Is it possible to create a sandbox that is a 'living, breathing world'?
No. It is possible to create one that to all intents and purposes resembles such a thing. But we fool ourselves if we imagine that paper, dice, pencils and a bit of imagination can actually produce, rather than create a simulacrum of, genuine complexity.
The reason this comes up is that, in a recent post, I made the claim that it was not possible to set up a sandbox in which the PCs are the good guys responding to threats posed by evildoers without a way of systematising how the evildoers behave. There was some pushback on this in the comments, but I stood by my position. If there is not a way of systematising evildoer behaviour - if one just says to oneself, 'Well, I will just treat the evildoers in the sandbox as though they were PCs and ascribe to them motives and agency accordingly' then one will end up producing what is in effect a railroad. What do I mean by this?
Picture a hexmap. And picture a campaign in which the PCs are the 'goodies' - Knights of the Square Table. Their job is to protect the weak from evildoers. They live in Bamelot, where they serve King Marthur.
For the sake of simplification, let's then say there are three factors of evildoers which you have created. There is the Red Baron and his minions, the Purple Vampire Count, and the Yellow Nosed Dwarfs.
Everything is set up - the hexmap is keyed and populated, Bamelot is filled up with interesting NPCs, etc. The campaign now begins: it's the 1st of January. Now, you want your sandbox to feel as though it is a 'living, breathing world'. So...what happens?
Well, if it is a living, breathing world, what do the Red Baron and his minions, the Purple Vampire Count, and the Yellow Nosed Dwarfs do? They pursue objectives. What are their objectives? Well, let's say the Red Baron, er, wants to abduct King Marthur's daughter because he wants to force her to marry him. And let's say the Purple Vampire Count needs the blood of innocent children to survive. And then let's say the Yellow Nosed Dwarfs want to raid Bamelot's treasure vaults to get King Marthur's gold.
Ok. So what happens now? Hmm. Let's say that the Red Baron sends his flying monkeys to kidnap King Marthur's daughter. And let's say that it will take them about two days to arrive based on distances on the map. Then let's say that the Purple Vampire Count goes off to raid the village of Autumnfield, and it will take five days for him to get there. And then let's say it will take a month for the Yellow Nosed Dwarfs to finish digging their tunnels all the way to Bamelot.
Right. So, in the meantime, what are the PCs doing? Maybe 'going out on patrol'. Maybe consulting the wizard Gerlin about what his soothsaying skills suggest about the emergence of future threats. Let's say they've chosen to go out on patrol. What happens? Hmm. Well, it would be boring if nothing happened when the PCs were out on patrol. So maybe they could get the opportunity to discover clues or hear rumours about flying monkeys, which would then give them the opportunity to pre-empt the Red Baron's kidnap attempt. Or maybe they could encounter some unrelated fourth threat? Or maybe they could get the chance to uncover information that might alert them to the activities of the Purple Vampire Count or the Yellow Nosed Dwarfs...?
Do I need to belabour the point further? I hope not. There is nothing about any of the above that is illegitimate and I do not mean to suggest that this way of playing a game would not be fun, but it is not a 'living, breathing world'. It is a world in which the DM is deciding more or less everything, either on the fly, or in reference to what he has pre-planned or thought up in advance - and, crucially, in light of his own particular tastes. Yes, the PCs do have a bit of agency in how they respond to events as they unfold. But they are really just living out an interaction with whatever the DM happens to think would be appropriate at any given moment. And they will not therefore be interacting with a 'living, breathing world' but in the end reacting to the DM's own implicit or explicit ideas about how he wants the campaign to emerge. What happens does not come about organically but because of what the DM wills, even if what he is willing does take place in response to what the PCs do at a particular time. And what happens in these circumstances will inevitably be led by whatever the DM happens to think would be good, or fun - in reference to his own tastes, desires, and vision.
A little of this is inevitable in a role playing game, as we all know, but by far the more authentic and, I think, rewarding way to simulate the existence of a 'living, breathing world' in such a way as to avoid the DM simply making things up as he goes along is to set up neutral systems of generation and decision-making. Instead of beginning with a Red Baron and a Purple Vampire Count and a tribe of Yellow Nosed Dwarfs, and ascribing to them motives, one instead comes up with a way of generating evildoers and then creating interactions between them, and the world around them, through the use of random tables. One comes up with neutral ways of determining, through the use of dice or other methods, how they pursue their objectives, and when. One creates methods for determining how new threats arise. One creates ways for seasonal and climactic factors to influence events. And so on. One, in other words, systematises as much as possible so as to ensure that the players are not in the end simply 'adventuring' in the DM's own hall of mirrors.
The result is not a 'living, breathing world' either, but it is one that is much less immediately a representation of the DM's own conceptions of what would be best at any given moment. And that is a world which, while not 'living and breathing', does at least contain space for player agency to develop.
Is this what happens when you play mostly D&D and games of that ilk?Post on what can't be done.
ReplyDelete*Sad alien*
I wrote a long rebuttal but then erased it as I don't want to get into a long argument. I'll just say I disagree, and my main point was about running a bounded sandbox with limited options vs creating a secondary world with unlimited options and the gameplay differences that result. I'll leave it at that.
ReplyDeleteI don't want to create the impression I was trying to have an argument. I was elaborating, and I always welcome thoughtful pushback.
DeleteI Meant I was being argumentative, lol, trying to avoid that
Delete@Lance Personally I find the main downside of a sandbox intended to provide unlimited options is that the sheer number of moving parts tends to make it too static. Especially as it grows. I think noisms wants/has a system that can handle an arbitrarily large number of non-player interactions in order to better simulate a secondary world.
DeleteIt doesn't have to be especially large. The crucial point is to do it in a way that makes it feel organic - i.e. not with the DM's hand on the tiller too much.
DeleteI’ve been following these reflections with great interest. As a GM, I’d love to have access to the kind of tools you’re describing. Kevin Crawford seems to have developed something along those lines in Worlds Without Number (see the chapter on Factions and Major Projects, pp. 322–335) and perhaps earlier in his Red Tide: Campaign Sourcebook and Sandbox Toolkit (2011). I’m sure you’re familiar with both.
ReplyDeleteHow do you think these works succeed—or fall short—in providing the kind of systematization you have in mind?
I’m genuinely eager to see this vision take shape. Thank you.
Yes, I like Kevin's work but I would have to go back and look at the books to give a good answer - it's been well over 10 years, probably, since I read the Red Tide and Sandbox Toolkit books, but they were very influential on Yoon-Suin.
DeleteCool! I just checked KC's work. I think that what's really worth re-reading is the section "Factions and the Faction Turn" in Worlds Without Number (pp. 324-6) and related sections. (If one doesn't have WWN, consider that he has a free version of WWN on DriveThruRPG.) The key concept is the faction turn. He goes to say "Every month or so [or after every adventure], the GM should run a faction turn (...)", and provides a procedure for triggering (or not) faction actions that includes an element of randomness. Also important is the section "Background actors" (pp. 334-335), where he talks about NPCs in the faction turn and their actions. I wonder whether that procedure is what your idea of systematization of a living world looks like, more or less. I'm no WWN expert, but it looks promising. Hope you'll take a look at it and tells us what you think!
DeletePS: When I mentioned Red Tide, I was actually thinking of An Echo, Resounding: A Sourcebook for Lordship and War.
ReplyDeleteHa! I've just seen this after replying to the previous comment. You're right - it is An Echo, Resounding. Which just goes to show how long ago it was that I read it and how old and possibly senile I am getting...
DeleteThe Adventure Creation System from Arion Games for Advanced Fighting Fantasy 2E does basically this - gives you a solo RPG or group RPG tool box for creating city, dungeon or wilderness environments, what factions and villains are there, and how does your hero embark on missions against them, with added rules for randomising villains, missions and quests if you exhaust the initial starting patch of 12 starting villains and 24 quests. All in just 430+ pages! :-)
ReplyDeleteNice. I promise to one day engage properly with AFF 2E!
DeleteI'm not seeing how the result would be any different if the DM follows random tables instead of his intuition. Wouldn't the tables have their own implicit or explicit ideas about how the campaign ought to emerge?
ReplyDeleteI'll have to ask my coffee table what it thinks about that. But nested random tables of sufficient complexity can generate more and sundry details - a lot more - than the DM's intuitive choices will alone in the time available, and more importantly they provide a measure of objectivity for the players to grapple with. Playing every faction and NPC as if they were a PC can work up to a point, but while it's perfectly viable in dungeons where the number of things to keep track of is small and very precisely known, it's much less doable on the scale of a hexcrawl. If the DM decides what happens based on what "feels right" it starts to lean away from a sandbox in the direction of narrative play. In a comment on noism's earlier post on the subject, I posted a (very clumsy) hypothetical behind-the-scenes system for how one might generate this kind of thing in a way that's reactive to the PCs' actions independent of the DM's whims, to try to marshal my thoughts on what it might look like.
Deletetl;dr Do you roll for wandering monsters or do they appear when your intuition tells you they should?
What John said, really. Obviously the DM is designing the tables, but that is the beauty of nested random tables - they generate results that he would never come up with himself. Whereas his intuition, by definition, only ever comes up with what he would have come up with himself.
DeleteImplicit in this is something that I think is worth making explicit: coming up with stuff the DM wouldn't have come up with himself isn't only for the player's benefits. Random tables (or something similar) can surprise the DM, much easier that thinking "let me think of something surprising (to myself!)".
DeleteIt's not the only way, and it's not a guaranteed result, but it's a tool.
I write or something similar, because I've had good experiences with intentionally more evocative and colourful tools than traditional D&D tables tend to be: In a Wicked Age's "oracles":
http://www.lumpley.com/oracle/4oracles.php
and, well, everything in Beyond the Wall:
https://www.flatlandgames.com/btw/downloads/
noisms, are you familiar with Beyond the Wall? It seems to me a very close fit to what you've been thinking about lately: OSR-adjacent, heroic rather than roguish, with situations prompted by the game's mechanisms rather than DM's fiat, sandboxy (with a supplement)...
There's a lot of free stuff to sample on the site and on DTRPG, but as a sample of the sample, the adventures, the characters, and the relationships between them are all generated off tables with entries like:
When did you first draw blood? The player to your right was there with you.
You haven’t yet, but might pretend otherwise. The friend to your right often helps you pretend that you are more seasoned than you are, and gains +1 Cha.
or
What makes the site of the quest dangerous?
This is the home of a great mass of beautiful butterflies with razor sharp wings. The air is thick with them.
I got excited about how awesome Beyond the Wall was in general, and forgot to go into how the Furthe Afield supplement actually works towards a proactively heroic sandbox.
DeleteLocations are generated off a simple table as you typically would in a D&D-ish game, 1d8, 1: major city, 2: ancient ruins, 6: monster lair, stuff like that. But it's each player in turn that gets to add one of these, so presumably by definition, the player will be invested into engaging with that bit of the setting.
There's also the expectation and the suggestion, but no hard requirement, that you tie this in with your (or someone else's) character generation tables results: if your "when did you first draw blood?" result was "defending a bridge", maybe that's the ancient ruin you got on the sandbox populating table.
However, to keep this OSR and not free form story game improv, what the player comes up with is what their character only thinks they know about the location. An ability check tells the DM how right or wrong they are.
In another pass, each player now gets to add some minor detail to another player's location.
There's more to the whole book that I summarise here obviously, but even with just this, you end up with a set of characters, each with a tie to another one ("the player to your right" thing), each with some unreliable information about and presumably interested in two different locations. Should be plenty to start an adventure, even in the absence of both an imminent threat or overweening ambition.
Thanks for this. It's interesting you mentioned In a Wicked Age. I played that quite a bit, but long ago, and this suggests it would be a good idea to revisit it.
DeleteI half-like what you describe about Beyond the Wall. It definitely seems to do a good job of remedying the story-game-improv thing of everybody having lots of fun co-creating a setting and then it being totally boring to explore it because you all know everything...
This is going to happen in any campaign structure where the players are a purely reactive force. Sandbox games need players with explicit, actionable goals they want to achieve precisely for the reasons you describe. Which is why I find "the PCs are the good guys" a bad campaign concept, because unless the players are very clearly expected to be proactive in finding and opposing evil, they are going to be a purely reactive force. So, in your example you'd need the players to have the foreknowledge that the Baron, the Count and the Dwarves are evil, *and* they would need some sort of idea about their plans from the first session, so the default option for what to do when there's no immediate problem isn't to wait around for one, or even go looking for something the DM might put in their way, but to actively try to disrupt their opponents' operations.
ReplyDeleteYes, I basically agree - but that's why I am trying to make the case that a 'good guys' campaign can be done by making reactiveness interesting. We all by now know (or should know) how to set up a good proactive 'the PCs are rogues' sandbox. I want to see if it is possible to flip that.
DeleteI've been running a very "reactive" game along these lines for the last two years and it's been really fun. It's gone a little against my sandbox-y principles but the players have been really enjoying the sense of being in a living world. What I think has made it work is having a lot of friendly NPCs to communicate the stakes and world in an organic way.
DeleteNo doubt it relies on putting a lot of thought into enriching the setting - making it as deep and interwoven as one can manage.
DeleteI'm a bit lost. I've read all the posts and the comments and I still don't quite understand what "systematizing" a world means. I used to think that meant a set of opinionated algorithms that a world builder used to determine the interplay of all the elements of the world. I thought that was typically somewhat stochastic, in order to challenge both GM and player imagination.
ReplyDeleteBut some of the comments are now leading me to believe that there is no world-algorithm per se, there is only a collection of disconnected rules for building individual elements in the world. That does not sound like a "system" for the world.
Besides my confusion on this point, I have a bigger question of where everyone thinks published modules like Strahd or Witchlight fit in. I know "railroad" is a dirty word, but I've only ever played with people who wanted the GM to have a cool plot with engaging NPCs, fights, traps, surprises, wilderness exploration, and urban socialization all worked out about one session ahead of time.
Maybe I have been privileged to have always been in such company, and if so, I'm belatedly grateful to all those GMs and players.
The short answer is that it could be either of the things you mentioned. Or something else! By 'systematising' I simply mean that there are systems for generating content that do not rely entirely on the DM just coming up with things from his own mind.
DeleteThat type of campaign - the one you mention, in which the DM has a narrative set up which the players play through - is by far the commonest type of game and perfectly fine and dandy if it's what you like. I tend to find that it places a lot of cognitive load on the DM and is also not very satisfying (to me) to either run, or play in. I prefer a sandbox. Horses for courses!
User51, if you play in more varied company, what you might find is that that while some people like to be treated to an interactive story, with a plot that if not precisely on rails is still largely decided in advance of play to create entertaining situations and opportunities for improv, other people prefer roleplay in the other sense: they like to take on the role of an imaginary character in an imaginary world, and would like that world to respond to them *as if they actually were that character* and not a person playing a part in a story so they can treat the world the same. That sort of play is rewarding in a different way and requires different approaches from the GM (and players). Usually when I've introduced a player to that kind of play they take to it like a duck in water, but I think the GMs for it are pretty rare birds.
DeleteThanks, John - I was being playfully sarcastic about my gaming experience, but that said, I'm still not sure I see how role playing in the sense you are talking about here relates to systematization of the world. I'd like to understand.
DeleteAt the same time, my point still stands: where is the system? If you want to create a living, breathing world, it has to have a form or principle of life. The world going "sideways" suddenly doesn't give it life. The GM (and players, collaboratively in some systems) must somehow integrate the prompts into a system. With that in mind, the nested tables don't appear to reduce cognitive load, they increase it. Sure, I don't have to come up with the twist on my own, but now I have to figure out how it fits in. I don't think that's a bad thing, but I do think it has the systematization reversed.
Anyway, it's really hard for me to follow this conversation intermittently because I keep forgetting what folks have posted elsewhere. I'd love a longer form explanation of the differences between the first and second kind of role playing you mention that has a bearing on world building. In practice, I see very little actual difference between the two at the table. Players slip in and out of persona and interact as the character and as themselves depending on the situation and their moods. It might be more useful as a theoretical than practical distinction, but I'm very, very open to being educated on this.
On the other hand, I do see a very different approach between GM world building and collaborative world building. In such a game, a true World System that gradually reveals the extensive interactions between newly discovered concerns would be extremely helpful.
That is what I'm building.
I don't think I can explain it to you properly because I don't think I, for my part, really understand what exactly you mean by systemisation or some of the other things you're saying, so I don't know what our common touchstones are. I'm trying - it's my fault not yours - but I'm floundering. Also I disagree with noisms that it reduces cognitive load - it front-loads it. My campaigns are prep-heavy at the start but I have to prepare very little between sessions in anticipation of the next, and all my attention during a session is free to be reactive to the players. That's not, however, why I do it - I do it because it provides the experience to my players for which I'm aiming. I'm speaking in vague terms because as I said I'm having trouble understanding exactly where you're coming from.
DeleteI will point out that collaborative world building is anathema to a certain kind of player. I'll say that using tables to determine what happens in the world is a tool more than a principle - I could theoretically run every active NPC in the game world like a PC while keeping track of them manually and striving to keep them acting like real people independent of the players, but it would take me an impossible amount of time and I'd lose the inspirational benefits of a good nested table. I'll say that what that second form of roleplaying needs that the first doesn't necessarily need is a prepped game world the contents of which have a sort of independent existence, aren't mutable in response to players' or GM's whim, such that when players make choices they're interacting with concrete elements with concrete consequences, rather than collaborating on a narrative - the narrative then is the retrospective story of what they've done. That doesn't have to be systems like noisms is describing, which is just trying to improve on the existing model. More commonly it's a relatively static sandbox in which the various adventure locations and NPCs are written down in black ink and not subject to change based on player input or mood, except where that input is actions taken by the PCs within the game world. The GM will, more usually in practice, then provide the active elements out of his or her own brain without using sophisticated tables. That's some of where I'm coming from. I don't know if I'm telling you stuff you already know or not.
I adopt the same sort of approach to you but I would maintain that it does reduce cognitive load in front-loading things. This is because the front-loading is not time pressured. I can spend six months front-loading, if I want to, or a year for that matter, and just start the campaign when I am ready. And then do very little work between weekly sessions. Whereas the DM who does not front-load things is working hard between sessions, under time pressure. The effect on the likelihood of burnout is extreme.
DeleteInstead of having active evil bad guys with active evil plans maybe it would be better to have evil bad guys without the plans. Either that or change timescale to something much slower. Villains don't need to want to destroy the world. Just preserving the status quo where they are on top can be enough. This could make systematization easier too. Instead of BBEGs you could have a table to randomly determine local, not-that-ambitious bad guys for them to work against. Small lords that they can overthrew or thwart as they so choose.
ReplyDeleteI think this would work well with a Kingmaker style game. At low levels they wander around trying to thwart the petty local lords, and then once they are high enough level they could try to start their own enlightened kingdom. If they wanted to, of course.
The Heretic
Rather than plans, I do think they need to have motives, or goals, so that things don't seem inert. But yes, I like the Kingmaker idea.
DeleteI think y'all looking at the wrong bit
ReplyDelete“GM decides yellownoses heist the treasury” vs. “ 7 rolled on events table: yellownoses heist the treasury” is not a meaningful distinction.
What matters is what the possibilities are for the players to interact: a challenge with several solutions, and the possibility of fun failure? Or a railroad thru planned scenes ?
It is a meaningful distinction - it is the difference between the DM imposing his own idea of what would be fun, versus responding to what the table generates. This perhaps does not matter at the level of the one-off occasion but it matters during the course of a campaign as these decisions accumulate. In short, is the campaign emerging organically or just as a collection of decisions by the DM about what would be fun? This is a very big distinction.
DeleteThe GM has already imposed their idea of fun when they created the table.
DeleteExcept they haven't though, because of the way a good random table works. I'm not talking about crappy tables with just a single column of results. I'm talking about properly designed tables which create connections no DM could individually come up with.
DeleteMeh, no need to overthink this past a certain point. Just roll with what works best for you. Its no different than videogames or fiction abstracting things. Nothing in fiction can reflect real life complexity so why bother? Just do what works for you and your group. This is basically ttrpg masturbation.
ReplyDeleteIt's game design. "Trying to improve your cooking is just culinary masturbation, just make what you like". Or painting, or writing, or whatever. Working on your skills isn't just fun, it can lead to more enjoyable and satisfying outcomes.
DeleteWhat John said. 'Just do what works' is the approach to life that a 3-year old, or a dog, adopts. Adults should expect better of themselves.
DeleteAfter reading a lot about the infamous BrOSR and their ideas on patron play, I'm thinking that one of the best solutions to this is to find players separate to the heroic PCs to control the villains, taking that side of things out of the DM's hand. Set everyone up with goals that are likely to bring them into conflict, and maybe it could work, but I feel like there are a lot of DMs who might not enjoy being reduced to the role of referee between multiple sides.
ReplyDeleteWhat is the infamous BrOSR? But anyway, I kind of like this concept - it feels very Amber Diceless.
Delete