Mailbag: Teaching Spatial Storytelling

A Twitter follower asked me for resources to teach students to pair space and story in a meaningful way, and they were already familiar with my article Plot-Shaped Level Design.

To state what will be extremely obvious to some of my readers, but probably new to others: this is classic craft territory for parser IF, where maps are generally developed in tandem with plot and puzzles.

The primacy of the map, in this tradition, is why Inform had a map index much earlier in its development than it had a scene index: charting the space, together with its doors and access points, was understood as more critical (and also easier to do programmatically) than diagramming a CYOA-style node structure.

Classic text adventures rarely experimented with treating space as continuous rather than room-based, even though the possibility of doing so cropped up in discussion at least as early as 1991, with another discussion in 1996. Some of that may have had to do with technical challenges, genre convention, and the relative difficulty of expressing quantitative information in prose. But I suspect another major factor was simply that the room-based approach to map design offered a lot of leverage in controlling which parts of the story the player saw at a time. Games such as Ether that allow for very free movement through a highly connected volume have to rely on alternative methods to control narrative presentation, or else have story content that can safely be encountered in any order.

In classic parser IF design, the companion of the map was the puzzle dependency chart. Puzzle dependency charts showed which barriers had to be crossed before which others; maps represented how this manifested in physical space.

In most parser IF, not all of the map is available at once, and the player has to solve puzzles to open particular areas, whether by unlocking a door, getting past a guard, throwing light on a dark room, etc.: many of the classic IF puzzles reward the player with access to new spaces, though there are many different ways of setting up the challenge initially.

That progression of spatial access was typically what let the author control the difficulty curve (only give the players puzzles that they’ve proven they’re ready for) and the plot reveals (put the more important clues deeper in the map). Often, reaching a particular location, or reaching it under particular circumstances, or interacting with an object there, would serve to trigger dramatic scenes marking a major advancement in the story.

Then there’s the question of pacing and content density. How much story material belongs in each room? How much real space does a given room represent, and how does that connect with narrative presentation? Adam Cadre’s review of Lost New York gets into detail about some of these topics, and the problem of representational space vs. literally simulating a large area.

So with all that background explanation, here are a few other resources beside the links already given, but if anyone reading wants to recommend others, please feel free to comment as well.

Exercise. I’ve partially developed, but never actually run in its current form, a workshop meant to give students practice in designing story to line up with a map.

The idea is to give them some example maps to start with that I’d pre-selected because they provided layered access to different spaces, and then have teams design stories meant to be set in these areas. I was planning to work with maps by Dyson Logos, a D&D map level designer whom I support on Patreon (see right), but there are other possible map sources available online.

Because my workshop notes are designed for myself to run, they’re a bit more terse than they would be if I were handing this exercise off to someone else. However, the instructions as written so far are in a PDF, and here are sources for the blank Dyson Logos maps. He does loads and loads of castles, dungeons, cave systems, houses, and other setting maps designed for RPG play, which makes them a convenient jumping-off point for this exercise.

Writings and talks. Some of these are old, but off the top of my head:

Here’s Ron Gilbert’s post on puzzle dependency charts from a graphical adventure game perspective; CE Forman wrote about them for XYZZYnews with a focus on textual IF, and Gareth Rees offered an interesting reply. Andrew Plotkin built a tool to model these problems explicitly.

Steve Gaynor gave a related (though not IF-specific) talk at PRACTICE in 2011, which looked at storytelling and spatial access as well as the tools available to the player. The slides can be found here and Leigh Alexander’s reporting of the talk here. Gaynor’s talk was focused on physical puzzles, but one can also gate on methods that require the player to demonstrate knowledge of the narrative.

In Where We’re Going, We Don’t Need Roads, Juhana Leinonen talks about the relationship between story node maps and geographical node maps, and how open world is not the same thing as narratively branching.

My post-mortem for Bronze is a very detailed look at how the map, plot beats, and puzzles work together; because it assumes familiarity with the game, it may not be useful to all audiences.

Likewise, this post-mortem for Counterfeit Monkey talks about how the puzzle/plot chart connects with the regions in the map layout (see the section labeled Puzzle Discipline) and also part two of the same post-mortem under Structural Work.

Adam Cadre’s classic work Photopia tweaks the player’s (then) expectations about the map-story relationship by doing a magician’s choice move: at a particular point in the story, you reach a certain location no matter which way you choose to go. Here’s Lucian Smith on this topic (scroll down to the review of Photopia, as it’s a post on all the competition games that year).

The IF Theory Reader has some pertinent articles about how space can be used to gate a story and control access to particular beats; I wrote about modeling geography there (“Challenges of a Broad Geography…”), including a section titled “The Map and the Plot,” which talks about designing the map to cause the player to encounter particular story beats in a particular order. (It’s hard to link into the PDF, unfortunately, but the title page is pretty clear.)

David Fisher’s IF Gems selection is a list of quotes lifted from different game reviews, about what’s expected from a parser game and how to achieve that; it includes some lines relevant to this discussion.

Here I am in 2001 having a gripe about a game whose open map structure made it difficult to play, in my opinion at the time; here is Brett Witty on the same game, with some different but related observations about the challenge of the open map.

Graham Nelson’s Craft of Adventure treats the prologue, middle game, and epilogue in very spatial terms. The whole thing is still very much worth reading, decades later: some of it does seem rather out of date now, but there is a lot of core game design that I first learned from this article.

There’s a lot else to find: how the map supported (or failed to support) the story is a frequent topic in reviews of interactive fiction ca 1990-2005. If you want to do more archaeology yourself, Past RAIF Topics is a big index of major discussions from rec.arts.int-fiction during the 1990s and early 2000s, and hugely useful if one is looking for historical background on these or any other topics. IF-Review and the archives of SPAG Magazine also have a number of in-depth reviews of many, many past IF works.