“SwiftKey Speaks”: Playing With Natural Language Processing Algorithms

I had hoped to have this ready for the “Dork Shorts” session at THATCamp Prime this weekend, but found myself pressed for time. So here’s a belated (and probably too long) Dork Short, in text form.

SwiftKey X is a keyboard program for Android that uses natural language processing algorithms– the sort that are used for discourse analysis, spam filters, and the like– to predict your next word before you type it. By analyzing the language from your text messages, twitter posts, email, and other sources, it becomes quite good at guessing what you might be saying. If you’re texting your friend that you need to go home to walk the dog before you meet up, it will see the string “I have to run home to walk…” and predict “the,” and if you select “the,” it will know the next likely word will be “dog.” If you’re using it for a while and give access to enough of your corpus of generated text, it might even suggest “Sparky” along with “the,” because “Sparky” is the name of your dog, and you frequently refer to him by name.

Swiftkey works pretty seamlessly, and I find that it makes for a much more pleasant and efficient keyboard program than the standard Android keyboard, or even my old replacement, Swype.

However, as well as SwiftKey’s predictive algorithms work, you really need to have a person at the reigns. Letting the algorithm drive and not steering the content can lead to results that range from comical to nonsensical to the almost poetic.

I thought I was the only one who got a kick out of letting the algorithm auto-generate a sentence without context until I saw this comic in xkcd last week:

The SwiftKey team, recognizing that even being jibed a bit by a site as popular as xkcd is good press, smartly went with this, and asked their blog’s readers to post the sentences that were produced by hitting their own “central prediction key.” (SwiftKey generally presents three predictive guesses for each next word or punctuation mark. The center key is the most likely.)

The responses were amusing: “I am not allowed to tell anyone so keep that in the US,” and “I am a giant popsicle of sleep.” They also, however, highlighted the powerful personalization that comes with using the algorithm with a corpus culled from your own writing: no two seem to be alike.

Personally, I get “I am on my way sorry lost track of time, meant to be a little late.” It’s a powerful comment on my punctuality (and on the unpredictability of public transit in the Metro DC area.) My SwiftKey corpus also seems to be hungry– a very popular sentence in my SwiftKey autotexts is “I am grabbing a sandwich.”

Discovering the sentence generated with the central prediction key is a fun exercise, a nice parlor trick or conversation starter. But it doesn’t really lend itself to repeated iterations. For this reason, it’s probably inevitable that I would create something for myself like this project:

SwiftKey Speaks” is a site where I record the outcomes of a predictive-text game I have created for myself on SwiftKey. The game itself is simple: I start with a single, randomly-selected word. And then I write, each word selected from one of the three suggested next words, until I feel I have reached an end or SwiftKey gives me no choice but to end the paragraph. Sometimes I reach a dead-end, and the prose degrades to the point where it’s no longer comprehensible. Then I go back a few words and start in a different direction.

I have long been interested in the poetics of spambots, at the ways that computer-generated language can so closely resemble real communication, but still lingers in this linguistic uncanny valley. Add that to a love of language games and found poetry– and a healthy dose of technophilia– and you can see why a game like this is so much fun for me.

Likewise, I have both a predisposition toward and an intellectual interest in what Mary Flanagan has called “subversive play,” that is, play that works contra the game’s design. To put it another way, it’s hacking a game from it’s insides. It’s the ludic pleasure derived from discovering that in certain games, the Master’s tools can dismantle the Master’s house. As long as you remember that you can repurpose those tools, and that a hammer can also be a crowbar.

As a game, “SwiftKey Speaks” is deeply subversive of the technology it uses. SwiftKey X was designed to assist in writing what you think. “SwiftKey Speaks”  a complete inversion–you are not interested in communicating anything, but simply making novel sentences. It’s taking a utilitarian tool and making it an instrument of whimsy.

Moreover, it subverts the typical game dynamic. Most games set the parameters and rules, and the player has to work within these. Playing “SwiftKey Speaks,” the user is constrained by the parameters of the software, but pushes back, setting parameters for the software. The game play feels more like an interplay, a dynamic relationship, where both you and the software are changing the rules on the fly, interdependent.

It starts to feel like a cyborg equivalent of Facilitated Communication— by selecting the next word from the field of three, you’re trying to help out a buried intelligence that doesn’t fully grasp the language it wants to use. Because it uses a corpus of your own words to feed its algorithms, SwiftKey starts to feel almost human, in a strange way. You and the bot in your phone share interests, passions, tics of language, and an enjoyment of a good sandwich.

The experience of playing with it almost starts to remind me of Ted Infinity and Nabil Hijazi’s excellent short story The Peacock, a love story about a man and a sentient spambot.

…At any rate, it’s a fun little game that I can play on the Tumblr app on my phone while waiting in line or riding the subway. I’m not sure how long it will keep being fun, but I intend to keep playing, and recording the outcomes, for as long as it is. And I’m curious to see if others find the project interesting, or want to play the game themselves.


Interactivity and Digital Storytelling…

NB– this is primarily cannibalized from a post I made on the class blog of the Digital Storytelling class I’m currently taking. I like to keep my stuff all in one place, though.

I have to admit that, while many of the examples of “digital storytelling projects” that my classmates posted to the class blog were quite interesting and well-done, and some where quite thought-provoking and evocative, they felt a little… old media?

Basically, they were just low-budget, one-person documentary shorts.

One of the really fascinating things about new media technology, however, is the interactivity of it. If you use the internet in 2010, you are almost certainly not just a media consumer. You are a producer. The most successful sites on the internet– from Youtube to Facebook to Twitter to Google itself– are not content creators. They are frameworks that host user-generated content, sort it, make it manageable, encourage discovery. From the moment Tim Berners-Lee began to conceptualize the World Wide Web as something interlaced, hypertextual, navigated by users, the web has challenged models of passive viewership. The web is interactive. New media is interactive.

So where’s the interactivity in digital storytelling? Well, it seems to be coming. Though it is still pretty primitive in its application.

A sidebar of sorts:

Is it still storytelling if it’s interactive? If the author relinquishes some degree of control to the audience, is it still his or her story?

I would argue that it absolutely is. While a was a voracious and omnivorous reader as a child, one of my sisters’ and my favorite series of books was Bantam’s Choose Your Own Adventure series. Essentially bound hypertext, the book would take a forking narrative format, where the reader was, at key moments, presented with choices. The reader’s choices determined the outcome, but the author’s vision remained at the center. Forked stories could fork back into themselves at time– especially in a time-travel story.

While most video games are admittedly thin on narrative, some of the best follow a similar course– allowing player decisions to influence the chain of events within several forked narrative outcomes.

That digression over, I have to say, I haven’t found exactly what I was looking for. I haven’t found any single example that illustrates well how exciting this possibility is. But let me run through a couple examples– all imperfect in some way– that illustrate what kind of thinking I’m talking about. All of these take advantage of Youtube’s fairly recent annotation feature.

“B-Boy Joker” is very well-implemented, though it’s more of a game than a story. Even by game standards, there’s not much narrative: The Joker and Batman are having a dance battle. You have to match your opponent’s moves or he will defeat you. Not really a story at all. But the action is compelling, the use of annotations is highly effective, and the stop-motion animation is top-notch. One could imagine making a project that was more narrative along similar lines.

logoSimilarly, “Youtube’s first weekly game show” Truth or Fail, is pretty lacking as a narrative, being more of a game. But while B-Boy Joker was more like a video game, Truth or Fail resembles a (highly eccentric) quiz show. Nevertheless, there is a beginning, middle, and end, and since many of us if not all of us are interested in the informative and pedagogical uses of digital storytelling, I thought it bore mentioning because it’s pretty easy to see how such a framework could be used educationally.

Finally, I found two more traditionally narrative interactive videos that unfortunately seem to be experiencing technical difficulties. Annnotations on some of the videos in these series seem to be broken, so clicking on the screen doesn’t always work. But go and check out The Time Machine: An Interactive Adventure and Choose Your Path: Find Sparta! and try to imagine them actually working.

At any rate, it seems obvious to me that interactivity is a pretty exciting possibility in digital storytelling. And that, unfortunately, we might not be quite there yet.


Myst is not a good video game.

First off, I had to post this when I found it: apparently, Nial Ferguson was so impressed by the Calm and the Storm that he went and got a job with them. It’s not really a conflict of interests or anything, but I found it very interesting.


That said, I’m still not loving Myst. I’m not exactly an avid gamer, but I do enjoy some games in moderation. I’m not one of those anti-video game people. But I do not find Myst particularly enjoyable.

I’m still trying to beat it, though, as Prof. P promises that there’s payoff at the end… I doubt it, though.

While reading the Gee article, it occurred to me that Gee has this almost platonic ideal of the “good video game,” which he then contrasts with a bad classroom. But his thinking about what good video games are made me realize that Myst falls short.

Yeah, I said it: Myst is not a good video game.

I’m not going to belabor the point, but a couple of principles of “good video games” that Myst V (the version I’m playing) violates:

  • Verbal information is seldom given “just in time,” but rather well in advance and often in a different location.
  • While there is a fish tank/sandbox in the form of the first couple rooms, it’s essentially useless, except for understanding the very basics of navigation.
  • This may just be me, but I know it’s not me alone: the game isn’t “pleasantly frustrating.” It’s just FRUSTRATING. The puzzles are very difficult, which may just be my unfamiliarity with the series, but they’re also often quite hard to FIND. This leads to situations like wondering around for a long time trying to find something to DO (this game is so action-less I find it PAINFUL. I can accept a nonviolent video game, but a game where you can’t even JUMP, get your feet wet in the ocean, or TALK TO ANYONE becomes pretty stagnant pretty quick), or, conversely, completing a puzzle without being aware that a puzzle had been there in the first place. (This latter situation occurred to me more than once.)
  • The game doesn’t go through “cycles of expertise,” it just stays at a fairly stable level of difficulty, with the occasional easier problems sprinkled throughout. When I had to resort to a walk-through at one point, I realized that I was doing the game essentially backwards. And the thing of it was, some of the earlier stuff was far more challenging than some of the later.

Oh– and just another desultory thought, here… Where do walk-throughs stand in terms of academic honesty and honor codes? Is it more like plagiarism, or more like a study guide? 


Here and here.


ARGs and the Classroom

I attended the annual conference of the Popular Culture Association and American Culture Association last week. It was fortuitous, maybe, to have this week’s James Paul Gee reading on the potential of video games as pedagogical tools, as I had the opportunity to attend several Internet and Video Games panels. One panel in particular made me reflect back to this course, and instead of just doing a gloss of the readings and the websites I’ve visited, I want to use my post to discuss the ideas I encountered in this panel.

J. James Bono, from the University of Pittsburgh, presented a paper called “Playing with Disaster:  Serious Games, Alternate Realities,and Atlantic Storm.” This paper brought up the pedagogical possibilities of something I’d never heard of– Alternative Reality Games. These are a new development, a web-based type of game that is without a single platform– the game is outside, it’s in the minds of the participants, it’s essentially research-as-gaming. Players find clues and put together remarkably difficult puzzles cooperatively, in a “game” the elements of which could be anywhere– on any website, in the form of an SMS text message, even in that dreaded IRL world. For those of you unfamiliar, as I was, with the idea of Alternative Reality Games, or ARGs,as I was, I encourage you to check out the Wikipedia article linked above– it gives a good sense of what ARGs are, and how they work, and it’s pretty well-written for a Wikipedia article.

Another presenter, Angela Colvert, of the University of London, discussed a project she undertook with two primary school classes she taught: she assigned her fifth grade students to create an ARG, specifically targeted at the fourth grade students she also taught.  While the project was, due to the students’ ages, a rather simplistic project about an alligator who lives in the London Sewers, the project immediately suggested a whole set of ideas in my mind– what if an assignment for grad students in CLIO was to design an ARG for students in an undergrad course, one based on an actual historical event or mystery? One class would acquire an invaluable set of skills based in information design, and the other could finding new approaches to research– in an environment of a “game,” which whether we’re gamers or not, is often more fun and engrossing than reading a textbook and memorizing dates.

The final paper in the panel that related to this class– I’m excluding a wonderful piece about the Japanese aesthetic principle of mono no aware in the Nintendo video game Pikmin 2, because it simply doesn’t apply– was by Terence Brunk of Columbia College.  While his paper was actually an analysis of the narratological principles that can be seen in two “serious” online games– the type of game that is created specifically with the social consciousnessof its player in mind.

This paper really brought home the potential of ARGs as opposed to more traditional video games– no matter how many options you present a player, video games are essentially goal-oriented and thus fairly linear. Eventually in the process of game design, you have to decide that the player must complete Level 1 before entering Level 2. While they’re interactive, video games still have much the same linearity of text. And this is reinforced by their very nature: they’re pre-produced, complete worlds. Add-ons like they have for the Sims or when they add new areas to an MMORPG are limited fixes, and must follow the rules previously established.

The role of the “puppet master,” the person who essentially creates and maintains the ARG, often modifying the next step, puzzle, clue, or plant based on previous outcomes, is in many ways essentially very similar to the role of an excellent educator– they challenge their subjects, altering results to outcomes, constantly pushing the problem further. I think it could be a really useful tool for this reason.