anyway.

Now, there's an important wrinkle to this: as long as you're collapsing toward the core game, the filament in the bulb, it'll work, but if you're collapsing some other way, it won't. THEN the game will plain break.

I think this is most common when the MC doesn't have an instinctive handle on the agenda, principles, and what to always say. If the MC instinctively goes to some other way to GM, that's what she'll try to collapse toward, and the whole enterprise will shiver itself to bits.

That's a great way to think about the game when considering hacks, too.  It's important to know at what depth you're reskinning things.

Yes!

I've played with layer 1 only several times over internet chat to introduce people to the basic mechanic. It works great. The added layers just round things out more.

Good stuff.

Very interesting stuff, Vincent! Building on what Weeks said, I think that being layered this way is what helped making hacks possible (or at least productive). The game manages very well to explain to the players how the pieces connect, and thanks to that, changing something isn't akin to a shot in the dark. Burning Wheel has very similar layers to those you described too, and there too they help people not get lost.

Ah yes! In Burning Wheel, Luke calls it hub and spokes, is that right?

Dread House also works that way very intentionally. Since it's intended for adults and kids together, there are certain rules which are the core experience of the game, while others give more complexity. So, like the Muppet show, there is something there at each level of understanding to engage with.

This layering aspect of AW seems more like the details and complexities that develop around a core mechanic. I'm Still haven't read that part, yet I keep using the term. Okay here's a quote:

"A game's core mechanic contains the experiential building blocks of player interactivity. It represents the essential moment-to-moment activity of players, something that is repeated over and over again during a game. During a game, core mechanics create patterns of behaviour, which manifest as experience for players." (p. 317, Salen & Zimmerman)

That implies that it's not only the rules they refer to, but the use of them. But their examples are similar.

So, did you design AW from the core outward?  How much did development of outer layers loop back into and affect the inner layers?

This model describes a Thing That Is. It's useful as a target, and as an evaluative tool for a finished (or nearly finished) product. But does it apply at all how you actually design a game from scratch?

Thanks!

This definitely rings true to my experience.

Also, as I've just started MCing the game semi-regularly, what I've found is that I started with just trying to get the basics right, even though I know I'm missing stuff, and then I'm gradually trying to add on the top layers with each later session.

For example, I don't think I've remembered to use the rules for gangs yet, even though there's been a couple times when I could have. But the moves have guiding us along fine, nevertheless. Still, next time, I don't want to be missing something!

Also, re: what other people are saying about designing... What I think you want to do when hacking AW is to collapse in towards the center and then build a new outer layer that's different from AW but has the same core.  A few hacks have struggled some because the core doesn't seem quite the same or because they've bolted on new layers that don't match up well with the layers below them. But if you collapse in before you build out, that gives you a stronger foundation, I think.

Jonathan, I get how Vincent is using collapse during play, but not how you're using it during the design process.  What does it look like to collapse and then build a new outer level?

Yes Vincent. The first layer is the hub, where the very basic of the game, such as how dice are rolled, and what are obstacles. Then you get the spokes that introduce various basic mechanics, such as advancement, Artha, linked tests, advantage die, etc. Then, the rim is where the various subsystems are, like the circles, Fight! and the Duel of Wits. Finally, both the Magic Burner and the Monster Burner have orientations and principles for creating new parts of the system.

Weeks: I thought I was just rephrasing what you were saying. Like, if you're designing a new playbook (level 4), you start by assuming levels 1-3 stay the same, yeah?

But if you wanna change, for example, how MC moves work, you may have to collapse the system down to just level 2 and rebuild the outer levels, not assuming that they'll necessarily stay the same, since they're built on the foundation of the lower levels. Love Letters may not make any sense anymore if the MC moves work differently, right? Or, at the very least, they won't look the same.

That said, there aren't necessarily clear ties between the levels that necessitate changes. Changing the way MC moves work may not actually change XP. I'm just trying to say that it's important to pay attention to which parts of the lower levels provide the necessary foundation for the layers above them.

Vincent, are you working out how to describe the process of developing a game's procedures?

OK, yeah, gotcha.

One thing that's been nagging at me is that I think I could change the vivid color (unless that's just supposed to note that there is some vivid color) and e.g. "barf forth apocalyptica" and leave nearly everything else intact and the game would work.  Changing _the apocalypse_ on the first level to _fairy tales_, doesn't necessitate completely revising the outer three levels.  I mean, you might change some stuff, but I'm thinking that you almost wouldn't have to at all.  That color shift means you'd access bits differently—different custom moves would be brought into the game, for example, but not the existence of custom moves.

I used the core ("1.") of the game twice in the past with not too much else.
I am happy to see this is not coincidental.

Hey Vincent,

This makes me think of the post you had way back about what the rules text says vs. principled decisions made beyond the rules, vs. ad hoc decisions.

Having transparent (or at least, easy to infer) principles to your rules means that it becomes easy fall back to them if you miss or forget sections.

Vincent-

This is neat! As you explain it, it makes perfect sense.

I am having a bit of trouble formulating a real-world example of how it might work in reverse, though? Is it even POSSIBLE to ignore the stuff underneath, besides the obvious one of imposing your agenda instead of playing to see what happens?

Like, would there be a way for a tactical system goober to ONLY use the advanced combat rules and not the underlying moves and mess it up that way, or something?

Weeks: I'm not sure I get what you mean. Are you saying you could change the color to "fairy tales" and I could still play the Battlebabe and open my brain to the world's Psychic Maelstrom. Because that sounds crazy.

Yeah, that's what I mean.  And some of the changing color or minor higher-order tweaks might include changing "when you open your brain to the world's psychic maelstrom" to "when you peer beneath the veil of reality" and similar.  But it doesn't sound as crazy to me as it does to you.

My concern with this concept is that it treats play as a fixed entity throughout the course of real-life play in real time.

For example, what about the rule in which you get moves from other character types, or the rules concerning removing a character from play? I consider these absolutely fundamental to the ultimate reward system of the game, which concerns what the character becomes (especially in light of the insights about the apocalypse which have undergone some development by this point). Let's take this personal consideration of mine as given for purposes of my following questions.

So: even though we don't see these rules in action until well along the real-world chronology of play, do they count as filament from the beginning? Does this mean we can't talk about the whole room, containing the bulb which contains the filament, until we've played this far? Or does it mean that the whole room, containing the bulb which contains the filament, actually changes into a different construct with different components, once we've played this far?

Ron, one of my first reactions was, "What?  Is the large-scale reward system not at the center of the game?"

After comparing it to the core mechanic of "Roll 2d6, 10+ = hooray, 7-9 = mixed, and 6- = something worse" though, it makes sense to me to say, "well, it's not as central as that."

We can ditch formal character change and still enjoy the way each roll advances the fiction.  Characters still change through making friends and enemies, losing and gaining stuff, etc.

If we ditch the rolls, on the other hand, we can't even really play.

That filament thing seems like the most interesting bit at the moment; it's like having a basic universal mechanic, but it's more than that. The basic mechanic, and the MC responces makes a tight little loop of gameplay. Ok, that's a wrong metaphor, it's very sparse, but it's a lightly defined gameplay loop.

But that loop being formed still depends on various stuff defined elsewhere, like the rhythm of play and implicit action/responce turn structure. But when you loose the rest you can default to it. Probably if you were going to make it into a minigame by itself without moves you'd have to spell out that rhythm explicitly.

I wonder what would have happened if the d20 system had been formed not just around a (semi-) universal resolution mechanic, but accompanied by principles etc.

Because it strikes me that that is what a load of old-school type gms do; you've got skill checks, difficulties, ac and saves, and every now and again you wing it from there using your own aesthetics of challenge and relationships to the players. It's a bulkier core than this, but it still works.

The other nice thing about is it makes me wonder I can look at whether I can decompose other games into the same structure, and by comparing what the other elements add compared to the basic thing, and get clues to fleshing out cute central mechanics of my own.

I'd say this kind of game makes people more likely to drop stuff, because the more baroque interconnected systems make it quite obvious when you're missing stuff. Wheras graceful collapse means that the game is still ok even when you're not using the whole book.

In fact it seems like this optionalness would work against the "introducing the unwelcome" thing. Not sure as I don't have the book, but I'm guessing most of the harshness and unwelcome-adding happens within the filament, and the more developed mechanics moderate that a bit?

Ron, David: You're right! There should be a notion of a reward system in the filament, at least. I didn't mention it but there is one. At that degree of design, there's nothing much formal to it - more like an impulse and commitment that the GM will allow the PCs to genuinely understand things in the world around them, develop their reach and position in the world, and deal with some of the things around them with finality.

It's at the bulb level that this reward impulse gets its mechanical expression. So, taking moves from other playbooks is part of fundamental Apocalypse World, while the particular assortment of moves available to you is part of Apocalypse World as it happens to be this time - the table and the room.

Now, Ron, I don't think that really answers your question. This is just an image of the game's design, not of the game's play. The real-world chronology of play isn't in this picture.

I have semiformed things I want to say about Apocalypse World's text and setup vs Sorcerer's text and setup! Both game texts tell you all about the bulb and all about the table, of course. But Sorcerer's text tells you more about what's possible than Apocalypse World's, and when you sit down to play, there's a process of narrowing Sorcerer down to this time's Sorcerer. Humanity means this; there are these kinds of demons and this is the assortment of powers available to them; these are the stat descriptors you can choose.

Bulb, table & room. In Apocalypse World, when you sit down to play, you get the bulb and the table, you make characters, and you can consider whether and when to bring things in from the room. In Sorcerer, when you sit down to play, you get the bulb and the room, you narrow it down to the table, and you make characters then.

It's neat.

Sorcerer's supplements expand or explore the room of Sorcerer - multiple humanity definitions, relationship maps, pacts, new demon types, the gender rules from Sex, for examples off the top of my head. They aren't just like new Apocalypse World playbooks. New playbooks are just more Apocalypse World, where a new Sorcerer supplement is new Sorcerer.

Does that make any sense at all? Am I completely raving?

Do Sorcerer's minisupplements provide narrowness mostly instead? That would be tidy. Does Demon Cops take the room of possible Sorcerer and say "when you want to play Demon Cops, here's how you narrow Sorcerer down to it"?

Lots of things ...

1. I think the basics of a reward system must be in the filament or it's not a filament. I do not think the details of rolling dice and moving things forwards are always in the filament - or to put it differently, in a good game, they are, but many RPGs are flatly not good.

2. This bulb et cetera idea is clearly still in development. I think it will benefit from considering whether we're equating the innermost pieces with the shortest-term components of play (short-term Techniques and Ephemera in Big Model terms) or not. It's easily read that way. I suggest that this reading wouldn't be all that great.

3. Vincent, I think you're not raving about Sorcerer. You've nailed the distinction between the Sorcerer supplements and the Sorcerer mini-supplements perfectly. I have to repeat, over and over, that the relationship map method is not fundamental to the game and indeed changes the game. I wish now I'd spent more time in 2000-2002 explaining the game without it.

4. I think there's more to the filament level of the reward system in Apocalypse World than you're saying. For Sorcerer, for example, it's about the big four outcomes. For Apocalypse World, I suspect that it's some kind of trade-off among security, hope, and knowledge. In other words, the reward system is *not* "kick ass, road warrior!", any more "dude! demons are real!" would be in Sorcerer.

Best, Ron

Where my brain went with the "concentric" idea was not about time at all.  It's about essentiality.  The outer layers are specific implementations of the inner layers.  Working from an inner layer, without further guidance, a play group would do something.  If they want further guidance, the next layer takes all the somethings they could do and picks one and says, "Do it like this."

Thus, the inner layer of "here's some post-apocalyptic color about life being tough" and "MC, make the character's lives not boring" finds more specific expression in an outer layer.  "How do we make life not boring in this setting?  With fronts!"  The next layer out specifies this even further, with lists of options for what your Front might contain (Grotesque Threat, Warlord Threat, etc.).

As for reward, I think the innermost layer needs a basis for incentives in play, but it needn't look anything like the finished game's formal reward system.  "When you deal with not-boring life in the apocalypse, you make trade-offs between security, hope, and knowledge," could suffice.  Though perhaps that's too unspecific to even register as part of a designed game.  Something linking that to the die rolls and when they're used is probably required.

Perhaps:

Does "danger, scarcity, ignorance" + "not boring" + "roll a lot" + "best/complicated/worse" = reward system?  (That doesn't cover "trade-off", but honestly, I'm not sure if the game does either, beyond multiple fronts with countdown clocks.)

In case it's not obvious, I'm correlating danger, scarcity, ignorance with security, hope, knowledge.

What I haven't made clear until now is: this post is about social context.

Here's the crucial background: 2011-01-10 : Social Context and Design.

"When I designed Apocalypse World for Meg and her ideal playgroup, I designed it to be fun and rewarding for them, yes, and I also designed it to fit into their lives. It gracefully accommodates sick kids, vacations, off-the-ball players, long stretches with no play, changes in the playgroup, individual cycles and levels of engagement and disengagement."

One of the specific ways I designed the game to accommodate individual cycles of engagement is to make it expand and collapse in play like this. This makes it learnable! You don't have to master the rules before you can have fun and rewarding play, and if you haven't mastered the rules, the game doesn't crash and burn on you. While you're learning to play, the game's already working.

Did you go through any particular steps to make sure it works this way?

Sure!

Setting it as a design goal;
Designing rules I thought would meet that goal (along with the rest);
Testing, testing, testing;
Redesigning as required.

No unusual steps, but yes, all the usual steps.