My gold standard for setting driven play is Amber, a fact which should come as a surprise to no one.

But what’s interesting is that there’s actually very little setting in Amber. This is the function of a few elements.

• Zelazny’s writing style is very sparse – colorful, but with a minimum of detail. He paints people and places with a few bold strokes, then just leaves it as that. As such, Amber has some geography – a castle, dungeons, a few place names – but it’s very loose.
• Because the core conceit of the series is multi-dimensional, much of fiction (and play) takes place in an infinite range of other places. What’s more, the very nature of the character’s power mean that details just aren’t that important, because they’re subject to change. A strong core idea is all you need for an entire world.
• At the center of everything is a large family, and the main drivers of play are tied to those characters. Importantly, the players are also tied to those characters – usually as their children – and the result is that everyone enters play connected along multiple vectors. And Zelazny’s sparse style paid off hugely in this regard, as each family member is pretty memorable with only a few details.

The last one is probably the trickiest and the most potent. “You’re all family” is the relationship equivalent of setting a game in the modern day. It sets up a whole array of dynamics and opinions right from the outset, in a way that most of us are familiar with to one extent or another. However, actual extended families can be tricky to put into play, so it’s often necessary to use a family substitute.[1]

But even if you can come up with another structure, and there are many, then the question becomes how you can convey those characters without huge hassle. And this is one of those horrible bugbears where the answer seems to be “really good writing and art”. The reality is that it’s possible to craft a really striking character in a paragraph or two, but it’s hard, and doing it well would probably require more rounds of revision and testing than most RPGs ever consider.

In fact, the very idea of playtesting a setting seems like a total goofball idea, but I wonder if it might actually be something of a necessity. It’s a weird thought in a lot of ways – I’m not even 100% sure how it would be done, but if you could set up a solid feedback loop (and, nontrivially, arrange enough testers) you might be able to gauge who people respond to and, importantly, write less material.

And that’s even more counterintuitive. Page could turns into prices, so it’s hard to justify writing less for a product. Hell, Setting books are usually bursting at the seams. But I definitely feel like we’ve done as much “more” as we can possibly handle, so now I’m wondering how we can do “less”.

While I have specific demands for maps in games, the issue if more muddled in pure-setting products, most famously defined by the boxed sets for things like Greyhawk and the Forgotten Realms. These are well-loved products, and their design sensibilities have influenced many setting products that followed, but they merit some examination. The questions that intrigues me is what the purpose of these products really is – are they designed to be played in, or are they just bookshelves, waiting to be filled?

To understand the distinction, let’s look at Greyhawk vs. the Realms. Greyhawk was designed to be played in if only because there wasn’t much choice otherwise. It was such an early product that it’s design as almost pure map plus gazetteer made a lot of sense. You could take it, pick a spot, and play your game. Yet even bearing that in mind, one of the interesting things about Greyhawk was that it provided a context for the locations of published adventures. I know that sent a little thrill down my spine the first time I discovered a note indicating which hex a particular adventure was taking place in. That was, I think, an inidcation of things to come.
The Forgotten Realms was subtly different. Not so much in content; there were some changes, but not enough to really change the type of product. It was, however, a different beast from a commercial perspective. The Realms were a container, one able to hold any number of smaller supplements, novels, video-games and lord-knows what else. In that sense, the initial boxed set was a skeleton to be steadily fleshed out, and TSR delivered on that promise. The realms might have been thin and disconnected at the outset, but they filled it in admirably.

(At this point there’s a requirement for an obligatory nod to the GMs of old whose insane notes provided the basis for these settings, and I hereby provide it, but only grudgingly. I applaud their creativity while I bemoan the fact that they convinced generations that binders full of data no one gives a crap about were going to be the next big thing.)

This kitchen-sink model has had a huge impact on setting design, but it’s fascinating to me because it’s so much at odds with the realities of play as I’ve seen them. By and large, I have seen games either drill down into a specific are or, if covering a broad area, touch upon the setting very lightly. That is to say, real games tend to be narrow and deep or broad and shallow, but the average boxed set aims to be broad and medium-deep, thereby serving neither need.

Now, product do exist to support these actual approaches. Many settings have “Gazeteers” or similar books – very slim (maybe 32 pages) volumes providing a very high level view of the setting, and almost all setting that produce subsequent books produce more detailed region books, those that zoom in on a specific area. Those products are much closer to the actual usage patterns of play, but they are secondary products.

This suggests a fairly cynical purpose for the main boxed set, which I alluded to above. It’s the stake in the ground that allows a publisher to tether those more-useful document to. The big setting with its big map is not necessarily there to be used on its own, rather, it’s a menu of sorts. It’s a resource that lets you find the glittering object that catches your eye and choose which area you want to zoom in on. At that point, perhaps you will flesh it out yourself, but ideally (from the publisher’s POV, at least) you’ll buy the book that deals with the part that caught your eye. Better still, you’ll be curious about a few areas, and pick up several books!

Does this sound like I’m asserting that the default model of presenting a setting is tooled more towards selling books than use in play? Well, yes, I suppose I am. Not to say that you can’t do both – I can think of several great examples where those two ideas have dovetailed awesomely (Birthright being the absolute best, in this regard). Plus, the people writing these things are almost always doing it out of genuine love, and that tends to muck with more cynical goals.

In some ways, it’s been very fortunate for the hobby that we’re so bad at business.

I did not expect to get to Dreamation this year, but My wonderful wife surprised me with a window of time that allowed me to day-trip up for the last day of the convention. Didn’t play anything, and didn’t get to see everyone, but still very much enjoyed myself.

Part of my fun was sitting down to talk with Brennan Taylor. Nominally, it was for a podcast, and he’s got 10 minutes of me talking about hacking games that will no doubt be unleashed on an unsuspecting world at some point down the line, but after that, we got to talking about setting. This was a little bit catalyzed by talking about his forthcoming game, Bulldogs, which I’m pretty excited for, but it was also kind of general. From this conversation came three points that I’d like to share with anyone looking to write setting material, mostly out of a selfish desire to make it all interesting to read.

First: Write Around The Holes

As a tribe, we’re smart people who like to show off how smart we are, and that lends itself to a completust streak. We also like our books, and we have an instinctive understanding that world building is something which should usually yield to the necessities of story. These elements combine to create complete, cohesive settings with satisfying narrative arcs and it’s terrible.

If you’re writing a setting for play, you are not telling a story – you are creating an opportunity for someone else. This doesn’t mean that nothing should be going on – quite the contrary, things should be hopping – but they should not be resolved. It is ok to set things up and then just stop. Our instincts may want to resist, but by doing so we create the opportunity for people to fill those holes with play, which seems rather the point.

Settings that fail to do this may be interesting and colorful, and they may be entirely satisfying as a pure backdrop of play (that is to say, interchangeable dungeon storage) but if you want the setting to matter, leave the holes in place and trust someone else to fill them.

Second: Playable Is Better Than Clever

In jokes, word play and other setting elements that have nothing to do with play and everything with the author squeezing in something cunning need to be used with the utmost caution. This is not to say text needs to be all serious – jokes which include the reader or the occasional easter egg can be fine, but if you’re just showing off, knock it off.

Short point, I know, but there it is.

Third: The Three Sentence Rules

Virtually any setting element can be fruitfully described in three sentences: a description, a distinction and a hook (they don’t literally need to be three sentences, but you get the idea). For example: Varn Kasi is an Ethari crime lord (description) based out of a gentlemen’s club overseeing the Alverado harbor (distinction). He’s making preparations for war against the Dwarf gangs horning in on his silver dust trade (hook).

That is not a lot of information, but it’s enough to play with. A GM can absorb that from a book and very quickly plug it into actual play. You really don’t want to use less information than those three sentences – doing so can cause the reader to wonder why you’re bothering to mention it at all. But on the other hand, you want to have a very good reason to provide any more information than that. Obviously, you will want to do so for the central elements of your setting, but it is worth challenging yourself to determine what more you really need to add. See, the three sentences is a sweet spot. Every additional piece of information you add takes longer to absorb and – almost paradoxically – often puts more limits on the element in question.

This need not be so. If the extra information is kept sharp and focused on how it will come out at the table, it will probably work out. And the best way to make that happen is to really ask yourself how you’re improving on those three sentence.

So, three guidelines. They’re not hard and fast rules, and they are probably a bad match when you are talking about worldbuilding in fiction, but gaming has different priorities, and it’s important to remember that a game’s setting should not be a reformatted novel. If you want to write a novel, then write a novel. Heck, it might even make for a great setting to game in once your done. But if you’re writing setting, then write for play.

I floated another question on twitter, since it tends to be an interesting sounding board for idea seeds, and this one’s sticking. That this is profoundly connected to my previous thinking on structure will either be incredibly clear or incredibly oblique – not sure which yet.

I asked whether any RPG settings had revealed all their secrets in the initial book, and then used supplements to respond to those secrets and reveals (rather than introduce new ones). This is not quite the same thing as asking which games have no metaplot – rather, it’s a structural question about how information is handled.

It is easy to find examples of RPGs that don’t do this. 7th Sea, for example, has placeholders for secrets in the core book (The White Plague or Die Kreuzritter for example) which are explained in later supplements. It also has non-flagged secrets scattered across its books.

While I’m not fond of this model, I acknowledge it’s practical advantages as a publisher. In short: it drives sales. People like being in the know, and this is a great way to tap that. But, thankfully, I see it a little bit less frequently these days. This is not to say we see no supplement cascades, but they’re often structured a bit differently. The new World of Darkness stuff, for example, is still supplementastic, but it’s more modular in its design. If you skip a book, you have a decent sense of what you’re skipping.

But the question of games that have turned the model of secrets on its ear, elicited some interesting responses, as well as some surprises, and I want to flag a few that came up.

Torg – This one got mentioned a lot, and I’m going to have to ask Fred how much was revealed in the boxed set because I have no idea. I have a great conceptual love of Torg, but it’s based entirely on people telling me about the game. I have never had a chance to play or read it.

Conspiracy X – Another one I haven’t read, so I have no idea. Any thoughts?

Feng Shui – This is probably the single best example of what I was thinking of. The supplements for the factions introduce ideas and plot hooks, but nothing that essentially changes things as presented in the core game.[1] Some of this was enabled by the fantastic flexibility of the setting and the general tone of the game, but there was also a decision to go in this directions which deserves credit.

Vampire: The Masquerade – My first response to this was surprise. Vampire is, after all, the poster child for the triumph of Metaplot. But thinking about it a bit, I realized that mostly came later. The core book is actually pretty open about things and there was no _necessity_ that things go in that direction. That they did was probably a good commercial decision, but it’s an interesting illustration of where these things happen. To see why consider…

Armageddon – This is pretty much a placeholder for most of the games out there which came in one book with no real expectation of supplements. The setting’s meaty enough that there COULD be supplements, but everything’s pretty much laid out on the table from square one.

Call of Cthulhu – This is an interesting one for reasons that are very relevant to the Dresden Files – how does external source material work into the equation? CoC could be said to be complete because all the material is out there[2], but that might also be viewed as a bit of a cheat.

In the end, there were more good answers than I anticipated, and I’m going to have to keep them in mind as I consider how one produces a setting today.

1 – At least until Friends of the Dragon (EDIT: Whups, meant Golden Comeback), where the need to introduce cooler-than-thou NPCs started messing with things. RPG writers – your NPCs will never be cool for the things they did to the setting. You take opportunities and focus away from the people who are actually playing your game.

2 – And I never noticed this until now, but the Cthulhu crowd does not seem to vigorously reimagine cosmology the way the Amber crowd does. Curious.

Equilibrium is a very tempting state in setting design. I has lots of fun trappings like a balance of power and broad opportunities for commerce and travel, and more importantly it lets the author really drill down into the things that make the setting interesting (at least in his eyes) without them getting all broken or overly complicated. The problem is that while this is very compelling from a perspective of creation, it’s a bad approach from the perspective of play.

Interesting things when systems fall out of equilibrium. Change, wars, revolutions, reformations and pretty much everything else, and all of these things are fertile grounds for play. When a game takes place someplace out of equilibrium, it has a sense of inertia and movement that is what many railroading games are trying to capture without realizing it. It’s a sense that the world if a moving, and you better keep up. By leaving things in equilibrium, that energy goes to waste.

As with many failings in setting design, I tend to pin this one on the terrible nature of social studies textbooks, which are the only model that many people have when it comes time to write up a setting. Having history presented in clean, digestible chunks warps the mind into thinking that’s how things should be, and overlooks both the narrative (which moves) and the reality (which is messy) in favor of simplicity and the least common denominator.

The trouble with equilibrium is, of course, that it has no trouble at all. If there was no game, things would proceed pretty much as they have, and even if there is a game, it’s likely to have a small impact as things play out. Now, a low-impact game may be desirable. Many styles of play emulate fiction where the main characters mostly drink and fight and while they may do hugely heroic things or even save the world, they’re likely to do it in ways no one particularly notices. Thinks like the earlier stories of Fafhrd an the Grey Mouser. But in such games, setting is usually designed very loosely, in broad strokes, with whole swaths of territory easily summed up in a sentence or two. Adventure is found in the exceptions and anomalies. Such a setting may well be at equilibrium, but it would also be almost silly for it to be more than a collection of notes, and maybe a really cool map.[1]

It is also possible to bring change to a system in equilibrium through the agency of the characters, especially if they’re the chosen ones or whatnot, but it’s a very brute force solution. It’s very nice and empowering, but it’s also not much of an improvement – unless the world responds to the change in a way that creates tension and problems, it’s just a kind of showpiece.

There’s been one interesting trend in setting design to address this, something I’ll call aftermath design. The idea is that in the setting, something big has just happened, such as the emperor being killed or whatnot, and the setting is going through changes as it sorts this out. This is a promising idea, but it bumps up against old habits. Too often, that change occurs (before play begins, natch) and it is then the ONLY change that’s going to ever happen. It’s just a push towards a new equilibrium.

And that, there, reveals the true rub. There is absolutely a tendency of system to move towards equilibrium, but even if they reach it, they don’t sustain it. Change is ongoing. For a GM, this is intensely liberating. For a setting designer, this creates a challenge of how to express that dynamic in a useful. Which is the thing I now find myself chewing on.

1 – Not to say this stops people from getting encyclopedic about it, but it’s a different beast.