There are real challenges to putting these two things together. One challenge comes from the mental perspective that goes along with emergent storytelling. Players in OSR games are often not investing their characters, initially, with a lot of backstory or thought. They prefer it to come out, over time in bits and bobs, as it's relevant, and fun. They want the story to accumulate moving forward, and they don't want to be doing homework. Since they don't view themselves as creating a narrative around their character, they also don't want to read thinly veiled fan-fiction, or massive setting documents. They want to sit down and start playing a game without having to do too much thinking.
Another challenge is framing meaningful player choice. For an OSR style game to be fun, the players need to be making tactically meaningful choices constantly. They need to know what they're dealing with, or have the fact that they don't know be part of the tactical situation they confront. They need to be making real choices about where to go and what to do, with real consequences. This is connected to the fact as well that PC death is an always looming possibility. (It's not fun to die if you aren't dying as a result of the risks you knowingly accepted.) It's also connected to the aversion to adventure paths and railroads of all kinds. This means the players have to know more or less what they're getting into when they choose to do that. The more alien the world, the less tactical information the players start with.
The last challenge is, I think, more general. Brendan S. talks about it very well here. The idea is that the extraordinary pops in part by starting from a baseline of what is known. If everything is wild and out there, then nothing stands out. This is partly an aesthetic point, and partly a point about information overload. If everything is always new and far out, then, in a sense, nothing feels new and far out. Also, it's hard to absorb and keep track of new information without a baseline of normal. What we remember are deviations from normal.
And yet, for all these challenges, my experience is that serious worldbuilding can jive with old school sensibility in play. Indeed, there is a special kind of joy that comes from combining high concept world building with sandbox style play in a tactical old school mode. For two and a half years, I have been running a game where players are dungeon and hexcrawling their way through the dreamlands. And its pretty far out there. They've explored the inverted jungle that hangs from from the bottom of a flying city in Wishery. In that inverted jungle, they've visited the shores of Lake Yannu that hangs like a suspended tear drop. They have opened the first two of the metaphysical locks of the Abyssal Dungeon created to hold the dread crown of the Hidden King, where they've battled the punishment puppets of the Inquisitor's Guild, stuttering nightmare automata made to entertain and terrify. Right now, they're getting into faction play with Phantamorians, travelers from the dreamlands of the dreamlands. This world is a snowflake's snowflake. TRUST ME.
Now I'm going to tell you how I do it. Here are some techniques for overcoming the challenges of combining high concept worldbuilding with old school play.
So You Want to Run a Snowflake Setting in an Old School Style Game?
Try These Techniques!
(1) Home Base in The Known
The first technique involves creating a home base where the players know more or less what's going on. The idea of "town" in D&D has always operated this way to some extent, as a safe place that operates on known principles, as opposed to the unknown howling wilderness or mythical underworld or Caves of Chaos, where reality might operate on different principles, and struggle with unknown foes, strange magics, and cunning traps, is to be had. In a game that employs a setting that is evocative and alien, the presence of this known quantity and safe zone becomes that much more important.
A good use of this technique comes from M.A.R. Barker's classic "Barbarians in The Foreign Quarter" set-up for playing Empire of the Petal Throne. That was a snowflake of a setting if ever there was one: a sword and planet world with highly-mannered cultures that are a melange of Indian, Egyptian, and Meso-American influences, all organized around a baroque religion worshipping alien intelligences. Barker's fix for the epistemic difficulty this posed for players was simple. The PCs start as barbarians fresh off the boat in the city of Jakalla. They are staying at an inn in the Foreigner's Quarter, a kind of polyglot neighborhood full of Conan type rubes, all working as gladiators or doing jobs for shadowy patrons. It's basically a classic sword and sorcery neighborhood lodged in a bizarre Tekumel metropolis. It is assumed that the PCs start knowing nothing about the complexities of the society and religion when they start. But they need to learn quickly in order to survive the ensuing intrigues, so they can make a buck and advance in station and experience as they venture beyond the known boundaries of the Foreigner's Quarter into the cultural unknown of native Jakalla.
I use a structurally identical gambit in my game. The home base of the players is a homebrewed sword and sorcery city-state in the Wilderlands. It operates on normal D&D type principles. But this is not the main place where adventure is to be had, at least not at first. For a door to the dreamlands has opened in the back of Ultan's print shop, and the PCs are the very first people through. This door leads directly into a dungeon in the sewers and catacombs beneath a flying city in the dreamlands. At first I had every session begin and end outside of Ultan's door with a potentially shifting group of players. (Over time this evolved into a more relaxed and organic style of play, with longer and longer forays through the door and a more stable group of players.) The fact that the players had no idea what they were getting into--what kinds of foes were on the other side of the door, who the relevant factions were, or by what principles things operate there, is part of the explicit fun and challenge of the game. Their lack of knowledge is part of the tactically relevant situation they confront, an obstacle that they have to overcome through play like other obstacles, as they foray from the known waking world into the unknown country beyond the vale of sleep.
(2) Setting Information Only in Connection with Game Objectives
People absorb and retain the information that is practical relevant to them. If you are playing an old school style game, it follows that you shouldn't introduce setting material through narrative or info dumps, since the game is not directly about narrative or information gathering. Instead, always tie evocative setting information to the kinds of hooks that move old school games: to sites of adventure, or to potent artifacts, or to rival factions. For example, suppose you want to get a little further into the alien religion of the city of the dreamlands. Well, why not put an inter-dimensional temple to the Unrelenting Archons on the map as a dungeon the party can explore? When the players finally choose to go there (if they do), they'll learn a certain amount of dreamlands theology in a fun dungeony form. Believe me, crawling into a new wing of the dungeon through the hellish portal born from the belly of a statue of Vulgatis, the Archon of fecund and unseemly growth, will make a greater impression on your players than three pages of text wall. Or again, suppose you want to develop the idea that this flying island in the dreamlands is surrounded by the Endless Azure Sea, a reimagining of the sky as a boreal, half aquatic elemental space. Then why not introduce a Prince of the Air as one of the major faction players in the wilderness hanging from the bottom side of the island? The party is bound to tangle with him in one way or the other eventually, and when they do they'll want to know where he comes from, and what his sources of power are. Because they'll want to kill or swindle or avoid him. They'll need to know.
(3) Lavish Descriptions Only Once
When the party first gets somewhere mind blowing, or first sees something amazing, especially if they've been trying to get there for a long time, they have a greater appetite for listening to a description than they normally have when in tactical game mode. They want to hear about it and immerse themselves in an experience of it. When this is likely to happen in a session, you should think about how you're going to describe the place in detail--you should think of this as an essential part of your prep. And you should think of this as your one shot to give such a description. Every time after that first time, you should have at most a couple of sentences to remind the players of what the environment is like they are moving through. Anything more than that will be boring--no matter how cool the environment is they are moving through, or how amazing the appearance of the NPC is they're interacting with, and so on. Seize the opportunity while you have it, and don't overplay your hand on later occasions. This has been very hard for me to learn, and I still stumble from time to time.
For example, the first time the party descended into the inverted White Jungle that hangs from the bottom of the rock of Zyan, I thought in detail about how I would describe the profusion of life, the sights of the foliage, the sounds of animal life, and the fragrant smells of this alien jungle. I thought about what it's like to move in space through it, on a system of ropes, descending ever deeper, walking on lattices of branches that can give way at any moment, and so on. And I shared that with them, because I knew they would be into it. I knew they would want to know what it is was like to be moving through an inverted white jungle in the dreamlands, in part because they worked so hard to get there, and in part because the wow factor of this setting bit. Every time after that that I made the mistake of launching into rich descriptions of the jungle, I watched their eyes glaze over. They were thinking, "Now is not the time for that; now we are trying to get from point A to point B." And they were right. This is a hard but important lesson for someone who wants to do rich world-building in an old school game.
This is part of the answer to Brendan S.' worry about the baseline of normal. The first time I described the jungle it popped. After that it was the new baseline of normal that operated in now familiar mechanical ways. Everyone knew what was what, and that meant that there was now room for the next thing to pop as new and exciting, when its moment came.
(4) Information Gathering in Optional Downtime Threads
If you're going to go into a more discursive mode with setting information, it's crucial that you make it optional for folks. One way we handle that in my game is that if some members of the party want to do research in a library, or to extract information by questioning an NPC about some setting element, then instead of stopping the action in our hangouts game, we save that for later in a "downtime thread", which is basically a social media play by post. This gives me time to figure out answers to their often unanticipated questions. But it also allows a division of labor among the players. Whoever finds it fun to engage in these downtime threads does so, others don't, maybe perusing them before the next game, or maybe not. The basic principle here is that you don't ever force players to learn about anything in a discursive mode. No one is ever forced to undergo excruciating (to them) setting information extraction (e.g. an info dump delivered by NPC monologue). To be sure, sometimes the PCs need to learn about the setting to get what they want to get done (or just because they're curious)--in those cases let a division of labor work where those who enjoy that kind of thing shoulder the burden of it. Memory and knowledge is a collective possession, a pooled resource, in a party of adventurers.
(5) Polished Canon Only After The Fact
If you're into worldbuilding, then chances are you like to keep records. You probably take more notes than most DMs, and may even write down canonical versions of things. This blog exists in order for me to do that. Speaking for myself, creating canonical, polished versions of my own private snowflake is one of the joys of worldbuilding. I want to write things up, I love doing it. There's a special aesthetic pleasure that comes from dressing up my fever dreams so they are presentable for company. What makes worldbulding fun is dwelling with something drawn from your fancy over an extended period of time, until it has a kind of life of its own in your imagination. Writing something up slowly, building it in prose, gives it that kind of reality on the page, and so partakes of the very same pleasure.
Of course, if you have a canonical version of things, you want to share it with your players. And you should! But yet, I just said that you don't ever want to force them to read anything. So here's how I handle that. I try to only put polished content on the blog after the fact, when my players have already learned the relevant information in game. This means that the players are never forced to read my canon, since they already know what they need to know by playing the game. I'll just post a link to it in the community for the game, and say "Here's a polished canonical version of that thing you guys learned about a while back. It has a little bit more backstory. Look at it if you want." Some people do, and some people don't as suits their taste. It's also handy to have those blog entries when, during a game, someone says, "Wait, what did that super-powerful Poem say that we picked up a year ago? Didn't it talk about this dungeon?" and as a DM I can just drop a link to the blog entry on In the Light of Other Moons.
Using these techniques you too can combine your desire to dwell to absurd degrees in castles spun from the spiderwebs and stardust of your fancy with your desire to run open world, tactical old school style games. You can have your fucking weird cake and eat it too. It's delicious.
