Episode 53. A game design interview podcast, from Soren Johnson of Mohawk Games.
Maps Not Meters
Listened to this episode of Soren Johnson’s interview podcast where he got Inkle developer Jon Ingold to spill his secrets. One thing struck me as incredible brilliant, yet in retrospect, completely obvious.
In making their interactive fiction version of Frankenstein, which played out like a novel being written ad hoc in response to player choices, players felt like none of their choices mattered--even though they very much did. Ingold attributed this phenomena to the output UI display of a novel seeming linear.
Their follow up, Sorcery, addressed these concerns by having a map-- which is technically just a UI visualization of the story flow. By marking areas on the map, and charting their course, and importantly, showing where players didn’t go, player’s immediately understood that they were being asked to make tough, consequential choices.
This is the same reason that games with skills now organize them into vizualized three-pronged trees (much replicated to the point of exhaustion); the non-narrative equivalent to a map.
One of the reason my 2D platformer/friendship&dating-sim game prototype failed to get anywhere was that NPC relationships seemed completely arbitrary. It seemed like NPCs had scripted behaviours. I couldn’t communicate to the player that their own subtle behaviours were being recorded by the NPC’s, that their choices were remembered, or that NPCs could have responded differently.
What I’m getting at is, what I needed was a map to chart player’s relationship with NPCs. A map is a weird way to phrase it, but It needs to be a map because it’s progress occurs across two dimensions. Instead of, say, the relationship meter that’s in dating sim like games such as Persona, which is one dimensional.
Meters are one dimensional.
The other benefit of using a map instead of a meter is that you get something as organic as a story-- or a relationship. A map is useful for content that is fundamentally non-linear, complicated, and organic. The problem is, Inkle came up with the brilliant conceit that a story map could be an actual map.
But what the hell is a character relationship map? Maybe there isn’t an elegant solution for this. Maybe it’s the kind of branching paths flow chart found in certain visual novel games. Which, as it turns out, looks exactly like one of those three-pronged skill tries. Go figure.
postscript After thiningk about this for a day, I think I actually have the answer to the problem of finding a clever metaphor to visually represent complicated NPC relationships and player choice. It’s: put them on a map. No really: has an NPC been affected by player behaviour? MOVE THE NPC. If you have a map, mark that NPC at this new location. And have that location reflect the player decision. Well, so this is a bit obvious, perhaps much like Ingold’s observation that a map is the perfect visual metaphor for a player-facing story flow-chart. For whatever reasons, I think of important story-related NPCs as stationary because most games fix them to a position so that the player can find them. Ignoring solving that problem, it’s hard to show the NPC as posessing an interior life, if, they are “sad widow character” who stays in side their apartment, or “gruff shopkeeper” who only lives to sell goods. Perhaps if you’ve contributed to their unahappiness they are found in the tavern drinking their sorrows. Or they go somewhere more out of the way specifically to avoid you. tl;dr on the entire thing: you can communicate possibility easier in a 3D world (or in a 2D representation of a 3D world, a map) for both player-centric story (the critical path) or for relationships with NPCs. Just, anything to get us away from putting actual flowcharts into games.