Aaron A. Reed has been attempting to be innovative with his interactive fiction for more than a decade, with occasional successes: his IF game Blue Lacuna has been widely admired by the community, and his IF-like-things 18 Cadence and Prom Week have been nominated for awards at IndieCade and IGF. He is the current organizer of the annual Spring Thing Festival of Interactive Fiction. His latest game The Ice-Bound Concordance merges explorable text, a complex NPC, and a printed art book driven by augmented reality.
AlethiCorp (Simon Christiansen)
IF authors might carefully consider which text adventure platform they should learn, but AlethiCorp is a smart reminder that we’re drowning in platforms for interactive text. We type text into calendar apps, search engines, online storefronts, social media apps and surveys; new text is returned to us after our input is digested by servers and shaped by algorithms and database queries. The fact that more of these platforms aren’t used to tell stories is due more to a lack of imagination than any technical obstacles.
There’s certainly an extant if small tradition of telling stories through mock-ups of real-world interfaces (such as many of Christine Love’s games from Digital: A Love Story on) but most of these titles have only a surface recreation of a recognizable interface. One of AlethiCorp‘s charms is that it’s actually implemented like the real deal, a genuine server-side C# based platform with side bars and top bars and real user accounts and passwords and ExternalLoginConfirmationViewModels (the source code is available). Many of the strengths of the experience come from embracing the realities of such a platform: the way it both constrains and enables our ability to communicate (and thus to tell stories).
Take the NPCs. They send you messages via the internal corporate email system, but your boss warns you to “keep in mind that most employees are way too busy to spend all day answering mails.” Here’s a new, disturbingly plausible excuse for underimplemented NPCs. (From the point of view of our employers, are we all underimplemented?) Obviously shopkeepers are too busy to talk to you about anything but their goods, and obviously the employees of AlethiCorp are too busy to answer emails. (To be fair, you can actually get replies out of emailing people, which seems a delightful surprise after such appropriate expectation-setting.) This a truth of the platform: plenty of open text fields to type into, most of which go straight into a black hole, carefully stored in a database table but never to be retrieved again.
The puzzles further develop this synergy with the platform. One relies on remembering a warning that your browser is not up to date, as easy to overlook as a casual clue in a room description. Another relies on exploiting a bug in the system to do with having two tabs open at once, which summons a cold sweat for anyone who’s been unsure what horrors might result from accidentally clicking the Back button three pages into an .aspx expedition. Form fields with autocomplete values feature prominently. Some of the failings of the puzzles come from similar failings in web interfaces: I missed out on a big swath of possible endings, for instance, because the game failed to tell me the proper format for a certain input field. To successfully mark someone as a threat, you must give them a rating of 6 or higher in an unquantified numeric field, which the author as acknowledges as a weak design move, albeit perhaps an in-character one: “What do you mean no one told you about threat ratings? Everyone knows about threat ratings!” (Quote from the pay-what-you-want walkthrough.)
The characters in AlethiCorp are all exceedingly earnest: the writer who won’t accept that his prose is terrible; the activists and corporate managers alike quoting philosophers at every opportunity; the spy who begins expense-reporting an increasingly elaborate romance with his target. In the end this makes them seem like they’re not quite taking this plot seriously, and neither should we. On the other hand, maybe this is what we’re all doing in the modern corporate or governmental surveillance state. Characters in a world like this suggest different motivations, imply different priorities. If all that matters is the numbers in your employers databases, or correctly posting to your social networks in the manner your social networks expect you to, you’re doing it right, aren’t you? Of course you put eight hours in the game’s timesheet at the end of each day. Why wouldn’t you?
With Those We Love Alive (Porpentine, Brenda Neotenomie)
Once again I have the unenviable task this year of explaining why Porpentine is innovative: unenviable not because the work isn’t worth writing about, but because playing it for yourself speaks far more adequately and elegantly to this point than I can. When the game asks you to draw a sigil of pain on yourself, it gives you this as context: “What is there to say? If you know, you know. If you don’t, nothing can show you.”
Explaining is inherently an externally directed action. This is a game about writing on yourself. Even though lots of people have shared their symbols online, it was a very internal experience for me. So I’m writing about how I experienced it, alone in my bedroom, trying to make feelings into shapes.
Innovation is “to make changes in something established.” Clearly drawing on your own skin is not an established game mechanic, but deeper than this are the implicit challenges in the act. Can a computer game smell like something? Sure: Sharpie and sweat. Can a digital game physically change you, leave tiny traces the next day, even if you scrub? Apparently yes. Deeper: can the act of playing a game have consequences? Can you express pain with an abstraction? Can you save a life by drawing a symbol? Oh yeah, hey, we call that one “writing.” It’s easy to forget we have that power.
I like the choices in Porpentine’s games. They’re playful, in a dangerous way. Take meditating at the lake. (Can a game make you stop breathing? No, but it can ask nicely.) After a longish pause, you’re given the option of a longer or shorter one next time. Is this to make the game more accessible for a wide range of lung capacities? Or is it a challenge?
As an artificer for the Empress, you have many choices constrained within spaces of horror. Do you try to make something beautiful anyway, or embrace the ugliness? You don’t have the courage to break out of her boundaries (no one does, until the end). This struck a chord, as a game designer who’s also been “noticed for my talents, the work I’ve displayed at festivals.” Standing on show floors trying to get my game noticed by people I’m not even sure I want to notice me, wondering if I would even want their validation or success, I’ve felt this: “You imagine yourself making more and more things for the empress of increasing intimacy, until you are making her bones themselves, and then the individual components of her soul. Until she is finally replaced by that which came from your hand alone.”
These choices are not about outcomes, but the choices themselves: their presence or absence, their limitations. Choices as traps, as offerings, as things to read into. Choices you can live with, or can’t.
I like Porpentine’s writing, too. Like many of her games, this one abounds with tiny koans to descriptive prose, little arguments against the bloated paragraphs that much IF (including mine) has been guilty of. “Oily stains unreal the wood.” A sky full of “disemboweled rainbows.” “A hotel as far as distance.” This extends to characters, too: I developed a wistful attachment to the slime kid, though she’s compressed into just a few handfuls of words. Every day I looked for her.
The drawing I spent longest on was the one to “influence the outcome” of your suicidal friend. There are hypothetical games with augmented reality or Kinect tracking or Leap Motion sensors that could detect whether and what you’d drawn on yourself, and respond. Those games would somehow judge your drawing on some precision criteria: is that close enough to an “e”, or to red? Is this enough like a circle? Is too much of your skin marked, or not enough? This is not a game like that. “Please remember: nothing you can do is wrong.” And yet doing the right thing was hard precisely because what I needed to satisfy was my own sense of rightness. What could I inscribe to keep from losing someone I loved? In the note you write just before this moment, many of the options are focused on you, not the person you’re trying to save: “I can’t live without you.” “I’m begging you not to do this.” Am I doing this for myself, or her? Am I doing this for some past self, some past someone I was trying to save?
It’s impossible to think that choosing the right symbol to draw could save someone who exists only as a few words on your computer screen, and yet this happens thousands of times every day.
Every day is damage.
You can wash your sigils away, but you’ll remember how they looked on your skin. A game can change you. We admit certain things are impossible, and we do them anyway. These are innovations, at least to me.
An Earth Turning Slowly (Mæja Stefánsson)
The rise of choice-based IF over the past several years has sometimes played out like a literary reformation: a simplification of spelling, say, or a modern update to Shakespeare. Proponents tout a new clarity and broader accessibility, while the old guard grumbles about tradition, dumbing things down, or losing some hard-to-articulate quality that made the whole endeavor worthwhile in the first place.
An Earth Turning Slowly is one of several experiments in the last few years that tries to capture some of the best bits of both parser and choice-based IF. A Colder Light is maybe the best-known example, trying to capture the putting-things-together-ness of parser puzzle games with the convenience of a click-based interface. Other games have experimented with incorporating the pleasures of Typing Stuff In with an otherwise hyperlinky format, including Ex Nihilo, Barbetween, or (in the broader world of indie games) Elegy for a Dead World and more recently Her Story. Here, the system makes a genuine claim to be a real hybrid of two distinct forms of IF, rather than bolting a single feature of one onto another, so it’s worth drilling into details about if and how this works.
The story is relatively brief (although lengthy for the minicomp event it was created for), but long enough to showcase the interface and get a good feel for it. There are four primary innovations by my reckoning, each of which can be discussed on its own.
- Verbs are explicitly introduced, but must be remembered after the chapter that first uses them. This elegantly sidesteps the guess-the-verb problem while still requiring some amount of focus and attention: while the game is puzzle-light overall, the mechanic creates something of a “meta-puzzle” in remembering the commands you’ve learned about and thinking which might be applicable in your current circumstances. I’ve long thought that explicitly introducing verbs is a really powerful idea (I used something like this in maybe make some change and have always wanted to flesh the concept out further), so I’m really happy to see this here. One downside is that several verbs show up in one scene only and aren’t useful again: in a longer game, it might be helpful to distinguish scene-specific and multi-purpose verbs on the UI level.
- Auto-complete options for currently valid commands appear while typing, and 2b. Invalid commands cannot be submitted. This removes another major stumbling block with parser games (the frustration of misunderstood inputs and the large, tricky authoring space of possible responses to them). A lot of why this works comes down to tuning: the auto-complete doesn’t kick in until after the second character, reducing the possibility of it spoiling the meta-puzzle of remembering verbs, and appropriate nouns (if any) don’t appear until you confirm a verb, preventing a too-easy reveal of the whole possibility space (which would effectively reduce the interface back to a multiple choice selection).
- Text explaining or justifying each command is shown before you commit to it. This seemingly-optional component might be the key to this whole interface paradigm, because it’s hard to imagine how this could elegantly be incorporated into either a pure parser or pure choice interface. It allows the game to both clarify what a given command means in a certain context, but also gives more information about what the player character is thinking and feeling about the situation, and what they think the consequences will be of certain actions. It makes the act of considering options a part of the gameplay itself: the opening scene, for instance, becomes about considering the social repercussions of subtle actions like talking to one person before another. This is a really lovely innovation.
- Context and timed nudges. Your action is contextualized above the prompt by a list of “immediate goals” at a higher level than individual verbs, which helps you remember what you’re doing. The most recently introduced verbs are shown next. If you run low on options, another window reminds you of older verbs that might be useful. Finally, the auto-complete tunes itself as time passes to first activate after typing a single character, instead of two, and then to just show a list of possible verbs, if you wait too long. This nudging could be annoying in a more puzzle-focused game, but it works quite well here. There’s something compelling about the idea of an interactive story that lets you choose how much you want to drive it: Versu games like the (now dearly departed) Blood & Laurels enabled this by allowing the player to sit back at many points to watch scenes play out without intervention, which worked well if you weren’t feeling particularly inspired or just wanted to see what the other characters would get up to without your prodding. Here, there’s an implicit sense that no puzzles should take more than thirty seconds or so to figure out, which might not work for everyone but for certain types of games allows more people to see a complete story rather than get stuck and give up.
How do these all come together? Quite delightfully, honestly. It doesn’t hurt that the UI as a whole is well-polished, easy to use and to understand, building on the already-solid presentation of Undum and extending it seamlessly. I can’t help but wonder if some of the parser slump is from a lack of interpreters that integrate so many of the convenience features we’re almost taking for granted here: autocomplete, or smooth scrolling with a marker for your eye to know where to jump to.
I’d be interested to see this approach exercised further in a game where your choices have more consequence. One of the presumed strengths of a parser-based system that a hybrid like this attempts to capture is the feeling of discovering a solution or avenue of exploration on your own. An Earth Turning Slowly doesn’t especially have puzzles: the closest it comes is the chapter where the viewpoint character tries to figure out a way to track down the stolen data pad, but it felt less like something I was expected to solve and more like being gently funneled down a lazy river towards the proper solution. Part of this is no doubt due to the compressed time frame of development, which gave the author little time for alternate solutions or different outcomes, but the lack of consequences for choices blunts the impact of the system. The opening scene, for instance, frames itself as a conflict between managing your professional duties and your personal interests, but replaying several times reveals you can’t actually make a choice about this: the same thing happens no matter what. There are a lot of options available, crucial to make the UI feel like an improvement over a straightforward list, but many of these options are synonyms for each other, which lessens the impact especially on replay.
But there’s lots to like here. I’d love to see the author’s Undum mods made public and more games developed in this format. It has a unique charm that I think both reformers and hardliners can get behind.
Hadean Lands (Andrew Plotkin)
The bit of innovation I’m fondest of in Hadean Lands is giving me a reason once again to fill up a notebook with maps, observations, and hopelessly backward theories.
It might be fair to ask whether this is innovation or nostalgia, but “paying attention” has become such a taboo of a game mechanic, it’s fair to say that making it fun again should count. In fact, this is typical of much of the innovation in Plotkin’s game: the surprise is not so much bringing something new to the table as in dusting off and perfecting something that’s been on the table so long, we’ve all assumed it’s just an unpleasant part of the scenery.
Others have written about the core structure in Hadean Lands: mastering a series of rituals to slowly build up a higher-level syntax for dealing with the world. Each mastered alchemical ritual begins as series of painstaking commands, consolidating first into a single typed action and later to a barely-noticed blip in a river of auto-completed steps flowing towards some larger goal. This approach lets players spend nearly all their time on the front lines, the sweet spot of puzzle games: thinking about the problem. Not struggling with the parser, not managing your save games, not remembering your way across the map or executing ideas, but sitting and thinking about them. Most of my hours of playtime were spent flipping between pages in my notebook, looking for connections between seemingly disparate elements. When I went back to the keyboard it was because I had an experiment to try, or needed to collect further data. This was nice. If the medium is the message, Hadean Lands turns IF’s message into one of distillation through iteration: figuring out why and how your view of a model world is wrong or incomplete in order to build the tools you need to master it. (This is interesting to contrast with Braid, where the iteration is more about perfecting your performance.) Mechanizing the repetition has been tried in IF before– there’s a whole IFDB category for “games revolving around a groundhog-day loop” — but never so systematically or wholeheartedly.
Part of this distillation is a love for the “good old days” of IF– a sprawling map, interesting nouns and obscure verbs to abuse them with, tantalizingly locked containers– but tempered with any number of modern conveniences, of which the auto-ritual syntax is merely the most visible: auto-moving, auto-remembering where you left things, auto-cataloging of nearly all the catalogables, the wide range of responses to not-quite-right commands and ideas. Several little-used or famously difficult parser mainstays are here, from setting things on fire to swimming through tunnels to sampling edibles and drinkables to usually-disappointing commands like LOOK THROUGH that here have glorious results. Marshalling of obscure knowledge comes into play, but all the knowledge can be discovered in-game rather than relying on your memory of Greek mythology, or not being color blind. Even the sequence of ridiculous actions which many adventure game puzzles boil down to becomes more plausible (and even forgivable) when framed as obscure alchemical ritual.
In a recent Jimmy Maher article about early computer games, he mentioned the baffling truth that many commercial adventures of the early ’80s (like Space Quest) weren’t even tested before release. Hadean Lands arrives at a future those games could only dream of: one with a puzzle structure tested by people, of course, but also by rigorous design methodologies and even computationally assisted verification. Not to put the game on too much of a pedestal, but I’m thinking of the rise of linear perspective and the mathematics of art during the Renaissance: saying “Hey, we’ve been doing this painting thing for a while now: maybe we should figure out how to do it right?”
For certain values of “right,” of course. The distillation of the puzzle game into an alchemically perfect form comes at the expense of some of IF’s other potential pleasures, like character interaction, a dynamic plot, or poking at irrelevant scenery just for the hell of it. There are plenty of interesting descriptions, but the environment is carefully constructed to be a transparent container for the puzzles: like a crystal goblet it is bereft of ornamentation that does not serve what it contains. One occasionally misses a world with things in them that are there just to make it a more grounded place, or to be funny, or to help tell the story. Not that the game needs to be more complex: but by choosing to be all engine and no car, by necessity it only attracts a certain kind of rider.
But on technical and design fronts, Hadean Lands might be that kind of innovation that is such a leap forward it’s hard to duplicate: I’m thinking of other works of groundbreaking interactive story, like Facade or Blood & Laurels, that seem so impressive it’s hard to know how to begin rising to the new bar. (Authors of such works have to solve the equally thorny and thankless problems of making your tools available and convincing people to use them; it’s rare there’s resources and energy to do even one, let alone both.) But in at least one way the game’s innovation has been catching. When the Kickstarter launched in 2010, it seemed a risky pipe dream; when it quadrupled its goal, either a happy fluke or a scarcely believable sign of things to come. Five years later, when that game (a hardcore, unapologetically text-only puzzlefest) has hit major gaming platforms and been covered by mainstream sites like Slate, when IFers like Emily Short have been written about in the New York Times and a text game like 80 Days has been named TIME’s Game of the Year, and dozens more people have crowdfunded or Greenlit their own text-based games, Plotkin’s modest Kickstarter may be noteworthy as much for its design as for the innovative idea that people could be ready again to pay for interactive fiction.