fantasy campaign was a direct result of a dedicated commitment to not having any "end game" in mind, but to "scene frame" based on what the players were giving me, with just the right amount of pre-authoring to make the frames coherent.

To follow up on the hypothetical Lord of the Rings example:

If I was the GM, the nature of the One Ring would be set in stone. But let's say the player running the "Frodo" character came to me and said, "What if I'm not entirely sure my uncle Bilbo is telling me the truth?" Maybe it's because he wants wants to explore something different in his character than "tragic heroism," so he imagines up that his uncle Bilbo isn't a good guy, but is in fact manipulating him.

The Ring is still the Ring, but now the Frodo character is exploring an entirely different set of fictional circumstances to react to / play against.

Would I as a GM be willing to change the fiction to potentially give the player what he wanted? Prior to 2010 or so, the answer would be a definite "No. That's not how the story is set up." Now? I'd strongly consider it.

Certainly, and I don't dispute this.
But I don't see you saying anything I didn't already agree with anyway.

As you said, the One Ring *is* the One Ring. The players choices may have HUGE impacts on what happens with that fact and how the story proceeds. And numerous new facts may come to be as a result. But the "pre-authored" facts remain set.

I already offered that a group could elect beforehand to play in a world where the one ring was *not* the one ring. The game canon can be as broken away from foundational literary canon as the groups wants.

And, obviously, if you go back and review a game after the fact, the number of "facts" that were "pre-authored" will be very small compared to the number of facts which were resolved "just in time". So again I strongly agree with you. It is certainly not at all the case that being strongly pro-preauthorship shuts out "just in time" as the main part of the game.