Hybrid interfaces: Texture; Contrition (Porpentine); Spondre (Jay Nabonne)

Lately we’ve been seeing more and more work that falls somewhere between parser-based IF and hypertext: in the past six weeks or so, I’ve run across two new games and a creation tool that push the boundaries in various directions.

Jim Munroe and Juhana Leinonen recently released Texture, a system designed especially to produce touch-based IF that will play well on mobile devices. Texture features the idea of applying verbs to passages of text:

When a verb is used on text, it replaces that text with something new, or else moves forward to a new page, mimicking the change-or-advance link distinctions in many Twine games. (With Those We Love Alive actually makes this distinction obvious by coloring these links different colors.)

The pairing of verbs and nouns means that navigation is a bit less obvious than in most pure hypertext Twine pieces, allowing for puzzles. The back end is still extremely simple, though, so although it might appear to be a system that would compete with the parser, in practice there’s no way (yet) to build up an extensive world model. The verbs that are available may change from page to page, and the author is handcrafting each verb-phrase interaction.

To the best of my knowledge there aren’t any released pieces yet that use Texture, but I’ll be interested to see what comes of it.

*

Contrition is a new Porpentine work inspired by Weird City Interloper. Weird City Interloper is a parser game that moves forward by keywords; Contrition is a Twine game that places a shifting palette of verbs and usable nouns on screen, as well as a list of reachable locations. The result has some of the appeal of a traditional parser world model — affordances are partially obvious and partially discoverable, and the ability to traverse the world and apply a sort of inventory makes it feel like the player has more effective agency than typical in Twine. Instead of verb-object, the player often needs to think in terms of verb-location: what will this activity do in that place? Which place/activity combinations will be interesting?

Porpentine’s places are explorable through the verbs: LISTEN is the equivalent of LOOK in a parser game, the verb that brings up the default description of a place, but with the difference that it focuses on social interactions happening there. Other, more mystical senses later come into play. Likewise, there are some objects that are essentially keys to locks in particular places. But these spaces also create a strong sense of emotional and even ritual context: to feel sorrow in one place has a different meaning than to feel sorrow in another, because the cause of that sorrow and the witnesses to it will change.

Contrition blurs the distinction between object-manipulation gameplay and internal-emotion gameplay, and that’s extremely interesting to see.

*

Spondre is a short interactive story about a man in an oppressive empire who has been imprisoned for unspecified crimes. The story itself is reminiscent of others I’ve read, but the interface gives it a new flavor. Executed in Quest, Spondre offers another new take in IF’s ongoing UI explorations: you may click on any word in the body text in order to continue the story. None of the words are explicitly highlighted as links. If you click on something that has no associations, you’ll get a short response reminiscent of parser IF, saying that there’s nothing more to learn about that object. There are also a handful of special keywords at the bottom of the screen that allow you to refocus textual attention on really important objects, such as the room that you’re in and the main non-player character.

Occasionally clicking on a topic will bring up a small menu of related actions, if there are multiple things you could hypothetically do with that object: this is reminiscent of the noun menus found in Sigmund’s Quest and a handful of other pieces, or the “point-and-click text adventure” idea raised by Dan Fabulich in our recent New Directions in IF talk.

There are some obvious strengths to this idea. For one thing, it encourages a kind of close engagement with the text to sift out which nouns (or other words, for that matter) might be important: this is the same kind of thing that makes ASK/TELL conversation appealing when it works well. At the same time, it gets rid of typing, for those who find that a cumbersome way to interact. And because the story doesn’t accumulate error messages, you wind up with a transcript that reads like a clean short story.

It retains some of the downsides of both parser and hypertext games, though. Discoverability is sometimes a challenge. Like parser games, it overpromises what the game can do relative to what it actually delivers. Where the parser says “type whatever you like”, Spondre makes the rather less grandiose offer that you can click whatever you like; all the same, the word coverage is such that many of the words you click aren’t going to do anything, especially in the later part of the game. Meanwhile, like many hypertext games and unlike the other pieces profiled here, Spondre doesn’t mostly allow for verbs: except when you’ve popped up a special menu, you’re clicking simply to activate topics, which removes the constructive agency that comes from thinking of what you want to do and then carrying it out. Maybe for this reason, Spondre is very much a reactive sort of story: you’re not so much pursuing goals of your own as you are choosing how to react to the proddings of an NPC.

Overall, I think the most serious issue here is that so many of the words do nothing. I can imagine some ways to reach full-text responsiveness that wouldn’t involve the author having to code a special response to every single word. For instance, one might use WordNet or something similar to assign emotional valences for all of the words that appear in the text, and use these to shift the narrative tone: if the player clicked on a bunch of negative words, say, the narration might become correspondingly darker and more pessimistic. (Allison Parrish’s NaNoGenMo novel I Waded In Clear Water explores using sentiment analysis of this kind, though for non-interactive output; so does the Georgia Tech paper on Scheherazade that I wrote about here.) Something like this would require a bunch of tech work as well as some UI elements to indicate what was happening with those tonal shifts, though; and it would also require training the player to understand and expect that sometimes clicking on a word just meant “I am putting more emphasis on this concept”. So while that’s one possible way of applying a UI like this to a consistent underlying model, it certainly wouldn’t be suitable for everything.

Still, a cool experiment, and definitely playable: the issues I listed are distracting but don’t at all get in the way of finishing the game. Meanwhile, the wall of undifferentiated text created a sense of mystery and potential that I don’t always get from conventional hypertext.