Almost four years ago, Valve released Portal, a little game stuck in The Orange Box alongside much more eagerly awaited games like [...]]]> (This essay, needless to say, is going to spoil Portal 2 like the recent tornado and subsequent power outage did to the food in my refrigerator. Don’t read if you haven’t played the game.)

Almost four years ago, Valve released Portal, a little game stuck in The Orange Box alongside much more eagerly awaited games like the new episode of Half-Life 2. It became a surprise success, and I fell in love with it. Portal 2 isn’t the astounding surprise package of awesome that Portal was, but it’s still a triumph in its own right. The single-player campaign is wonderful and a joyful celebration of puzzle-solving, the co-op campaign is well-crafted and provides an experience that echoes the newness of the original game, and the whole game exhibits great game design from the sound to the visual cues to the writing. What intrigues me the most about Portal 2 is how it has the best adventure game story I’ve seen since adventure games died¹.

Way back when it first came out, I mentioned that Portal had less story than it did backstory. The game focused on establishing an atmosphere that evolved as the game progressed. Portal began with a sterile atmosphere, punctuated in part by odd, off-putting details, only to become creepier and more sinister. You moved from the cool blues of the testing chambers to the rusted red and brown of Aperture Science’s backstage areas. It worked because it was so tightly focused and so new. Trying to re-create that same feeling would have been too derivative unless many of the original elements of Portal, such as GLaDOS and the testing chambers, had been scrapped². Instead, Portal 2 took the approach of having an actual story, one in which characters evolved and changed, and in doing so used many of the techniques of 1990s-era adventure games.

Portal 2′s story, as with the game itself, is extremely linear and completely on rails. Gameplay is a series of puzzles set up as beads on a string³. Each puzzle room begins and ends with a bit of narration. The puzzles are the game’s lowest-level pacing mechanism, serving as gameplay and as gates to the next bit of story. Portal had the same structure and rhythm, but the pattern becomes much more obvious over Portal 2′s extended playtime, even though it’s occasionally interrupted by set-pieces in which you escape down catwalks.

The story isn’t expressed explicitly through gameplay, though the overarching theme of escape fits well with how you use the portals. Instead, the story is grafted on top of the game. As in many adventure games, the story exists separate from the gameplay, and you could strip out that story and be left with a serviceable physics puzzle game more akin to Cut the Rope.

You don’t even play a developed character. Chell isn’t quite the ageless, faceless, gender-neutral, culturally-ambiguious adventure person of adventure games past, but you don’t know anything about her other than she’s female, young, and a whiz at solving puzzles with portals. As her name suggests, she’s a shell, an avatar for the player. Chell never speaks, has no personality, and is functionally an amnesiac. She exhibits the same bloody-mindedness as an adventure game character. She may not take everything that’s nailed down, but she incessantly applies her one inventory item — the portal gun — to every puzzle she comes across.

Even worse, you’re not the story’s protagonist! To borrow a phrase from Christine Love, don’t take it personally, babe, it just ain’t your story. GLaDOS is the real protagonist. Chell remains the same throughout, never developing or changing beyond exhibiting an increasing skill with portals. GLaDOS is brought low, reduced in status and stature, only to discover who she really is and where she came from. She has character-defining revelations that, in the end, make her a different person, one capable of caring about Chell. True, she chose to erase the part of her that was Caroline in a manner reminiscent of how players in Paul Tevis’s RPG A Penny For My Thoughts can choose to forget their traumatic recovered memories, but in doing so GLaDOS remained true to who she is as someone distinct from Caroline.

Why, then, do I think the story is so great? For one, while you aren’t the protagonist, you have agency. You’re the person who drives the story forward. The actions you take, not only in solving puzzles but also in moving some of the game elements such as Wheatley and turrets around, are central to the story. In an unusual approach for videogames, Portal 2 explicitly acknowledges your agency. Wheatley comments on your character’s past actions in Portal on several occasions. His fear of GLaDOS cements her as a credible threat and Chell as someone who performed a mighty deed. “You know who took [GLaDOS] down in the end?” Wheatley asks. “A human! I know! I wouldn’t have believed it either.” He also acknowledges Chell’s prior actions in the big boss fight at the end. “I took the liberty of watching the tapes of you killing [GLaDOS], and I’m not going to make the same mistakes,” Wheatley tells you. It’s great to have a game recognize what your character has done and praise it in the context of the game, even if they’re the character’s actions from the previous game.

In creating the story, Valve blended techniques from static fiction and from videogames. They used a lot of foreshadowing and thematic resonance to give the story heft. The story’s silliest elements are heavily signposted through jokes that function also as foreshadowing. The announcer in the early chapter mentions that Aperture Science personality constructs remain functional in low-power environments of as few as 1.1 volts, and at one point you run across a swath of potato batteries entered as science projects during Bring Your Daughter to Work Day. Both of these are jokes, the latter referencing a joke from Portal, but then Wheatley sticks GLaDOS on a potato, which can indeed produce 1.1 volts with the right choice of electrodes. Even the bit with the moon at the end was foreshadowed both by Cave Johnson nattering on about moon rocks being pure poison but a great portal conductor and by the art in the opening room changing to include a large moon after you awaken the second time.

Portal 2 also included the Oracle Turret, a turret that plaintively says, “I’m different!” as it rides on a conveyor belt to be incinerated. If you pick it up, it babbles meaninglessly — until, later in the game, you realize that it referenced upcoming plot points as well as the game’s play on Greek mythology.

Oh, yes, the game’s symbology and themes. The game has strands of larger themes woven through it, from capture and escape to children and abandonment to the fluidity of identity. Greek mythology is referenced throughout. It’s no coincidence that the oracle turret mentions Prometheus, who brought fire to man only to be punished by Zeus by having his liver constantly pecked out by eagles. You and potato-GLaDOS are thrown down into a deep pit where the pillars are labeled “Tartaros”, wherein you find GLaDOS’s potato being picked at by a bird, which is also foreshadowed by the bird attacking Wheatley earlier in the story.

Portal 2 also relies heavily on storytelling through the environment. Not only is the setting nearly a character (a long-standing interest of mine), it provides optional backstory that players can pay attention to or ignore as they see fit. The first underground section, set in the 1950s-era Aperture Science labs, includes a lobby filled with ashtrays for all of the smokers and a trophy cabinet. The cabinet is filled with trophies proclaiming Aperture Science as the #2 applied science company, reinforcing how the company has always been behind Black Mesa. The environment echoes the story’s arc. You begin in ruined chambers that show glimpses of the sky, foreshadowing the ending. You then move to repaired chambers before being thrown into the rusting underbelly of Aperture Science, only to climb back up to the testing chambers and eventually to freedom.

The NPC characterizations are fabulous, aided and abetted by the hands-down best voice acting I’ve ever heard in a videogame. Wheatley and GLaDOS change and develop without losing the core of who they are. Throughout Wheatley remains an idiot who always takes a brute-force approach to solving problems, whether he’s a small robot eye who breaks open windows instead of picking door locks or a giant robot who fixes a test chamber without an exit by ramming another test chamber into it. GLaDOS, who changes the most, still remains a right bastard despite her Caroline-driven softening towards you.

This is also the funniest game I’ve seen since Psychonauts. Portal 2 has a blend of character-based humor (“I’m going to be honest with you now. Not fake honest like before, but real honest, like you’re incapable of”), slapstick (“SURPRISE! We’re doing it NOW!”), absurdist humor (“Then ten years in the chamber I built where all the robots scream at you”), and straight out jokes (“To help you remain tranquil in the face of almost certain death, smooth jazz will be deployed in three…”). It’s the kind of funny that makes you want to grab people and quote bits at them⁴. There’s a reason why my ringtone is the space core’s monologue and my four-year-old daughter has been marching around the house saying, “How are you holding up? Because I’m a potato.”

There have been plenty of adventure games that have the same story structure as Portal 2, used the same puzzle pacing mechanism, and had the same disconnect between gameplay and story. Nevertheless, just as Portal succeeded by combining familiar design elements with verve and skill, Portal 2′s story succeeds not because it uses new techniques but because it uses old ones in a coherent and skilful way.

[1] There have been some great reviews of the game as a whole, if you’re into that kind of thing. See, for instance, Kirk Hamilton’s review in Paste. (back)

[2] Early on, Valve did toy with a version of Portal 2 that had no portals or GLaDOS before ultimately abandoning that approach. (back)

[3] This is where I’d mimic Kirk’s domino pictures with pictures of beads on a string that I’d then spin into an extended metaphor about gameplay and love, if only I weren’t so lazy. (back)

[4] But please don’t, at least not excessively. That kind of behavior is why no one in the world is capable of watching Monty Python and the Holy Grail any more. (back)

The IF Theory Reader began its [...]]]> The IF Theory Reader is out! It’s a collection of essays about interactive fiction that run the gamut from theoretical discussions of the form to nuts-and-bolts advice for describing rooms and designing puzzles. You can download the PDF for free or buy a physical copy of the book for $15.

The IF Theory Reader began its life as the IF Theory Book some nine years ago. While the articles were written for the book back then, many have been updated and re-worked. Nick Montfort re-wrote his article “Toward a Theory of Interactive Fiction”; indeed, the first page of the article is now a footnote describing some of its history. Emily Short’s article on IF geography has been similarly overhauled, and I expanded my article on the history of short works of interactive fiction to encompass changes since its first draft in 2002.

The book begins with the classic article “Sins Against Mimesis”. When Roger Giner-Sorolla first posted it to one of the IF Usenet newsgroups, it sparked wide-ranging conversations about interactive fiction theory. Many of the articles in the book owe their existence, at least in part, to that post and the theoretical debates it touched off. If you’re interested in theory about IF and seeing how it’s been evolving since the late 1990s, The IF Theory Reader is a great compendium.

It’s also not where text-based interactive fiction parsers should be going.

Interactive fiction, at least [...]]]> Watson, the IBM program that beat two top Jeopardy! champions, is an eye-catching advance in natural language and reason-engine-style processing. It crushed its two human competitors at reading answers, teasing out the clues in those answers, and responding with the appropriate question.

It’s also not where text-based interactive fiction parsers should be going.

Interactive fiction, at least as I’m talking about it here, is a type of text-based game in which you type imperative sentences to move around in and interact with a game world. You type the sentences at a “>” prompt which in theory promises, “Type in anything and the game will understand you.”

That’s a lie, of course. What the prompt really means is, “Type a sentence that matches the pattern of commands the game understands and it might respond appropriately.” The natural language processing behind interactive fiction hasn’t changed much since Infocom set the standard back in the 1980s even though computers have become much more powerful. Could you make interactive fiction better by improving its natural language processing capabilities? Brian Moriarty, former Infocom implementor, sees NLP as a near-necessity for IF to be better. And in the wake of Watson’s victory, others have wondered why IF parsers don’t take advantage of computers’ increased processing power to do better parsing.

Would Watson, or something similar, make IF better? Watson’s requirements of 90 IBM Power 750 servers with some 2,880 processor cores and 15 terabytes of memory puts it out of the reach of the general IF audience, but you could certainly improve the natural language processing capabilities of interactive fiction without going to those lengths. Other games with text input, like Façade, respond to any typed input without requiring you to follow IF’s established imperative sentence structure. Why shouldn’t IF?

Many other genres of games, from first-person shooters to role-playing games, have a limited interface. Xbox and PS3 game controllers have eight buttons, two joysticks, and a D-pad. That limits the number of actions you can perform with one button press, and to get more you either have to make certain buttons context-sensitive, like the “use” button common to many games, or ask users to chain together a long string of button presses, like in old-style fighting games. Mainstream PC games use a mouse, arrow keys, and perhaps a set of function keys.

IF, on the other hand, has a much larger interface. Few works of IF let you type in “USE DOOR”. Instead it’s “OPEN DOOR” and “SEARCH DRESSER” and “PUT THE BOX ON THE TABLE”. That gives you a wide range of possible actions at the cost of a complex interface. To help guide players, authors adopted a standard set of commands. If your game requires new or less-usual commands, you have to spend time guiding players to learn and use those commands.

Now imagine a game accepts any input. You can type any English sentence you want and the game will attempt to parse your input. What should you type?

Game interfaces are about expressing agency in the game world. They’re how players communicate their intentions to the game and affect what’s going on inside the game. Modern videogames spend the first part of the game teaching players the available game mechanics and how to use the interface, helping them climb up the game’s learning curve. They guide players explicitly. If your game accepts any text input, then you have to work much harder to teach players what to type. To overcome option paralysis, you have to narrow those options.

Even if you had a perfect parser that could understand everything you typed, the game has to know what to do with it. Parsing is no good if you don’t do something with the results. Watson’s processing power let it parse text input and, based on that and its knowledge of how Jeopardy! answers are structured, make inferences about what related question fit the input. How much power would a game need to respond appropriately to sentences like “What have I been doing?” or “Measure out my life in coffee spoons”?

Take the case of an IF parser that accepted adverbs. Current IF parsers accept commands that are of the form VERB THE ADJECTIVE NOUN, occasionally with an added preposition and second noun: “PUT THE BOX ON THE TABLE”, “OPEN THE RED DOOR”, and similar. Now add in adverbs, so that you can “OPEN THE RED DOOR SLOWLY” or “PUT THE COFFEE CUP DOWN QUICKLY”. Now the game must decide the difference between putting something down quickly or slowly. What does it mean in game terms to TURN THE KNOB ANGRILY? You’ve added more nuance to a player’s interaction with the game world, and the IF author has to handle that nuance. It’s more work for the IF author; does it add enough to the game to be worth that work?

To rein in that increase in complexity, previous and current attempts have restricted this kind of accept-anything natural language processing to conversation with characters in the game instead of to affecting the game world as a whole. In those games, it quickly becomes apparent that the characters you’re conversing with don’t really understand what you’re saying. Worse, sometimes they mis-understand you in ways that mar your the game experience. Player agency is reduced, and you soon get the feeling that you can’t know what effect your actions will have on the world at all.

That’s why I don’t see Watson-style NLP taking IF by storm. The promise that you can type anything and the game will understand and respond appropriately has not yet been fulfilled, but even if it were, I don’t think it would make better IF.

As you might imagine, his comments have raised hackles in the adventure game community. Some of that is a reaction to [...]]]> Jonathan Blow’s new game The Witness is going to modernize adventure games. The creator of the hit indie platformer Braid claims that his new game will avoid what killed off adventure games in the 1990s.

As you might imagine, his comments have raised hackles in the adventure game community. Some of that is a reaction to a perceived outsider riding in and saying, “I know what you lot have been doing wrong all of these years!” as if he were starring in Dances With Adventure Games. My negative reaction, though, comes from Jonathan’s apparent lack of information about what’s happened to adventure games since the 1990s.

He starts out promisingly enough, talking about how video game design has gotten better as time has passed.

If you go to conferences, designers are always talking about how they’re doing things and how to make games more fun. And that’s true, it’s pretty obvious. If you go back, get an emulator and play some games from the eighties on home computers, they’re kinda unplayable. You know, people say, “Games were just as good then as they are now.” It’s just not true. Things are way better design-wise.

Where he goes off the rails is when he then turns his eye to adventure games.

[Streamlining gameplay] happened to all the genres, but it never quite happened in adventure games. The core gameplay of a racing game, for example, has been refined. It’s way more interesting than Pole Position was in the arcade, you know. Much more sophisticated. A first person shooter is a lot about knowing what’s happening on the map. Especially if it’s multiplayer, like, who is where? And all this stuff. It’s been iterated and refined.

Adventure games are still what they used to be. And what the core gameplay actually is, is very different from what the designer intends. The designer wants it to be, “It’s going to be cool puzzle solving. There’s going to be a story and stuff.” But really what’s actually going through the players head in adventure games is, “I don’t know if I should be clicking on this thing” or “I don’t know if this is a puzzle” or “I don’t know if I need an item to solve this that I don’t have yet, or if I’m just not thinking.”

Adventure games are all confusion. If it’s text, it’s “Why doesn’t the parser understand me still?” So the core gameplay of adventure games is actually fumbling through something, right? And that’s true with modern [versions]. All the episodic stuff that’s coming out. And there’s a whole community that makes modern interactive fiction games and all this stuff. And it’s true for all these games.

Gameplay in adventure games can certainly be improved, but it’s not all confusion. Adventures aren’t what they were in the 1990s. Jonathan claims passing familiarity with the modern interactive fiction community, and yet has missed how it’s been addressing this confusion. Games like Blue Lacuna and Aotearoa use keyword highlighting to make it more obvious what you can interact with. Ones like Lost Pig and Violet respond to a tremendous number of commands to make it less likely that a player will type commands that the game doesn’t understand. We’ve got better help for learning the command pattern a parser expects, Emily Short and others deconstructing the parser and whether or not it’s necessary for interactive fiction, and Aaron Reed researching how to make the parser more user-friendly.

These are not obscure, hard-to-find developments. Blue Lacuna has shown up on everything from G4 TV to Gamasutra. Violet and Lost Pig were on JayIsGames and PlayThisThing and are often cited as games newcomers should play. Aotearoa won this year’s Interactive Fiction Competition. Emily Short is one of the two best-known names in all of interactive fiction.

Meanwhile, in graphic adventure games, you’ve got Telltale Games refining what can be done with episodic graphic adventures and Dave Gilbert at Wadjet Eye Games exploring what can be done with adventures intended for casual game players.

But what gets Jonathan excited? Riffing on Myst, especially the idea of a player with amnesia.

PC Gamer: That’s presumably why Myst is an inspiration?

Jonathan Blow: It’s a classic video game trope. I mean, you start the game. You don’t exactly know who you are –

PC Gamer: Or you’ve got amnesia.

Jonathan Blow: Yeah, or you have amnesia or whatever! And then through the course of the game you find out who you are. Like, BioShock did that. Tons of games do that. This game does it but in a very self-conscious, self-referential kind of way.

So the most over-used adventure game trope, the one so prevalent that it’s the name of a 1986 text adventure written by someone who wasn’t familiar with adventure games, is what gets Jonathan excited?

When I entered physics graduate school, I had big plans. I was going to learn a little physics, but not too much, because that way I could see clearly what others had missed about physics and then perform world-changing research. Later I realized how cutely naive I’d been. Outsiders to a field can make original contributions, but more often they end up going over old ground and repeating past mistakes.

Look: you don’t have to be full to the brim with adventure game knowledge to want to design one, or to take elements of their gameplay and use them in other games. But if you’re going to claim to be fixing what’s broken with the genre, it’s best to know what the genre’s been up to since you solved Myst.

I was lucky on two counts: I’m one of the about 80 people interviewed for the documentary, and I got to see an early cut of it at PAX East this year. GET LAMP is funny, affecting, and informative, which isn’t a bad trifecta to [...]]]> GET LAMP, Jason Scott’s documentary-slash-love-letter about text adventures, is now available.

I was lucky on two counts: I’m one of the about 80 people interviewed for the documentary, and I got to see an early cut of it at PAX East this year. GET LAMP is funny, affecting, and informative, which isn’t a bad trifecta to hit. If you have any interest at all in early computer games, in how technology can shape lives in unexpected ways, check this documentary out. It’s a labor of love, and the details from the metal coins that come with every order to how the DVD itself is structured like a choose-your-own-adventure story, add richness to the subject.

None of that mattered. In the end, Digital: A Love Story told an affecting story superbly, bolstered by its evocation of a specific moment in [...]]]> Here’s the thing: I very quickly figured out what Digital: A Love Story was on about. I could see where the story was headed. I suffered through some sketchy story mechanics.

None of that mattered. In the end, Digital: A Love Story told an affecting story superbly, bolstered by its evocation of a specific moment in online history.

It’s set “five minutes into the future of 1988″, and takes place entirely in the proto-cyberspace of bulletin board systems. You’ve been given a brand new Amie (an Amiga-alike) with a modem, leading you to dial into your first local BBS. You read through the posted messages, replying to one user, Emilia, who’s posted a bit of poetry. As your relationship with Emilia deepens you find yourself hopscotching across multiple BBSes, using phone codes to steal long distance so you can call the ones that are further away.

And that’s all there is to the gameplay, really. You dial into BBSes; you read messages; you hit reply. You don’t even see what you write, only what (if any) response you get. At first I found that approach very distancing, since I didn’t know what I was saying. But as the game went on, I became more and more of a fan of this approach. It helps immersion, since you’re less likely to say, “Hey, I wouldn’t have written that!” It keeps you focused on the other characters in the story. And in one notable exchange between Emilia and me near the end, I was replying to messages as fast as she was sending them and felt like I was having a real conversation.

That helped counterbalance the other glaring weakness in the game. Since the gameplay hinges on you replying to others’ messages, there are times where the story pauses while the game waits for you to read and reply to the right message. At those points, I quickly began lawnmowering through the messages, dialing up every BBS whose number I had and hitting “reply” for every message until the story proceeded again. It’s the same problem often seen with dialog trees in games, where you select every dialog option without paying attention, pressing the conversational lever until you’re rewarded with a food pellet of story. More side-discussions would have helped, like my argument with a guy who introduces his thesis that Japan is taking over everything by saying, “Ni hao, bitches!” At times the game’s world felt empty, every message read and my replies gone unanswered. But, then, that was part of early online culture, where you might send a message to Usenet and see no replies for days, or log onto BBSes with ten users who were more interested in playing door games than chatting.

Why does the game work so well? Digital: A Love Story does two things absolutely right. One, its interaction fits the story being told, even if there are sections that you lawnmower through. The story unfolds, paced by the rate of messages and the occasional light puzzle that you have to work through. Two, it’s rooted in a very particular time and virtual place. It captures the heady days when being able to talk to people on a computer was new and amazing. It’s a tour de force that’s made more astounding by Christine Sarah Love, the author, having been born in 1989.

Digital: A Love Story is free to play, and will take about an hour of your time. It’s a neat demonstration of how digital storytelling can make stories more visceral, and it’s touching and poignant. Go give it a try.