So an interesting problem now (to me) is where the commentary goes. There is commentary, but it’s generally not here, which is fine because this isn’t an open forum really. Even if I open up comments it doesn’t feel like one — it feels like someone elses space. But everything is crosslinked to whatever WordPress’s limited linking capabilities provides, so that’s G+, Twitter, and Facebook basically. Not linking G+ because it’s dying nor Facebook because I don’t check it often.

Anyway, that’s where the good commentary has been.

I’ve also manually linked to Reddit and Mastodon. That’s had good commentary too but I fear that manual linking isn’t something I’m likely to maintain. And Reddit doesn’t seem like a place that will tolerate a link a day unless they are crazy interesting. If you dig something and link it there, that’s cool of course. Same with, say, — doesn’t seem appropriate somehow to link my own stuff there.

Another possibility is the VSCA discord. That’s always on but I don’t know to what degree I’m interested in moderating it. I ban pretty reflexively anyone that aggravates and I probably aggravate too easily. If discord’s your bag, the RPG Talk discord is also pretty wonderful, packed with cool people.

So in the spirit of self-reflexivity, where would you most like to engage? I’ll hunt down comments in all these places, but if there’s one place that most people go, I’ll be there.

One thing I’m not interested in is any sort of tribalism. If you have a hard-on for some place, that’s awesome. But if it means you have a hate-on for some other place, I’m not interested.

So I guess the bottom line is that I’ll follow up wherever I know there’s discussion. If you start it, hook me up and I’ll participate. Hopefully something will shake out.

old ramblings #5

Last old post. This one about killing.

So violence in games is often treated as a special case, that is it gets a fun subsystem. There are a number of ways we can react to this in design.

We can counter it with fun subsystems for other kinds of resolution. Subsystems that have equal weight in mechanism, tactics, and so on. This makes them, in theory, as fun an alternative: you still get to play a tactical game with it.

We can neutralize it by designing the system so that violence and non-violence (and all the shades of gray) have the same mechanism and impact. So if combat is never “special” then maybe violence will be an equally compelling option among many.

But the heart of the problem feels like it might be narrative rather than mechanical to me, and not about the general case of violence but the specific case of killing: whatever the mechanism for resolution, killing an opponent is typically a permanent solution within the fiction of the game. And pretty much nothing else is. So morality aside, it’s the best tactical solution since you don’t need to revisit the problem.

Morality aside.

What can we do about that?

Some ideas here. We could make killing expensive. So mechanically it’s the same (say) as everything else but the permanence of the solution is balanced against a cost to the killer. This opens some doors — psychological harm from being a killer and exception-based powers to ignore it (the sociopath stunt). Maybe not all doors I’d want to open, mind you. Reign does this with its haunting, but not for all killing. But when you murder someone in cold blood, their spirit haunts you, yelling in your ear forever and making it hard for you to do what you do. Niftily this is a narrative solution to a narrative problem: no mechanism enforces it.

You could make killing impermanent. If everyone you kill always comes back to life later, stronger, and very angry then you might reconsider it as an option. Especially since it not only no longer has the advantage of permanence but also comes with a price tag. This is a setting-dependent solution, though, and would have a pretty deep impact on your setting. Not for everyone. Pretty cool though. Does anything out there already do this?

You could make all resolutions permanent. Something like a weapons-grade let it ride: if you resolve a conflict by any means, it stays resolved. You talk that person out of stealing your stuff? It stays unstolen. You escaped your captors? You escaped. They will never find you. This is not entirely satisfying as it has very gray boundaries and it limits the kind of big bad villain story arcs you might want to tell. But I bet there’s a way around it.

For me, none of these are quite right (though mechanical harm for being a killer is something I’ll use). One powerful thing I like to do is actually extrinsic to the game: create a play environment where it feels wrong to kill things. Where everyone has a life and loves and killing is just never okay. It might be necessary, but even when it is, it’s never okay. I have no idea how to codify that or to otherwise make it available to you. I’m still thinking about that.

old ramblings #4

From late September, just before publication.

Starting is always a creative obstacle. The blank page gives you no clues as to how to begin; it all has to come out of your head.

So we have a good many “oracles” to keep you from facing the blank page. Now these aren’t adventures or even adventure hooks (well maybe) but just ideas, imagery, issues, to spur you to fill out that ref prep form with confidence.

The Association. The most important oracle at the beginning is the Association. This is the organization that all of the player characters belong to. It has three characteristics that are rolled randomly and then the players need to make sense of them, using them to create a company or an institution that their characters will be dedicated to. And that association starts with a debt: a problem that the characters are immediately tasked with solving. This problem is vague (like “pursued by an enemy”) and the details are all yours. But it will be intrusive. When you try to find an idea to start some shit, you’ll start here. At first.

The King. The King is the wrong king. They are oppressive. But you’ll also roll for their details: what kind of primate are they and what is their specific failing? While there are some notes about how to apply oppression to the scenario, the type of failure of the King is your oracle to paint the campaign a special colour. Is the King a colonialist? A hedonist? Or maybe an ivory tower intellectual? The whole nature of the world will change. Or at least the opposition.

The Land. Each Land that the characters visit is painted using a set of oracles that suggest size, proximity to others, altitude, and trade connections. This is enough to make you image that Land sufficiently to find the adventure in it. It’s warmer and more political at high altitudes. It’s colder and more self-interested lower down. And right near the bottom it’s freezing and full of political dissidents.

Armed with these three random elements, it’s certain that your campaign will have motivation, opposition, and colour that’s novel each time you start. And after that I guarantee the game will go its own route. That’s what it does.


I’ve been using a simple technique for game development lately that has paid off extremely well: scaffolding. Now I’m certain this is not novel and that someone smarter than me has already elaborated it in a more beautiful blog under a more euphonic name. Perhaps even in a book. But since you’re reading this now and I’m writing it now (whenever now is in this context) here’s my personal (re)discovery.

A rough sketch of a spaceship somehow evokes “scaffolding”. It doesn’t really but it’s a cool spaceship.

The idea is simple and predicated on two things that may or may not be true for you. They are true for me:

  • a piece of a game can be tested without the presence of the whole game
  • setting elements can be tested without any particular system

Now this isn’t an article that will claim that there is no such thing as an integrated system (where all pieces work together to create the experiences) and nor is it going to claim that setting and system are separate. Far from it. But I do think that you can reveal information about little bits in isolation from each other. Valuable information. The whole is necessary to create the final experience, but the bits can be tested individually and discarded early if they suck.

So I build scaffold games. These exist only ephemerally, and only to test something. I might take one and build it into a real game. I might not. But the principle is:


That is, game elements that haven’t got to the table haven’t been tested. You have to play. But how do you play when you have no game? Invent enough game.

So, for example, let’s say you have a cool piece of a setting idea. Giants with villages for heads. What happens when you encounter one? What do they want? How do they impact the rest of the world? You can invent this and hope it works. Or you can test it. The advantage to testing it is that not only does it get tested, but it also tends to get elaborated in play. Often in ways you would not have anticipated.

So go to the table with the setting idea, set up a scene, and just roll a d6 for resolution as called for. That’s your scaffold: that d6. It’s not your system, it’s just there to make the other element testable.

Conversely, let’s say you have a rough idea for a resolution method and no particular setting or them. Invent the barest, stupidest, most obvious setting and drop some players in the middle of it with only the new system (incomplete as it is; wing the missing bits) and drive the system around.

So go to the table with the system idea, set up an arbitrary scene, and get going. Hit the ground running. That stupid, obvious setting is your scaffold.

These events are scaffold games. They aren’t your finished product. They might not even turn out to be part of your finished product. But they let you talk about your development in terms of play rather than hypotheticals. It’s one thing to say (and believe) that this mechanism causes this, this, and this other kind of tension. It’s much more valuable to say it actually happened at the table. Or didn’t.

And you know what else you can avoid? Calculating odds. Deciding what odds you want and then contriving dice to make that happen has a missing premise that you should expose: the assumption that those odds are fun. Play will tell you whether the dice are fun without requiring that you do the math (or more likely fire up AnyDice). You can always come back to the math. But if it’s fun then you don’t really care.

So is that obvious? It wasn’t for me. Build a minimum game that plays to test the bit you need to test.

Digging’s hard work. But you gotta dig.

One side effect is that a scaffold can sometimes become a load-bearing structure. This is what happened with the Soft Horizon system and what will soon be Sand Dogs: the desert-dwelling dieselpunks were just a dumb idea to let me quickly test the core system. And I kind of got into it. Did some drawing. Liked the characters, as distinct from just their representation. So I kind of got a second game for free just by building the envelope around the scaffold.

And of course there’s something buried in here: I don’t really want to talk about design unless we can frame it in terms of play. This isn’t news. But it’s not a principle that’s used enough any more. The first response to “what do you think of my dice game; are the probabilities right?” should be, “well how does it play?”

Play your damned games.

old ramblings #3

From September 2018.

One of the objectives of The King Machine is to make reffing the game as easy as possible. I love making maps and spending time on prep, but I don’t want to need to do it. I also love ad-libbing, ideally based on the other players’ actions, but I do it better when I have cues. So we built in all these things.

Risk. Every resolution roll has a risk that’s selected by the referee. And many rolls will realize these risks even if they succeed. So the risks can propel the narrative in new direction: if you’re being chased and risk spillover, for example, you might successfully get away but injure or kill a bystander in the process. That’s going to have consequences I can build into future scenes!

Cues. Sometimes the back-and-forth at the table stalls or slows. Everyone experiences this. Whenever this happens, the ref has a list of options to jump-start the narrative. Things like start some shit: introduce an immediate threat. Or recall a missed hook: that hint you planted earlier that no one grabbed? Make it important now. These are things any good ref depends on anyway, but we’ll list them explicitly because not everyone is a veteran. And not every veteran remembers to mix it up a little and try something different. It’s nice to have a list.

Deadlines. When things get urgent, we make them mechanically urgent. You are now fighting a deadline and there are specific things you might do during the game that will mitigate that deadline or, more likely, tick that clock one closer. This gets people focused on the action. Characters no choosing a specific direction and acting on it? Make it urgent. Make every action count towards or against the deadline.

Mechanized difficulty: The ref never sets a difficulty level so you don’t have to worry about setting it correctly or fairly. The system handles that. You just set the scene, listen, and then guide the conflict with a risk. The story will flow out of that with minimal direction from the ref. There’s all the usual room for creativity, but you don’t need to fabricate numbers to oppose an action. Just get on with the action.

a little business

The energy and space (notional) I have to create comes from Patreon. Without my incredible backers, lengthily thanked in every 2018 book, the VSCA would not be coming back the way it is to bring games I care about. So please consider dropping a regular penny in the jar.

Sales through DriveThruRPG (RPGNow) and Lulu, especially of physical books (just because I like making books a lot, not because I make more on them. I don’t), are also a big fat motivator. Seeing people buy, read, discuss (for heaven’s sake link me into that conversation!), review, and play these games motivates me to make more. You’ll also find links to some free toys at the VSCA main page if you dig.

All this business stuff, this money, does is make room in my life for me to write and draw and construct and playtest. It lets me wonder about ways to adapt role-playing games to my changing life and my changing needs and since we are all growing older, I think these adaptations will have an impact on you as well. So that’s the semi-regular plug: if you dig what you’re seeing, please consider supporting it. Get a book, drop a buck in the jar, or re-share something you read and thought was nifty.

Addendum: if you’re from Sweden, could you leave a comment and let me know what drew you here? The majority of my visitors this week are from Sweden. I’m curious!

magicians in the soft horizon

Sometimes those aetheric goggles show you more than you wanted to see.

Magic is always a pain in the ass. How do you balance it, how do you differentiate it, how do you keep it under control: these all wind up either driving a boring system or a subsystem with all the complexity that implies. Games that are about magic fare better since you don’t need to differentiate it at all and you can go berserk with the system.

So Soft Horizon is obviously tending towards the simple. A subsystem is right out. So how can we manipulate the existing system, following the style rules of change being additive and contextual, to get a magic system?

First off, anything can be magical by describing it as magical. This is the undifferentiated method: magic is just magical narration. This is fine if you want to magical incidentally in the world. But what if the world demands magic? Then you want something meatier. So let’s balance effect and cost to get something we didn’t get before. And let’s try to make it get better in a useful fashion without adding too many new rules.

We’ll lean on risk. In Soft Horizon games every conflict has a risk. And we’ll make magic a specialization. So if you add a specialization with the prefix sorcerous (say you specialize violence with sorcerous six-shooter) then that specialization starts at d8 instead of the usual d6. That’s your magical benefit: bigger die. And the narrative oomph of having a sorcerous revolver of course.

Now the cost. We’ll define sorcerous as running the additional risk of self-harm. So whenever you bring that die into play, you add harm to the existing risk for the conflict. If it was originally waste, now it’s waste and harm. If you screw up, you get injured in addition to all the other crap that goes down. Just to cover special cases, if the risk was already harm, now its scope expands: everyone in your group gets wounded.

Of course as soon as you build a special case you want to generalize. So what do the other risks look like as schools of magic?

Spillover. Demonic magic. Every time you use magic innocent people get hurt. If that’s already a risk then you cause a plague.

Waste. Well you gotta draw your power from somewhere. Call down a pillar of fire on that swarm of scarabs and if the risk gets realized, it uses all the gas in your car and the air in its tires.

Revelation. A seer or oracle perhaps. When you weave your magic you have visions and they are always bad and true.

…and so on. They all work. Of course if you roll well the risk is not realized. You get your effect without paying for it. And as your die gets bigger, the chance of this goes down. So advancement rules as they already exist take care of reducing negative effects of your magic.