Maybe it's "Authored" vs. "Emergent"

[kyoryu]

We keep talking about sandbox and railroad/linear games. But I think maybe that's not the key point.

Maybe it's "authored" vs. "emergent".

In an authored game, the GM or module author write the story of what will be happening. This doesn't mean it's exactly linear, though it can be. There can be optional branches, there can be some shuffling of what order things are done in, etc. But the key is that all the things that are done, all of the situations handled are things that were predetermined.

An emergent game, on the other hand, doesn't have that. Sure, there can be some initial elements, but what the players do is fundamentally unknown, and how the world changes in response to their actions is also fundamentally unknown. Players can do basically whatever they want.

This doesn't mean that authored games have no agency, though they have very limited or no agency in specific areas - usually in terms of what specific encounters/scenes happen. Sure, you can change some aspects of hte result, or you can have side-effects, often significant, on the world, but you're really going to mostly go along the same path as others - you might do things in a different order, or skip a few things, but you're mostly going to do the same things.

This has advantages! Since the GM knows what the players are going to be doing, the GM can prep very cool, detailed, prep-intensive things for the players to do - custom actions, cool terrain, super balanced encounters, etc. And if the players are mostly concerned about that kind of stuff, it's a great way for the GM to give them what they want.

But fans of emergent games want different things - they want their decisions to matter. They want the game to take on a different shape than it would have if they had made other choices. They want to be able to solve the problems their own way. They want to do things that the GM didn't plan for, and they want to make that take the "story"/game in a way that the GM could not have predicted.

This is great for players for whom the major point is "how do I react to this situation". It's not as great for players who really want the resolution of the encounter to be the major focus, since by definition it's not going to result in as detailed/designed/tuned encounters.

I think this could be more useful, and I tried to keep the distinction as neutral, and point out the value of each side of this, and how they mutually oppose each other - the design of each serves certain purposes, but makes the opposite style harder to achieve.

[kyoryu]

Is a dungeon emergent or authored?

Maybe

A dungeon could be either - early megadungeons were highly emergent - any given session, the players would choose what to do and where to go. And while the dungeon and its contents were set, how the dungeon and its denizens responded wasn't.

A lot of more modern dungeons, however, are more authored - individual rooms are individual encounters, and have little effect on each other.

[KorvinStarmast]

Maybe it's "authored" vs. "emergent".

Interesting way to frame it, but there's still a middle ground somewhere, isn't there?
Maybe, "it's still a continuum, not an either/or choice" is a better way to get my thought across.

[kyoryu]

Interesting way to frame it, but there's still a middle ground somewhere, isn't there?
Maybe, "it's still a continuum, not an either/or choice" is a better way to get my thought across.

I'm not sure.

I do think they're fundamentally different, and I think every attempt at a "continuum" has basically been "I'm going to give you some leeway in a fundamentally authored game". Which.... is still an authored game, from the perspective of someone that prefers emergent games.

So, I think there's a fundamental difference. Either you know what's going to happen, or you don't. In Mass Effect 1, everybody's game ends up facing down Saren on the Citadel. In Civilization, everybody's game ends entirely differently, and different events will happen. Mass Effect has a cool story and setpiece encounters, Civilization has an unpredictable situation where your decisions ripple out and have long-lasting repercussions.

I mean, you can have elements that are authored in an emergent game, but those are usually primarily the initial situation. Like, running an emergent game boils down to "here's the initial situation, go!" while running an authored game is "here's the content I wrote, let's go through it." As soon as you start saying "the players have to go here, then here, then here" it's not really emergent any more. And as soon as you say "I don't know what's going to happen" you lose the prep abilities, and the story-writing of the authored style.

Like, for me to author stuff, I have to know the state of the world well enough to do that at some point in the future. And as soon as that's true, it's not emergent. And if it's emergent, then I don't really know the state well enough to predict it and write stuff for it.

Maybe there's some kind of middle ground? But I think that the goals are pretty opposite of each other, the work to run them is pretty different from each other, and it's perfectly okay to recognize that these are very different playstyles with different success criteria. What doesn't really interest me is "how do we make authored styles more appealing to emergent players" because, from my perspective, I just don't want authored styles, and want that respected.

[Thrudd]

I think it's possible to alternate between them - that's not really a implying a continuum, but it is possible to have a game that can be satisfying to both types of players, as long as they are willing to compromise and share time a bit. You can have authored adventure paths happen as events occasionally in a game that is otherwise run as emergent. It would be "fair" to all, so long as an equal number of sessions are spent playing in each style.

[KorvinStarmast]

I find this insistence on "it is either or" to be a bit too restrictive ... but I still like your framing. The Civ example is a good one about ripple effects, even if it isn't a role playing game.

[Vahnavoi]

A distinction between authored and emergent elements exists, but in practice trying to build a dichtomy around that doesn't work. Most games and virtually all open-ended games have both types of elements; only very simple or tighly scripted games have no emergent game play.

[icefractal]

I'm not sure I'm getting the classification here. What would you call these?

#1: The game is a sandbox - a setting that players can travel to any part of, ally with or oppose almost any group, take on mercenary jobs or not, help people or not, seek out treasures or not. However, there's an agreement that at the end of a session, the players decide what they're going to do during the next session, and they stick to that. And that's because the GM will then prep the appropriate material between sessions.
So the players have a lot of freedom on a macro-level, but during a session most things that happen will be pre-written by the GM rather than improv.

#2: The GM is using a huge existing piece of material, like Ptolus or a mega-dungeon. Within that, the players can do whatever they want to, but they can't decide to leave the city entirely (well they can, but it means the end of the campaign).

#3: The GM has a definite plot in mind, with a beginning, certain milestones, and a pre-defined ending. However, the game is run in a very improv style, and between those fixed points almost anything can happen. The GM uses subtle railroading / quantum ogres to make all the infinite possible paths converge on those fixed points, but does almost no actual "authoring" of material - the entire campaign notes fit on a single page.

[Satinavian]

Maybe there's some kind of middle ground? But I think that the goals are pretty opposite of each other, the work to run them is pretty different from each other, and it's perfectly okay to recognize that these are very different playstyles with different success criteria. What doesn't really interest me is "how do we make authored styles more appealing to emergent players" because, from my perspective, I just don't want authored styles, and want that respected.

I do think there is a middle ground and i also think most of the games i have been part of in recent years are there. That is why i am sceptical it is more useful than sandbox vs railroading was.

You say in authored games the plot is basically predecided by the GM and in emergend games it simply is not. But what i experience the most is that GM do have a general idea of the potential plot and that is what is prepared best, even with scenes, but pretty much nothing of it is guaranteed and the story can go elsewhere and suddenly need improvisation. But often the GM just guessed right and most of the preparation can be used. The better the GM knows their players, the more often that happens. But ending up in completely unexpected developmets and results is not really rare either.

So both extremes don't really fit. The GM prepares mostly as for "authored" but instead of using railroading to make the prepared plot happen, the plot is propagated and indeed made by the "emergent" method and allowed to end up anywhere. Though often exactly where the GM planned.

[NichG]

It does feel like most of the mixtures are more about one thing at one scale and another thing at another scale. The binary view is more natural to me if we specify what scale we're talking about, especially if it's an asymptotic one (the arc of an entire campaign + conclusion + topic/goals = asymptotically large; individual character decisions/moments = asymptotically small).

Given the scale, the natural question to ask is then - looking at a summary of events limited in detail to only the scale you're talking about, would you get a different summary if the game had been played by different players?

So I feel like part of the reason the mixtures feel unsatisfying is they're usually proposed in a 'up to X, you decide, above X I decide' way - for any finite X, if you want to control the X->infinity limit it's going to be unsatisfying.

Do games where e.g. there are a small number of authored sequences (small scales can be authored) but whether they get triggered or not totally and only depends on the emergent parts - e.g. nothing that happens within an authored part automatically has consequences beyond the players' choices that led there in the first place - feel authored or emergent?

[King of Nowhere]

I do think this catches better the difference between sandbox and railroad, and it's probably a more accurate way of putting it.


As for the matter of prepared content vs improvisation, my table solved it by asking the players what they plan on doing next at the end of the session.
this way, the dm has a week to prepare in detail a reaction to what the players freely decided to do. it's the better of both worlds

[Quertus]

I don't think that the useful version of these terms describes "Civilization" as "Emergent".

What?

Yeah, all of Civilizations possible endings are pre-scripted. I can't win Civilization by banishing the other civilizations to another dimension, or by owning a monopoly on fish, or any other user defined objective, without some serious mods / hacking.

In Civilization, the set of end goals are already Authored. Even though each individual game will play out differently in many ways, I can already tell before you start what your set of possible end conditions will look like.

So, for me, for what I care about between "Authored" vs "Emergent", Civilization is Authored. It does not give me the freedom to choose my end state.

How does this relate to Linear vs Sandbox(y) vs Railroading?

As I understand the terms, Linear and Linear (Branching), the scenes are pre-built. Like a module. You can transition between them in a linear or branching fashion, but the scenes (and transitions) are pretty well established.

Railroading is when the GM negates the logical consequences of a PC's actions in order to prevent the game from leaving the established / expected scenes and transitions.

The more Sandboxy something is, the more open to user definition the scenes and transitions are. A Sandbox is more about providing scenarios and content than having expected flows / outcomes.

Which... sounds like Linear maps to Authored, and Sandbox(y) to Emergent.

Am I missing anything here?

[Olffandad]

I think both of these ideas are useful tools in a GM or storyteller's toolbox and should be viewed as a useful scale, not an absolute.

A new GM may need an "authored" pre-written scenario to learn GM skills, or new players may need simpler, controlled adventure content for example.

My group is somewhat passive, so they have struggled to find footing when the plot or missions aren't laid out clearly. Sometimes they aren't so engaged or don't have time or creative energy free to engage in a collaborative process, also.

Of course, a GM may want to prepare maps, lore, and encounters so they don't have to be totally pulling content out of thin air. Not every GM is a perfect improvisation machine.

On the other hand, too much preparation can be a straitjacket - for instance, my DM ran battles with complex mixes of enemies, height and distance where the negotiations of distance/spell effects took much time. I would have used a more narrative mode, but she was more comfortable with the content and method she had planned.

She enjoyed creating detailed maps and setting up encounters very exactly, so we were probably going to ride the bus to all of the planned encounters.

My own personal interpretation is a mix of these concepts - I would tend to think out the main points of an encounter and the participants, then think through the most likely scenarios of things, and use that to inform any deviations from my main plan while trying to honor the players' interpretation of what I presented.

The exact balance of authorship to emergent content will vary with the players and DMs strengths and interests.

[Lord Raziere]

I don't think that the useful version of these terms describes "Civilization" as "Emergent".

Okay look:

1. first of all Civilization is not a ttrpg, its a strategy game, different genre, not really relevant to a discussion about ttrpgs.

2. if it was, its a videogame, you can't reasonably expect the same level of flexibility as a human mind. having five victory condition is more than most videogames ever give you. most only have one.

3. it is in fact emergent, because you don't know which civilization is going to win or how. the fact that it has five set victory condition is a luxury for the medium and offers a lot of ways to play this out. you can be conquest and destroy all over nations, you can be diplomatic to win without destroying other nations, you can win a cultural victory by making your culture dominant again without destroy other nations, you can do a science victory to simply colonize another planet before everyone else, guaranteeing you will always be a step ahead of them on the space race, and a religious victory similar much to the cultural one, makes sure you have a dominant belief winning out. and since its multiplayer, the actual winner depends on emergent interactions between those players trying to win over the other, and thus could depend on a number of factors.

4. No one is talking about civilization, you just came in and suddenly started talking about it out of nowhere as if it matters. Only you are talking about this, and to somehow make this discussion about a completely unrelated game I don't think is in the spirit of the discussion. don't talk about oranges when everyone is discussing apples.

[Cluedrew]

I'm not sure. [...] So, I think there's a fundamental difference.

I'm going to actually going to argue that there is a spectrum (perhaps not a continuous continuum) with options between the two extremes. In fact although "linear campaign"* might be one extreme in terms of how authored it is, while still definitely being a table-top role-playing game, but on the other hand I wouldn't call "sandbox" the opposite extreme.

Although, sandboxes are definitely emergent, in the archetypical sandbox (or maybe just a CRPG-like sandbox) the content is authored. The setting exists before the characters/players interact with it and hence is authored. Yet how they interact with it and in what order is very open ended and because of this new experiences/combinations can emerge from the game that were not planned, were not authored, and that is a fundamentally different experience than even a linear campaign with branching paths.

Yet I would argue that I have played and run games that are even less authored. Well we still made them up, in fact we make them up during the game. And by "we" I do mean the table, the players usually participate in this indirectly but it still isn't a sandbox because the GM has prepared toys ahead of time. They are created from the character's drives and motives** and it is something that is also fundamentally different than a sandbox.

So there are more than two fundamental structures, possibly more than three even, and as Thrudd and NichG have pointed out, you can splice these together in different ways.

* With may or may not be railroading if we want to get into that divide.
** I can go into more detail if people like but really all that matters is A) it is a legitimate way to play and B) it is fundamentally different than both a linear adventure and a sandbox.

To Lord Raziere: kyoryu brought up Civilisation up in the post I quoted from. I don't have much to say about the other 3 points either way.

[Saintheart]

We keep talking about sandbox and railroad/linear games. But I think maybe that's not the key point.

Maybe it's "authored" vs. "emergent".

In an authored game, the GM or module author write the story of what will be happening. This doesn't mean it's exactly linear, though it can be. There can be optional branches, there can be some shuffling of what order things are done in, etc. But the key is that all the things that are done, all of the situations handled are things that were predetermined.

An emergent game, on the other hand, doesn't have that. Sure, there can be some initial elements, but what the players do is fundamentally unknown, and how the world changes in response to their actions is also fundamentally unknown. Players can do basically whatever they want.

This doesn't mean that authored games have no agency, though they have very limited or no agency in specific areas - usually in terms of what specific encounters/scenes happen. Sure, you can change some aspects of hte result, or you can have side-effects, often significant, on the world, but you're really going to mostly go along the same path as others - you might do things in a different order, or skip a few things, but you're mostly going to do the same things.

This has advantages! Since the GM knows what the players are going to be doing, the GM can prep very cool, detailed, prep-intensive things for the players to do - custom actions, cool terrain, super balanced encounters, etc. And if the players are mostly concerned about that kind of stuff, it's a great way for the GM to give them what they want.

But fans of emergent games want different things - they want their decisions to matter. They want the game to take on a different shape than it would have if they had made other choices. They want to be able to solve the problems their own way. They want to do things that the GM didn't plan for, and they want to make that take the "story"/game in a way that the GM could not have predicted.

This is great for players for whom the major point is "how do I react to this situation". It's not as great for players who really want the resolution of the encounter to be the major focus, since by definition it's not going to result in as detailed/designed/tuned encounters.

I think this could be more useful, and I tried to keep the distinction as neutral, and point out the value of each side of this, and how they mutually oppose each other - the design of each serves certain purposes, but makes the opposite style harder to achieve.

I prefer to analyse this whole problem through the MDA Model of videogame design, which basically says: there are essentially - but not exclusively - 8 kinds of fun/enjoyment that people look for in playing games. Two of these types are narrative - the satisfaction of experiencing a well-told story as it unfolds - and discovery - the satisfaction of exploring and learning new things, colouring in a blank piece on the map. And a third, which clouds the whole thing, is Expression, or flat-out pleasure from expressing yourself creatively.

Again, this is not to say that people necessarily want only one kind of fun. Indeed people often don't know what they want, that's one of the curses of our two-speed brains. Often people want several kinds of fun at once. TTRPGs often supply a dose of several.

I think this is where the fundamental screaming arguments about 'railroading DM' versus 'lazy improv DM' come from. It comes down to people arguing about the kinds of fun they like to have, and projecting those likes onto what a TTRPG "should" be or whether a TTRPG is a "real" TTRPG if it isn't delivering primary Narrative versus primary Discovery.

Overlaid onto that is a fundamental difference in preference by way of DM preparation. Simplistically: you're either a DM who thinks their created world should exist in heavy detail ahead of interaction with the players, and continue to follow those pre-formed principles of existence when the players aren't directly interacting with it ... or you're a DM who doesn't.

Putting it by way of an exaggerated stereotype, DMs fall on a simple continuum. At one end are the frustrated authors who write fanfiction for their parties of PCs. At the other end are frustrated stage directors whose actors cease to exist once they're behind the curtain and not visible to anyone in the audience, all that really matters is what's going on right at this moment in the game. You're either trying to play in the game as a player with everyone bar the PCs as your characters, or you're a (overly?) detached observer who sees themselves as delivering an experience to the players.

Spoiler: An Editorial Aside

Show

At least in third edition, my personal view is that WOTC devoted a hell of a lot of time to the first end of the continuum and little time to the second. There are tables and tables in DMGs and across sourcebooks which, if followed religiously, would demand a DM create massive verisimilitude that the players will never notice and never even bother with. And it is apparent that the utility of following these tables and tables is vanishingly low to pointless.

This is a problem. A very big problem. Because, especially in a world which is getting better and better at enhancing sensory pleasure and automating complex experiences, people as a group are gravitating to experiences that require them to exert as little effort as possible to obtain pleasure. Complexity rings on certain of the 8 pleasure types (challenge, notably) but it's death on types such as fantasy and submission, which videogames can offer and unquestionably are winning at when you compare the size of the TTRPG industry to the videogame industry.

I only speak for myself, but I don't think I am speaking radically or uniquely saying there's a strong vein of DM advice that says: never prepare more than one session ahead. There's all sort of reasons given for that advice, but one of its virtues is that it avoids DM burnout. And DM burnout has to be avoided if the TTRPG industry is going to survive. As I've said before elsewhere, most games can survive the loss of a player, but not many games survive the loss of a DM. And the problem with mountains of material trying to make a DM a player in the game rather than the key person delivering the experience to the players push hard against that sort of vital advice.

And please: don't quote me Ernest Hemingway's 'iceberg' theory of fiction writing and argue that your game is more plausible if you prepare all this absolute manure ahead of a session. Not unless you're writing a novel, which you aren't, because you're trying to run a TTRPG, where you will have to adapt and you will have no room to allow interesting inconsistencies to drive better emergent stories. TTRPGs and passive fiction reading are vastly different forms of entertainment, and it's like nails down a chalkboard when someone starts quoting three-act structure in reference to RPGs. Not because it's an invalid pursuit if your kind of fun is Narrative, but because that stupid advice has been more responsible for more DMNPCs destroying entire campaigns than any other.

Like it or not, most players of D&D are not experienced grognards. The guys who wrote third edition, fourth, and fifth all express either openly or by their actions that the rules (and modules) were designed for game store casuals, teenagers, and beginners. And therefore the authors injected huge amounts of material to try and suck the DM into the world of fantasy as well, because that's what all the tables and tables of stuff down to goddamn demographics of entire cities seem to have been intended for. The problem with it being that those beginners, casuals, and teenagers became veterans, regulars, and thirtysomethings who mistook something to draw them in as something that couldn't be dispensed with and therefore to be pushed on new, incoming DMs 'rediscovering' older editions. The world's changed, attention spans have shortened faster than the odds of Elon Musk acquiring Twitter, you can't throw 300-page books of dense two-column text at DMs anymore without really thinking about what you're expecting the DM to do with it or to use it or without bearing in mind the person you're throwing it at can likely only concentrate for 160 characters at a time. People whinge that fifth edition is too simplistic, I agree but I also shrug, because that's the world we live in and increasing complexity of a system actually only makes it easier to abuse or break: witness third edition.

When you step back and look at an adventure from the point of view of asking what kind of enjoyment is this thing trying to deliver to its players, the analysis of whether it is "railroading" or not becomes a lot simpler. Or at least less emotional. You also get a better idea, I think, of what needs to be done to the adventure to shape it to address given kinds of fun.

Spoiler: Another editorial rant

Show

Published adventure modules, by which I mean the big WOTC ones at least from third edition onward, appear to try to be as generic as possible. They'll lean on narrative because, like it or not, in our present world it's the easiest form of adventure for a brand new DM to run. Their narrative direction is clear so a casual DM can pick it up and run it as written without the whole session (necessarily) collapsing.

D&D might have been built as an improv game in its early editions. It might have been built as a game where the primary focus was discovery and exploration, i.e. the great gonfalon of OSR. That's life. Third edition came out, massive customisation (and therefore attachment to) characters happened, and clear narrative-based modules came out to support it. The miracle is that third edition and the later editions can still be run as improv-heavy games, but one just needs to recognise the bias and look up the tools to help you overcome them.

For clarity, I define 'railroading' pretty narrowly. The basic sequence of any RPG comes down to:

1. Introduce the scene.
2. Invite the players to act.
3. Adjudicate the outcome of their actions.
4. If the scene isn't over, return to step 2.
5. If the scene is over, transition to the next scene and return to step 1.


"Railroading" is, to me, removing step (2) from the equation, whether explicitly or implicitly. And it usually becomes apparent in how step (3) comes out. If there is only one way a given encounter can play out, then there is no actual encounter. There is no invitation to the players to act. There might be an invitation to spectate - most archetypically and obviously in the case of the party forced to watch as the DMNPC screams "Witness Me!" before exploding the dragon with their staff of awesomeness - but no invitation to act. And is it not the case that DMNPCs show up more frequently in the type of DM leaning on the first end of the spectrum I've identified, i.e. the frustrated authors? Choice is the engine of all roleplaying games, without it there is no actual game. This does not mean that every choice is always available, because unlimited choice is the same as no choice, but there surely is no RPG - or at least, no real encounter, only description/fluff text - if no choice can be made.

Outside of this, I don't define an adventure with a strong, very defined narrative sequence as "railroading", or at least I regard it as a sarcastic joke by the players. A game built primarily to induce narrative is not of itself a railroad; that only happens when the players have no capacity to act in reality. And yeah, that's a balancing act, but nobody ever said DMing was easy.

Spoiler: Other random complaint

Show

This does go both ways to some extent.

I have a corollary to the party that feels cheated when their DM said they'd be getting a sandbox and instead they got a host of DMNPCs with the party expected to be its cheer squad.

It's when the DM asks for players interested in an open world sandbox; gets a bevy of players who all swear black and blue they're looking for a good open world and want to play an open world; works up a world; works up a list of things the players can do (adding every two paragraphs or so that there's all sorts of other things they might do, they have only to think and ask) ... and then sits back as the players make like Buridan's Ass the moment they enter Medieval Town #1 and get a list of things they can do. People are just the worst.

[kyoryu]

A distinction between authored and emergent elements exists, but in practice trying to build a dichtomy around that doesn't work. Most games and virtually all open-ended games have both types of elements; only very simple or tighly scripted games have no emergent game play.

It's not the elements that determine if a game is authored or emergent - it's the path that the players take. You can have as well-defined of a setting as you want, if the players are still largely in control of how things play out, it's an emergent game.

To go back to the Civilization example, you could take a saved game on the first turn and hand it to five different players - each one would likely get a different game, and often significantly different. The "history" of their events would read very differently. And yet all of the elements of the game are set in that saved game.

OTOH, you can take those same five players and have them play Mass Effect - and they're still going to hit exactly the same major beats along their path. Some of the side effects may differ (do they save the Rachni or kill them?), but they're going to go on the same major five or six missions, and they'll share a subset of the side quests. Some of the orders may change, but they're essentially going through the same sequence of events, even if shuffled around a bit.

I'm not sure I'm getting the classification here. What would you call these?

#1: The game is a sandbox - a setting that players can travel to any part of, ally with or oppose almost any group, take on mercenary jobs or not, help people or not, seek out treasures or not. However, there's an agreement that at the end of a session, the players decide what they're going to do during the next session, and they stick to that. And that's because the GM will then prep the appropriate material between sessions.
So the players have a lot of freedom on a macro-level, but during a session most things that happen will be pre-written by the GM rather than improv.

Probably emergent.

#2: The GM is using a huge existing piece of material, like Ptolus or a mega-dungeon. Within that, the players can do whatever they want to, but they can't decide to leave the city entirely (well they can, but it means the end of the campaign).

100% emergent. It is not the presence of prepared elements that makes a game emergent. It's the presence of a set path for the players to follow.

#3: The GM has a definite plot in mind, with a beginning, certain milestones, and a pre-defined ending. However, the game is run in a very improv style, and between those fixed points almost anything can happen. The GM uses subtle railroading / quantum ogres to make all the infinite possible paths converge on those fixed points, but does almost no actual "authoring" of material - the entire campaign notes fit on a single page.

Has a game like this ever existed? Still, it's authored - the path is set, and nothing is going to change that, even if there's some leeway between the points.

You say in authored games the plot is basically predecided by the GM and in emergend games it simply is not. But what i experience the most is that GM do have a general idea of the potential plot and that is what is prepared best, even with scenes, but pretty much nothing of it is guaranteed and the story can go elsewhere and suddenly need improvisation. But often the GM just guessed right and most of the preparation can be used. The better the GM knows their players, the more often that happens. But ending up in completely unexpected developmets and results is not really rare either.

That sounds like a very common style of primarily authored content. That may be your experience as to the most common style, but it's not how I or a great number of others run games.

And honestly this may be a big part of the issue - I think a lot of people (not necessarily you) assume that certain things are just part of how games are run, and haven't had experience in more emergent game styles. So it's hard to have that conversation when there's no shared experience. I see this all the time in the strawman of "if you don't railroad there's no plot, and I'd rather have a plot, so railroading is necessary".

It does feel like most of the mixtures are more about one thing at one scale and another thing at another scale. The binary view is more natural to me if we specify what scale we're talking about, especially if it's an asymptotic one (the arc of an entire campaign + conclusion + topic/goals = asymptotically large; individual character decisions/moments = asymptotically small).

Given the scale, the natural question to ask is then - looking at a summary of events limited in detail to only the scale you're talking about, would you get a different summary if the game had been played by different players?

Yes, exactly.

So I feel like part of the reason the mixtures feel unsatisfying is they're usually proposed in a 'up to X, you decide, above X I decide' way - for any finite X, if you want to control the X->infinity limit it's going to be unsatisfying.

Usually the "mixtures" offer some leeway of freedom within a larger context. The problem is that the set nature of the larger context is exactly the thing that people that prefer emergent games are trying to avoid.

Note that I'm not using words like "good" or "bad". These are just preferences. But they're valid preferences.

Do games where e.g. there are a small number of authored sequences (small scales can be authored) but whether they get triggered or not totally and only depends on the emergent parts - e.g. nothing that happens within an authored part automatically has consequences beyond the players' choices that led there in the first place - feel authored or emergent?

I'd say if the overall experience is still emergent, it's probably emergent? I dunno. It's very non-concrete.

I do think this catches better the difference between sandbox and railroad, and it's probably a more accurate way of putting it.

Thank you. I'm trying to hit on the key parts of what "sandbox" people really want.

As for the matter of prepared content vs improvisation, my table solved it by asking the players what they plan on doing next at the end of the session.
this way, the dm has a week to prepare in detail a reaction to what the players freely decided to do. it's the better of both worlds

This was proposed above. I'm not sure how I'd feel about it in practice.

I don't think that the useful version of these terms describes "Civilization" as "Emergent".

What?

Yeah, all of Civilizations possible endings are pre-scripted. I can't win Civilization by banishing the other civilizations to another dimension, or by owning a monopoly on fish, or any other user defined objective, without some serious mods / hacking.

It's still a computer game. There are limitations.

Even if there are only certain win conditions, different players will go through very different paths to get to whatever end they choose to pursue, even starting from the same seed/save game. One player may war with England, make peace with India, and so on and so forth, while another player just huddles up and works on their tech tree. The "stories" of their game will be

In Civilization, the set of end goals are already Authored. Even though each individual game will play out differently in many ways, I can already tell before you start what your set of possible end conditions will look like.

This feels very disingenous to me. Computer games have inherent limitations, and if you can't see the difference between Civ and Mass Effect in terms of how emergent/plotted they are, then you're willfully ignoring them.

Which... sounds like Linear maps to Authored, and Sandbox(y) to Emergent.

Am I missing anything here?

More or less, yes. However both "sandbox" and "linear" (or especially "railroad") have connotations that aren't useful. Specifically sandbox often implies a lack of things happening, which is not necessary - many if not most emergent games have various things in motion, but how they end up is not predetermined.

I think both of these ideas are useful tools in a GM or storyteller's toolbox and should be viewed as a useful scale, not an absolute.

A new GM may need an "authored" pre-written scenario to learn GM skills, or new players may need simpler, controlled adventure content for example.

Oh, for sure. And to be clear, neither is good, and neither is bad.

My group is somewhat passive, so they have struggled to find footing when the plot or missions aren't laid out clearly. Sometimes they aren't so engaged or don't have time or creative energy free to engage in a collaborative process, also.

It's totally possible to have more emergent games that still work with passive players - what's required is really a motivation.

Of course, a GM may want to prepare maps, lore, and encounters so they don't have to be totally pulling content out of thin air. Not every GM is a perfect improvisation machine.

It's not really about prep, especially for maps, factions, NPCs, etc. And it's not about not doing some prep work for possible encounters. It's more about why you're having them, and where you go from there.

She enjoyed creating detailed maps and setting up encounters very exactly, so we were probably going to ride the bus to all of the planned encounters.

A valid style. But one that some people like, and others don't.

The exact balance of authorship to emergent content will vary with the players and DMs strengths and interests.

Exactly.

4. No one is talking about civilization, you just came in and suddenly started talking about it out of nowhere as if it matters. Only you are talking about this, and to somehow make this discussion about a completely unrelated game I don't think is in the spirit of the discussion. don't talk about oranges when everyone is discussing apples.

That was me, as an example of emergent vs. authored games slightly out of context (I often use out-of-context examples to get away from things being quite so contentious)

Although, sandboxes are definitely emergent, in the archetypical sandbox (or maybe just a CRPG-like sandbox) the content is authored. The setting exists before the characters/players interact with it and hence is authored.

No, when I refer to "authored" I don't mean the world - I mean the path that hte players take.

Yet how they interact with it and in what order is very open ended and because of this new experiences/combinations can emerge from the game that were not planned, were not authored, and that is a fundamentally different experience than even a linear campaign with branching paths.

Precisely this.

Yet I would argue that I have played and run games that are even less authored. Well we still made them up, in fact we make them up during the game. And by "we" I do mean the table, the players usually participate in this indirectly but it still isn't a sandbox because the GM has prepared toys ahead of time. They are created from the character's drives and motives** and it is something that is also fundamentally different than a sandbox.

Yes, those are emergent games, and why I'm avoiding the term "sandbox". Amount of prep is fairly orthogonal.

I think this is where the fundamental screaming arguments about 'railroading DM' versus 'lazy improv DM' come from. It comes down to people arguing about the kinds of fun they like to have, and projecting those likes onto what a TTRPG "should" be or whether a TTRPG is a "real" TTRPG if it isn't delivering primary Narrative versus primary Discovery.

I didn't talk about narrative vs. discovery at all, and I think it misses the point.

I'm also not sure where "lazy improv DM" comes from so.

Overlaid onto that is a fundamental difference in preference by way of DM preparation. Simplistically: you're either a DM who thinks their created world should exist in heavy detail ahead of interaction with the players, and continue to follow those pre-formed principles of existence when the players aren't directly interacting with it ... or you're a DM who doesn't.

That's pretty orthogonal.

When you step back and look at an adventure from the point of view of asking what kind of enjoyment is this thing trying to deliver to its players, the analysis of whether it is "railroading" or not becomes a lot simpler. Or at least less emotional. You also get a better idea, I think, of what needs to be done to the adventure to shape it to address given kinds of fun.

I've played these silly games for forty years now. I'm aware of what styles I do and don't like, and have played enough to really know. Kthx.

For clarity, I define 'railroading' pretty narrowly. The basic sequence of any RPG comes down to:

1. Introduce the scene.
2. Invite the players to act.
3. Adjudicate the outcome of their actions.
4. If the scene isn't over, return to step 2.
5. If the scene is over, transition to the next scene and return to step 1.

This is the part I really wanted to address. I think you actually missed a step in 5.

5. If the scene is over, let the players decide their next move.
6. Start the next scene and return to 1.

or something like that.

"Railroading" is, to me, removing step (2) from the equation, whether explicitly or implicitly.

So I think it's actually in step 6. In step 6, it's railroading if the choices made in 5 are irrelevant - either through outright negation or manipulation. That's an extreme level of railroading, of course, but it's the fundamental problem from my PoV as an emergent-preferred player. That negation can be through lack of choices (All exits but one are cut off) by inaction (nothing will progress until you do the right thing) by illusion (quantum ogre) or something else. And in less extreme versions, you can probably do some optional things or change up some ordering (again, see Mass Effect). But it's still primarily authored.

[Saintheart]

I didn't talk about narrative vs. discovery at all, and I think it misses the point.

I know you didn't, but it is completely the point. I can see how you might have taken my post the wrong way, but I'm not actually disagreeing with you. What I'm saying is that if DMs focus on the type of enjoyment they're trying to convey, then moulding adventures becomes an easier analysis. That is, know your crowd and you don't have to worry about whether the narrative in a TTRPG can be easily categorised as authored or narrative, you just build your adventures playing to narrative or discovery impulses as desired.

I've played these silly games for forty years now. I'm aware of what styles I do and don't like, and have played enough to really know. Kthx.

I wasn't saying otherwise. If my post came across as patronising, or suggesting that you don't play linear just because you don't know what you like, that was not my intent at all. Quite the opposite, emergent RP is a valid playstyle and one I wish I had the chance to run more often. It's also (arguably) easier on a DM in terms of prep, so I have some self-interest on that one.

This is the part I really wanted to address. I think you actually missed a step in 5.

5. If the scene is over, let the players decide their next move.
6. Start the next scene and return to 1.

or something like that.

...

So I think it's actually in step 6. In step 6, it's railroading if the choices made in 5 are irrelevant - either through outright negation or manipulation. That's an extreme level of railroading, of course, but it's the fundamental problem from my PoV as an emergent-preferred player. That negation can be through lack of choices (All exits but one are cut off) by inaction (nothing will progress until you do the right thing) by illusion (quantum ogre) or something else. And in less extreme versions, you can probably do some optional things or change up some ordering (again, see Mass Effect). But it's still primarily authored.

As said above, I don't think defining something as authored or emergent really helps as such - we're already into gradations of one or the other, "primarily" authored for example.

I am interested in the phrase "it's railroading if the choices made in 5 are irrelevant", though -- what do you mean by the word "irrelevant"?

Irrelevant to the PCs?
Irrelevant this round but not next round?
Irrelevant to the NPCs?
Irrelevant to campaign world, or to the nation in which the narrative is occurring?
Irrelevant in that the same encounters took place?
Some combination of any or all of these?

This is not me trying to nitpick, it's not a gotcha question, I'm really trying to nail down exactly what is meant when a player says "my choice was irrelevant" in a TTRPG specifically. I suspect we're going to get no closer than the old definition of pornography, i.e. "I can't define it closely but I know it when I see it", but is working out what we mean by "irrelevant" worth looking into? (I mean, one of the big, big criticisms of Mass Effect 3 was precisely that - that the player's choices were irrelevant in large degree, in that the ending sequence came down to one of three differently-coloured cutscenes when Bioware had spruiked/promised that your choices would have meaningful effect on what happened. In passing, a TTRPG of that series could have handled ME 3's plot problems, easily, but that's another story entirely.)

[Satinavian]

That sounds like a very common style of primarily authored content. That may be your experience as to the most common style, but it's not how I or a great number of others run games.

And why do you call it authored instead of mixed ? It has from your description of emergent games :

but what the players do is fundamentally unknown, -> There are expectations, but no knowledge

and how the world changes in response to their actions is also fundamentally unknown. -> World always reacts in the most plausible/natural/sensible way but as player actions can only be guessed, not known, the same goes for the changes of the world.

Players can do basically whatever they want. -> They totally can.

But fans of emergent games want different things - they want their decisions to matter. -> they do matter

They want the game to take on a different shape than it would have if they had made other choices. -> Which is absolutely the case

They want to be able to solve the problems their own way. -> They can do that

They want to do things that the GM didn't plan for, -> they can do that, but obviously only by doing surprising things instead of sticking to their regular routine

and they want to make that take the "story"/game in a way that the GM could not have predicted. -> As above. They would need to do things the GM had not predicted to take the story in an unpredicted direction, but they totally can.


So why do you classify it as "authored" when it also shares all those properties of "emergent"?

And honestly this may be a big part of the issue - I think a lot of people (not necessarily you) assume that certain things are just part of how games are run, and haven't had experience in more emergent game styles. So it's hard to have that conversation when there's no shared experience. I see this all the time in the strawman of "if you don't railroad there's no plot, and I'd rather have a plot, so railroading is necessary".

I think you should not mix a discussion about classifying styles of play with an attempt to present "emergent" play to people who don't know it yet. One of the reasons i am so certain, that a mixture or continuum between "authored" and "emergent" exists, is that i also have experience with both extreme ends of the scale. And this experience certainly does help me to recognize i am more happy in a sweet spot in the middle, seek out like minded players and play this way most of the time.


But enough of the mixture and back to the extremes. When you describe what people like from authored/emergent games, i think you missed an important point : IME emergent games require more metagaming during play where authored games can offload most of that into the prep phase of the GM and go full immersive experience during play.

[Vahnavoi]

It's not the elements that determine if a game is authored or emergent - it's the path that the players take. You can have as well-defined of a setting as you want, if the players are still largely in control of how things play out, it's an emergent game.

That's irrelevant for what I just said. Let me rephrase what you just said: "a game with authorially defined elements can still have emergent paths between those elements."

What I'm saying is that paths - connections between elements - are elements of their own. Additionally, that this kind of emergence is so god damn common that trying to build a dichtomy between game types on it is fruitless. I'm pretty sure Mass Effect has it's fair share of it despite strictly scripted over-arching plot. It might not have as much of it as, say, Civilization, but it definitely has more than two-branch visual novels or simple point&click adventure games, so.

This said, I don't advocate calling it a spectrum. Spectrum implies a range between values, a start and an end point. Neither authorial nor emergent elements have an end point in a vacuum. There doesn't actually exist, and it isn't even possible for a game to exist, that's "all emergence" as opposed to "no emergence", because it's always possible to add another point element for path elements to connect to. Similarly, there doesn't actually exist, and isn't even possible for a game to exist, that's "nothing authored", because games are defined by their rules, and rules must be authored. (Yes, the prior statement includes incomplete games such as roleplaying games. Just because all rules of a game aren't defined doesn't mean the rules that have been defined aren't definitional to a game.)

To go back to the Civilization example, you could take a saved game on the first turn and hand it to five different players - each one would likely get a different game, and often significantly different. The "history" of their events would read very differently. And yet all of the elements of the game are set in that saved game.

Wrong. Players are elements of a game. The individual skills, thoughts and strategies of players aren't recorded in the save file. That's chief reason why giving the same save file to different players produces different results. If I dug deep into programming of some specific Civilization game, I could probably find even more technical elements that aren't saved, mostly hidden variables such as random seeds.

[Yora]

I've been using the term "scripted" for some years. It gets the point across and avoids having to discuss what railroad really means every time you talk about it. I guess authored does the same job, though I think I've seen that term used in regards to players authoring the story. Which is the exact opposite of a scriptrd game.
Though regardless, it's an inferior way to run RPGs that loses the best parts of what makes RPGs unique as a medium.

[Vahnavoi]

Scripted is often sufficient, when used in specific theatrical sense. It starts to lose it power once you look at scripts and scripting on a deeper level and realize there are a lot of different ways to make a script, and several discussions about giving expressive freedom to actors that are nearly one-to-one with common discussions about player agency.

[Morgaln]

Wrong. Players are elements of a game. The individual skills, thoughts and strategies of players aren't recorded in the save file. That's chief reason why giving the same save file to different players produces different results. If I dug deep into programming of some specific Civilization game, I could probably find even more technical elements that aren't saved, mostly hidden variables such as random seeds.

Isn't that Kyoryu's point, though? The way a game of Civilization develops largely depends on how the player approaches it (=emergent). Whereas Mass Effect will mostly develop the same, regardless of player approach (=authored).

And why do you call it authored instead of mixed ? It has from your description of emergent games :

but what the players do is fundamentally unknown, -> There are expectations, but no knowledge

and how the world changes in response to their actions is also fundamentally unknown. -> World always reacts in the most plausible/natural/sensible way but as player actions can only be guessed, not known, the same goes for the changes of the world.

Players can do basically whatever they want. -> They totally can.

But fans of emergent games want different things - they want their decisions to matter. -> they do matter

They want the game to take on a different shape than it would have if they had made other choices. -> Which is absolutely the case

They want to be able to solve the problems their own way. -> They can do that

They want to do things that the GM didn't plan for, -> they can do that, but obviously only by doing surprising things instead of sticking to their regular routine

and they want to make that take the "story"/game in a way that the GM could not have predicted. -> As above. They would need to do things the GM had not predicted to take the story in an unpredicted direction, but they totally can.


So why do you classify it as "authored" when it also shares all those properties of "emergent"?

I'd like to second this question; the play style Satinavian describes is very similar to my own. I usually give the players an initial problem to start the game off and then see where the decisions of the players take us. However, I have a few scenes (2-3) in mind that I would like to be part of the game. I wait for opportunities to insert these scenes into the emerging story at points where they would fit. Since I know my players, I usally have no problem finding an opportunity like that, but it occasionally happens that I never use one of these scenes since the game never reaches a point where that scene would make sense. These scenes are never the end of the story. I have no idea how the story will end when we start.
I'd classify that as emergent play with some authored content, i. e. mixed.

[Vahnavoi]

Isn't that Kyoryu's point, though? The way a game of Civilization develops largely depends on how the player approaches it (=emergent). Whereas Mass Effect will mostly develop the same, regardless of player approach (=authored).

We're not in disagreement over existence of emergence in Civilization. I'm refuting a particular point about how the game stores information.

But, for the sake of argument, let do away with computers entirely. I set up a Chess board with some legal board position as a puzzle. Two players then continue this game. The one thing that the board position does not tell is who's turn it was before that game was halted, but it doesn't actually matter: since Chess is a deterministic perfect information game, it's in principle possible to calculate every legal outcome from that point on for both black and white.

For some board positions, this number is in the single digits. For others, it's in the low dozens or hundreds, but nonetheless within capacity of me, as the puzzle maker, to calculate and put in writing if I want to. Basic rules of Chess are conserved beyond the starting position.

I don't know which moves the two players will make, but I do know every possible move they could make. All possible paths have been accounted for and are visible to me as the one who set up the puzzle. So is the game authored or emergent?

[The Glyphstone]

I'd definitely say that it is a sliding scale rather than a binary, as well. To consider my last game, the party was effectively given the premise - 'an ancient evil stirs and will soon awaken, gain strength and gather allies before confronting it'. But then they were dropped in the top corner of an unexplored 20x20 hexcrawl grid, full of stuff to loot, fight, or make friends with. In theory, like the chess game mentioned above, every potential outcome 'could' be calculated, since the grid map was drawn out ahead of time and I knew what each hex contained. But they only had enough time to actually explore about 1/3 of the map, or less if they devoted more time to currying favor with other factions (some of whom were mutually opposed) rather than exploring. I don't have the mathematical acumen to calculate how many thousands of potential pathways they could end up following, including the ones where they actually switched sides and joined the ancient evil instead. So is this an authored story since it has a (mostly) fixed endpoint, or is it emergent because of the massively unplanned and unpredictable game that fills the remaining 90%?

[PhoenixPhyre]

I've actually run the same scenarios with multiple groups. Same planning document. Very different outcomes and gameplay, with entirely different scenes and challenges as actually played. All departed from my planning. The intended goal was the same, the starting point was the same. Authored? Emergent? Both?

I'm on team "multidimensional chaotic solution space, with some attractors and common themes" here.

[KorvinStarmast]

I'm on team "multidimensional chaotic solution space, with some attractors and common themes" here.

And it's dialable, based on the tastes of the player group. (As noted in the other thread, some people like a bit more railroad/progress/milestone/objective arc, some people prefer a lot less structure / more discovery.

[PhoenixPhyre]

And it's dialable, based on the tastes of the player group. (As noted in the other thread, some people like a bit more railroad/progress/milestone/objective arc, some people prefer a lot less structure / more discovery.

And some people prefer "authored" more in some areas than others. There's a whole set of layers (very squishily-defined ones as well) here. And interplay between them.

Trying to boil all of this down into a single term seems rather futile. Sure, it takes more words, but explaining what parts you like and what you don't in plain language seems more useful to me.

[Telok]

Using the computer game examples:
Strict - you start at A, follow events B to C to D, and end up at Z. These are the scripted first person shooters.
Less strict - you start A, you may or may not do a number of different things from B C D, and you get ending X Y or Z based on those. The 'choose your mission' games with multiple endings.
Fairly Open - you may start A B or C, choose to do or not do stuff in the middle, and end at X Y or Z. The 'choose your mission' with different starts & endings, or the Civ style games I guess.
Way Open - choose from lots and lots of possible starting points, do anything possible in the game, no defined ending. DwarfFortress (mermaid farming, Turing comlete minecart & pressure plate computers), maybe Minecraft.

Funny thing, when I start a campaign it usually starts strict then walks quickly through the spectrum to fully open. Its like, run the first session strict, then less strict to fairly open during the first "mission"/mini-adventure, then fairly open next, and after the second mini-adventure it gets to fully open. Interesting, hadn't thought about it that way before.

[Stonehead]

It's not the elements that determine if a game is authored or emergent - it's the path that the players take. You can have as well-defined of a setting as you want, if the players are still largely in control of how things play out, it's an emergent game.

I'm gonna add my vote to the "it's a continuum" crowd. Most "authored" games I've played do give the players freedom to do whatever they want, they just have stakes set up in a way that the GM can predict generally the sort of things they'll do. During character creation, or maybe partway through the campaign, the GM learns what the characters care about. Then things happen that involve those things, and the characters naturally want to intervene. The go to cheap-and-easy way to do this is have a big villain try to destroy the world. All PCs want the world to not end, so they'll almost always go to stop the villain. Occasionally, the GM predicts wrong, and when he does, the players are free to go their own route.

I can think of one specific example in which a dimension hopping army secretly started invading, planning to take over the world. When the PCs discovered their plans, the odds seemed so overwhelming, that they decided to use their portals to try to evacuate their dimension, instead of fighting off the entire invading army. The GM readjusted his plans, and the campaign continued down this new route. That example really seems to bridge the line between authored and emergent, because the GM clearly did author a story, 90% of it even came to be, but the players drastically altered the direction of the story, so it would be pretty hard to say it was not an emergent game.

Now, not always, there have been campaigns whose plot hooks aren't much more than "your boss told you to do this" where we didn't have much say in the matter. But that's why I think it's definitely a continuum, because those games were way more authored than the first example.

IMO the biggest difference between the traditional "linear vs sandbox" dichotomy was how active the world was. It isn't always true, but typically, a narrative game will have pretty clear stakes, where the players react to the threats made by the world (and the npcs that inhabit the world). A sandbox game will have more active players, and a more reactive world. The players are given a map (usually), and actively decide which part of the world they'll interact with, and the world reacts to their choices. So it is subtly different than your "emergent vs authored" dichotomy.