D&D General [rant]The conservatism of D&D fans is exhausting.

[Campbell]

There are circumstances where people in real life have very limited agency. A prisoner in jail only has a small amount, someone who is very poor will typically have less than someone who is financially stable and so on. But no person in real life is always going to achieve their goals, they are not always going to know the odds of success, will not always know the outcomes. Those things can be added to a game, but at that point you're no longer talking about the real world definition of agency.

In a linear games you're very limited on what options you have to have real long term impact, in my living world sandbox game you have far more options. Infinite options? No, because we all agreed on broad outlines of the campaign when we started the game. Control over how other NPCs react, or fictional world events that you haven't interacted with? Again, in my game, no. Just like I can't control whether my wife wants to go to a movie, all I can do is ask and attempt to convince her to go. I can exert my will and have a meaningful chance to change my environment but there are no guarantees of success.

All I'm talking about here is basic interaction design. Is there a reliable to get information about the environment before you need to act upon it? Are the consequences of success and failure knowable? Are they a bunch of secret backstory or unknowable setting elements that create chain reactions that result in the sorts of changes we are striving towards resulting in unknowable consequence chains? It's just basic game-ability of the setting and scenario.

Not looking to control things. Just have reliable indications of how to influence them.

Can we please reset this conversation with the idea that I have multiple decades of experience playing and running roleplaying games and not someone you should have to explain basics to.

 

[robertsconley]

You can see how the two rewards fit together to create a story without ever having to try to create a story. You can get them just by playing your character(s). IF there are characters with potentially conflicting values and consequences.

At its core, a story is simply an account of events. However, that’s not what is meant by “story” in Narrativism. In that context, “story” refers to a thematic structure shaped intentionally by character decisions and their consequences.

That kind of thematic structuring doesn’t exist in my Living World Sandbox campaigns. The world exists independently of the players, and while meaningful outcomes may emerge, they aren’t designed to support a particular theme or narrative arc. They’re the byproduct of player choices interacting with a consistent world.

Take your example of the beggar leader seeking revenge. You said:

In your scenario about the wolf, there's a beggar character who is trying to avenge his son. If the scenario was that the beggars are in danger by staying (and it kind of suggest they are). Then the beggar leader is caught between two values (the safety of his followers V revenge/justice). If, as a group, we're interested in what he chooses, then that's a kind of fun, Narrativist fun. Then we see what that decision brings him.

If the scenario was that the beggars are in danger by staying (and it kind of suggest they are).

Yes, the beggars were in danger if the Demon Wolf turned its attention to them. But even in your reductive example, that danger wasn’t constructed to generate a moral dilemma. The beggars were there because, a few weeks prior, it was plausible for them to be fencing loot for the local bandits. It was bad luck that they happened to be present during the Demon Wolf’s attacks, which led to the chief’s son being killed. The leader then chose to stay and seek revenge, and the clan agreed to support him.

The outcomes you listed are all plausible:

• He leaves, the beggars are safe, and they flourish.
• He leaves, but the lack of justice causes internal conflict.
• He stays, fails, and the beggars are slain.
• He stays, succeeds, and the beggars survive.

If the players chose not to engage with that situation, I would still resolve it, because the clan’s decision to hunt the Demon Wolf had already been made. I’d roll to determine the outcome of that hunt. Based on that result, I’d then roll or judge which of the possible consequences occurred, and update my notes accordingly.

Again, the point isn’t to create a narrative but to maintain a coherent and reactive setting. The focus is on how time and events unfold whether or not the players intervene. That’s a fundamentally different procedure and design goal than what Narrativism is concerned with.

Narrativism isn’t emergent, it’s authored through play. Ron Edwards is clear: story is the point. Thematic structure arises because the system is designed to foreground premise and character choice. That’s a different goal than simulating a world and letting consequences unfold without regard for theme, and primarily considering plausibility.

The Living World sandbox offers a world you survive, change, or leave behind, while Narrativism offers a story you shape, wrestle with, and resolve. Both are powerful, and both speak to different creative instincts. Each delivers a distinct kind of payoff, and both deserve respect for how well they serve their goals.

Yes, Narrativism is a type of fun. But that fun doesn’t happen in a vacuum. It’s built on procedures and structures designed to foreground premise, character choice, and thematic resolution. Living World has its own set of procedures and structures designed to foreground visiting a setting with a life of its own as a character having adventures.

 

[hawkeyefan]

What you are both engaging in is not neutral analysis but rhetorical sleight of hand. When you frame sandbox campaigns, particularly those rooted in traditional procedures, as “vehicles for GM prep,” and then position that as ideologically suspect, you are not offering critique; you are advancing a position through mischaracterization.

I have not framed sandbox style play as “ideologically suspect”. I’ve repeatedly pointed out that it’s a perfectly valid way to play, and one that I engage in myself.

However I do think that a large part of what you’re talking about is GM led play. The focus of play is the setting more so than the characters. Again, that’s not bad.

And although the players do have the freedom to move about the setting and engage with different elements, their ability to do so is largely dependent on the GM.

This is why I don’t think it’s as player driven as is often claimed.

It's like if you give someone a car, they might be able to do what they want with it but, at the end of the day, it's still a car.

Except ... they might decide to pull it apart and use it to build all sorts of things that have nothing to do with what a car is for. At which point, the car is no longer what the automotive company made it to be or even what anyone who worked at designing and building it imagined it could be.

When the town enters play, it's what the GM made. But what it becomes is, to very significant extent, up to the players.

Sure. But to lean on your analogy a bit, the GM’s the one who also made the parts. So what the players are able to turn the car into very much depends on what the GM has provided.

 

[Lanefan]

My own feeling was it had advanced the design for D&D significantly, but was still hobbled by the system's excessive tendency toward one-off exception based design.

Which, if you are looking for a rules-based system, seems odd; in that a game that has to cover so many different bases and types of things that require rules would seem ideally suited for exception-based rather than unified design. Exception-based design allows this rule to apply to that situation while that rule applies to another, similar situation - i.e. the rules can be made more bespoke to what they are trying to accomplish when modelling or abstracting the fiction.

Unified rules tend to push against or even outright disallow unusual situations (i.e. the type that require non-unified rules) from occurring, preferring to try to shoehorn everything into the same system whether it works well in that instance or not.

I saw this with 3e - a lot of 1e-2e subsystems had been shoehorned into the d20 resolution mechanic, which while convenient for play also meant a great deal of precision and detail was lost. 5e did the same thing again by shoving so many previously-individual-to-situation bonuses and penalties into the universal advantage-disadvantage mechanic.

 

[thefutilist]

At its core, a story is simply an account of events. However, that’s not what is meant by “story” in Narrativism. In that context, “story” refers to a thematic structure shaped intentionally by character decisions and their consequences.

We're getting somewhere but you're still misunderstanding a few things.

Pretend me and my group are playing 5E or whatever edition you play in. Forget techniques for the moment. The defining feature is the group reward, I'll state this again because it's the thing that you need to beat people over the head with. Narrativism doesn't refer to a set of techniques. It refers to a group reward. If the group notices and enjoys that the beggar leaders decision is thematic/ethical, then that's a Narrativist reward. This can happen in all sorts of play and it doesn't mean you're playing Narrativist, to play Narrativist just means you want to more reliably generate that reward.


That kind of thematic structuring doesn’t exist in my Living World Sandbox campaigns. The world exists independently of the players, and while meaningful outcomes may emerge, they aren’t designed to support a particular theme or narrative arc. They’re the byproduct of player choices interacting with a consistent world.

Yeah that seems like my way of doing default Narrativism.


If the players chose not to engage with that situation, I would still resolve it, because the clan’s decision to hunt the Demon Wolf had already been made.

Yeah same here.


The focus is on how time and events unfold whether or not the players intervene. That’s a fundamentally different procedure and design goal than what Narrativism is concerned with

Another common misconception. See clocks in AW. The entirety of Dogs in the Vineyard. Trollbabe. many others.


Narrativism isn’t emergent, it’s authored through play.

I'm not sure what you mean by this.


Yes, the beggars were in danger if the Demon Wolf turned its attention to them. But even in your reductive example, that danger wasn’t constructed to generate a moral dilemma. The beggars were there because, a few weeks prior, it was plausible for them to be fencing loot for the local bandits. It was bad luck that they happened to be present during the Demon Wolf’s attacks, which led to the chief’s son being killed. The leader then chose to stay and seek revenge, and the clan agreed to support him.

This is interesting because there is no 'few weeks prior', there is just stuff the GM or scenario writer made up. You can't cite the reason as being 'it was plausible they were fencing loot' because there is no loot to fence. When not in the prep stage I'm more sympathetic to appeals to plausibility but in the prep stage it makes no sense because there is nothing to extrapolate from.

 

[hawkeyefan]

I was told that the difference is only one of scope. Is that wrong? I think it's wrong.

I’m not sure… it depends on the context in which it was said. I didn’t see where that was mentioned, so I’m not sure.

I mean… I can see the idea of why someone might say that. I mean, the geography we’re dealing with is pretend. So you could probably design both a dungeon and a sandbox as like a flowchart. You get to this node along that path, and that node along the other path. So I get the idea.

But I imagine that most games that rely on sandboxes would likely wind up having many nodes and many paths to the point where it’d be hard to track them all.

Is that only a matter of scope or has it become something else at that point?

So you're trying to alter definitions to confuse things. Now we're supposed to accept that there's Game Design and game design, instead of just saying what they are? Not me. There's Game Design, Setting Design and Playstyle. That way, you know, we can actually understand what is being talked about instead of confusing things

Max, I’ve been talking about a GM and how they prepare their game. What they do when they are doing that, the decisions they make, the elements they create. I think that’s very clear.

I don’t care about this semantic nonsense you’re talking about or why you mistakenly insist on bringing system into it. For someone who cries “strawman” all the time, you sure like to build them yourself.

So of course your One True Way is the right and unharmful one.

Or perhaps there are more than two views?

Is The Forge where the disparaging term, "play to find out the GM's notes" that gets bandied about by that side of the discussion comes from?

No, that came from describing games without poetry.

 

[Bedrockgames]

No, that came from describing games without poetry.

But the problem with this is pretty evident in the way you guys talk about prep and sandboxes. It is treated as a kind of lifeless matter, for the players to discover, perhaps interact with a bit, but not something that is hugely dynamic and responsive to what the players do. This is pretty consistent when you guys are talking about sandbox (even when you speak of it and talk about how you like it). You can call it poetry if you want, I think there is more than poetry going on in our descriptions of sandbox, but I think on the opposite end from poetry you have a kind of sterility that is overly reductive and as I have been saying, misses many of the important details and nuances of the interactions between player and GM (which we have all gone over exhaustively so I don't see any value in revisiting)

 

[Lanefan]

The 2024 DM guideline’s emphasis on “will this change/ruling make the game more fun for the the table / improve the game? If not yes to both (and get feedback from your players), don’t it.” Is pretty neat. It spends a lot of words on establishing the idea of mutual respect, communication, ensuring all players are heard, etc. it also explicitly calls out the idea of a social contract around what we’re here to play, and that if you’re here to run an adventure and the players decide they just don’t want that - have a chat, sounds like wires got crossed or you need to re-tailor your hooks.

The problem comes when the two bolded pieces directly clash, which is inevitable sooner or later. Players in general about 99% of the time prefer rule changes that make life easier on they or their characters, while it's often in the long-term interest of the game to go the other direction and make things more difficult.

A simple example might be speed of level advancement. In 5e as written, characters advance pretty fast; and if played roughly to RAW will reach capstone level in - at about 4 sessions per level after 1st - about 18-ish months of regular play. A DM looking to run a longer-term or even perpetual campaign will obviously want to greatly slow down this advancement rate, improving the long-term game at direct cost of the "fun" of levelling up all the time.

A more in-the-moment example comes when a player discovers an exploit or loophole in the rules that breaks something. Fun for the player(s), sure, but in the interests of improving (or maintaining) the game's playability the DM has to shut the loophole down even though the players might want the loophole kept open.

Hence, I see this as poor advice as written, and would rather see the "improve the game" part emphasized with fun as a necessary but secondary consideration.

 

[robertsconley]

OK, sandbox: the GM authors an environment filled with locations. This environment generally allows for fairly free navigation, such that the PCs, acting as a group, have the option to interact with these locations when and if they wish.

That's a part of it, but what is ALSO relevant, and gets pushback is the question of the direction and content of play. I assert that, absent certain types of techniques which are not usually present in trad forms of play that the GM generally has an outsized role here. You can describe that how you want, ideology be damned.

Heck, Doskvol (BitD) has the first part of this, albeit created by the game's developer. But it plays far differently than Majestic Wilderlands due to different techniques used in play to decide what happens, and who decides it. Nobody is attacking sandbox by describing that as more player driven and this, reasonably, higher player autonomy. It's just a logical way for us to describe it.

An outsized role in what, exactly? This is not about challenging your statement. If we want to avoid talking past each other, we need to be specific: what is the referee controlling, and what are the players controlling? Too often “referee control” gets thrown around without clarifying whether it’s about narration, adjudication, pacing, or the setting.

As I’ve said before, just like with railroading, this isn’t a simple on/off switch, it’s a matter of nuance. The impact of referee procedures depends entirely on the creative goals of the campaign.

The point of Blades in the Dark is to recreate the feeling of a heist story in the world of Doskvol. Its mechanics, including flashbacks, flow from that goal. Sharing narrative authority is part of the system’s design, and clearly, that style resonates with many, given how much it's created a family of Powered by the Dark RPGs.

By contrast, the point of my Living World Sandbox campaigns is to present a setting that feels real, where players feel like they’ve been there as their characters. Everything I do flows from that. I handle the World in Motion because the players only have access to what their characters could plausibly know. Their agency comes from choosing what to pursue and how to act, not from framing scenes, asserting authorial control, or sharing the fiction.

And to be clear: player agency in my campaigns goes far beyond just “choosing where to go.” If the players wander into a village of basket weavers and decide they want to become basket weavers, that’s the campaign now. It’s not a joke, I’ve run campaigns where the players built and ran an inn after retiring from mercenary life. The game shifted entirely based on their choices. That’s not limited agency. That’s deep integration into a world that reacts.

 

[AlViking]

All I'm talking about here is basic interaction design. Is there a reliable to get information about the environment before you need to act upon it? Are the consequences of success and failure knowable? Are they a bunch of secret backstory or unknowable setting elements that create chain reactions that result in the sorts of changes we are striving towards resulting in unknowable consequence chains? It's just basic game-ability of the setting and scenario.

Not looking to control things. Just have reliable indications of how to influence them.

Can we please reset this conversation with the idea that I have multiple decades of experience playing and running roleplaying games and not someone you should have to explain basics to.

So keeping to real world for a moment, do I have 0 agency if I'm lost? What if I just pick a direction in hopes of finding something familiar? What if I have a little information and know the general direction of the nearest road? For that matter do I have agency if I'm on roads that were built by someone else?

We make decisions all the time with limited understanding, so how much is needed? If we sometimes lack knowledge does that negate other times when I did have more information to go on? Because these kind of things seem to be what we go round and round about. Most of the time people in my games will have a fair amount of information to make a decision and I'll even go out of my way to remind them or give them appropriate knowledge checks if I think it's appropriate. But sometimes? Sometimes it's little more informed than a flip of the coin because the players don't know any more than the characters do. If you're always guessing or your decisions don't really have any impact? Then you're lacking autonomy. But I don't think autonomy is binary and should not be judged by the fact that sometimes you're basically guessing even when the outcome can be critical.

I'm just trying to establish a baseline because some people have very different definitions of what I consider common terms.