Emily Short
Emily Short is one of the leading figures in the Interactive Fiction (IF) scene, winning awards for her work, including her debut game, Galatea. She talks about her career and her process, as well as the space IF occupies between literature and technology.
It feels like IF is having a renaissance, why do you think people are embracing text-based narrative games again?
You’ll find some people who say it’s having a renaissance and some who’ll say it’s not like it used to be and the kids need to get off the lawn. Right now, though, we’re in a place where people do a lot of reading on their mobile devices, anyway, with chat and texting and websites and other, at least partially, language-based content and there are lots of distribution channels set up to get them more of the same. That means that it’s possible to make interactive stories that resemble interaction paradigms people are used to, and sell them in stores where people are used to buying things.
There are some other contributing factors too — Twine and ink being accessible, free tools that make it easier to create work at home with little investment for instance.
What was the IF landscape like when you first started and how has it changed?
When I got involved in the late 90s, there were a couple hundred people, maximum, who were regulars on a Usenet newsgroup — a thing that predated forums and reddit and the rest of it. The people playing IF were the same people writing IF, and they were for the most part people my own age or older who had fond memories of playing text adventures in the mid-80s. No one was making any money, there was no “pro” IF scene, and there was not much communication with the mainstream video game industry.
That group was also very heavily white male, incidentally — there were some exceptions like Judith Pintar and Suzanne Britton and Irene Callaci, but when I showed up there was vocal discussion over whether I was really a woman at all, because that would be so surprising. And I got a certain amount of what I would now describe as harassment, though at the time people mostly shrugged and said “well you’re a woman on the internet, what do you expect.”
Despite its issues, though, it was also an environment of intense creative energy and mutual support, and we would have these very intense conversations about our art form and what we could do with it.
How has it changed?
Now both the exclusiveness and the intensity have shifted. Way more people write IF now, and apply a range of different standards to what they’re calling it, and I’m not the only person from back then who has made the jump to actually working in the game industry. But if you aren’t all in one place and you aren’t all trying to achieve the same goals artistically, you don’t have nearly as deep a set of conversations about how to achieve them.
To be clear, I think some of that shift has been really good, and I’ve personally benefited from it and also done what I could to make the discussions more inclusive. But I occasionally do also miss being able to have those deep-dive conversations into quite obscure aspects of interactive fiction craft.
You are one of the leading figures in IF, how did you get involved in interactive storytelling?
I played interactive fiction as a child in the 80s. My parents bought an Osborne 1, and on it they had Infocom’s Deadline, a text adventure murder mystery about a man killed in a locked room. I was six or so at the time, and I was not really equipped to understand the plot (which involved an affair and embezzlement among other things), but I played it over and over again. Never solved it, but was fascinated with it.
Even then, I really wanted to make my own text adventures. There was something intensely appealing about making a world that someone else could enter and explore; and the purposefulness of those games also appealed. It felt as though every object you encountered was there for some important purpose, which made the imaginary world legible the way the real world wasn’t.
When did you start programming?
I spent a lot of childhood and teenage time drawing maps of possible game environments, attempting to program interactive fiction in various programming languages, etc. until I got to college and one of the other students in my dorm told me about a language specifically for text adventure creation, called Inform, which I could download for free.
So I got Inform and I got the designer’s manual that explained how to use it. I taught myself to program in it and I got involved in the online community around Inform and published a few games. Then Graham Nelson, the creator of Inform, approached me for help with the new version, Inform 7, which was meant to be much easier to use for non-technical people. The result was that I got very involved helping with Inform 7’s design, documentation, and community management and that I also married Graham and moved to the UK.
You also work in AI — how do the worlds of IF and AI collide?
Both IF and AI offer strategies for telling stories in which the player has a lot of narrative agency — where the things they do are reflected in the outcome of the story, in detail. The reason this is usually hard is that it requires a terrific volume of content to achieve: if you’ve got twenty or fifty or two hundred choice points in a story, it means the possible set of outcomes is enormous. So, if we want to let the player feel like they’ve really done something interesting in the course of the story, we have to somehow make all those different configurations meaningful and rewarding.
What possibilities does this intersection offer creators?
IF is very story focused, and its textual nature means the cost of producing an individual bit of content is much lower than with other conventional methods. It also has a really strong canon and tradition when it comes to exactly these kinds of topics. In the late 90s and early 2000s we spent huge amounts of time and energy talking in detail about kinds of choice, about what they meant, about signalling consequences and stakes to the player, about how to model the significance of what someone has done.
AI tackles the problem from the other direction: if we can’t write all the variations ourselves, in all the possible combinations, can we teach a computer to do so? If the machine knows how to make a character’s dialogue a touch more or less friendly, more or less domineering, then it can reflect those subtle shifts as a result of all the small relationship shifts the player has made over the course of a game. And the result is something that feels both lively and intensely personal.
A lot of people think of AI-driven storytelling as something soulless and dull, where the machine tells a hundred million random stories and none of them are more interesting than Dick and Jane. That’s not what I’m trying to achieve. I want the AI — need the AI — to help me deliver on my own artistic vision. I might have written a very specific plot and set of characters, but still need a system to show all the many subtle variations in how that plays out in response to what a player does.
What is IF’s relationship to traditional literature? What do they share and what sets them apart?
Like traditional literature, IF can portray experiences that are novel to the reader. It can communicate setting — IF is very good at making a place you can explore. It can communicate character, and by setting out a series of choices for the player, can express the limitations that a protagonist faces (or believes she faces). Like traditional literature, IF can explore questions of morality, look into the justice or injustice of the current social system; can transport the reader to another time and place. Like traditional literature, IF is written by many people in many types of voices, though some groups of writers have been under-recognized or left in the margins.
Unlike a traditional novel, at least, IF is — or can be — quite bound to structural concerns. For some types of IF where an entire world is being modelled in code, it doesn’t make sense to have long descriptive passages or extensive flashbacks; the writing must be compact and highly disciplined about what it is conveying to the reader. Sometimes, to people unaccustomed to IF, this can read as simplicity, but simplicity is not the same thing as artlessness or a lack of craft.
Do you ever see traditional publishing embracing IF, beyond a few experiments?
Traditional publishing has experimented with IF a few times, and there was a point about five years ago where it seemed like that might be an interesting direction. In practice, though, it never did take off, and I see three major reasons. One, writing and editing IF is a very different task from writing and editing traditional fiction, and traditional publishing houses are not necessarily particularly equipped for it, either technically or conceptually.
Two, the audiences is not the same: IF is readerly from a gamer’s perspective, but many people who identify as readers do not want to interact with the story they’re reading, they just want to read it. Audience uptake hasn’t been great in the experiments I’ve seen.
And finally, there are some format challenges; early ebook formats could have developed in a way that would have better supported meaningful interaction, but they didn’t.
Do you have any writing rituals, places you like to work in or places you go to inspire you?
I sit in my living room, or occasionally in a café, and I do research on the internet. There’s nothing particularly fancy here. I could talk about inspiration I’ve taken from walking around historic cities in Europe or about turning out the end of my last big puzzle game while I was supposedly on vacation in Hawaii, but that would be kind of pose-y — the truth is I just work where I am, and sometimes that’s an interesting location and sometimes it isn’t.
What advice would you offer to someone wanting to get involved in IF?
Play a lot of work, find what you like, find the pieces that make you wish they were just a bit better or different in some way. Then make your own improvements. There are a lot of other suggestions I could make, but they aren’t one size fits all. I have gained a lot from reviewing the work of others, because I am the kind of person who learns by doing written analysis — but not everyone likes that kind of thing, or gets much from it.
