Casey Muratori's Treatise on Interactive Fiction

Casey Muratori wrote a series of blog posts on future of Interactive Fiction about a decade ago, and they are some of the most thoughtful views I have read about this little hobby of ours. While the blog has dissapeared down the entropy machine of time, I was able to recover its content using the Wayback Machine. I am posting it here both for archiving and for discussion (since it seems that these never got posted here). I have not been able to contact him, hopefully he wouldn’t mind me posting this here.


The Cave Inside the Terminal

I’d like to take you back for a moment, all the way back to the dawn of human history, when mankind subsisted on gathered fruit and felled mastodon: the late 1970s.

You are in a secret canyon which exits to the north and east.

The plains of rural Massachusetts stretch out to the horizon. The baby Jesus, still clad in Frank-incensed diapers, rides majestically by on a newly-domesticated velociraptor. And there, in the ambiguous space between the dining room and the living room of a suburban home, a tiny three-year-old Casey sits on his father’s lap facing the great hulking frame of a computer terminal.

It is a formidable machine, part cathode ray tube, part tangle of circuit boards, all encased in thick plastic moulding so inoffensively beige that it makes the walls of doctor’s offices seem garish by comparison. Incapable of performing any substantial computations on its own, it is wired to a modem, which is wired to the telephone jack, and so connected it patiently waits to receive single characters from the real brains of the operation: a mainframe at the headquarters of the Digital Equipment Corporation.

The transmission speeds are so slow you can physically see the arrival of each letter on the flickering monochrome screen, and as the letters form words and the words form sentences, Casey’s father reads them aloud so he can understand the predicament he’s gotten himself into…

A huge green fierce dragon bars the way! The dragon is sprawled out on the Persian rug!

OK, essentially what just happened — and I know it will be difficult for you to fully comprehend this right away because your brain will take some time to fully extricate itself from such a lavishly rendered vision — is that you were transported to another place and time by the gentle, rhythmic, linguistic prodding of my carefully-chosen words. These words made their way from the screen through your eye sockets onto your retina and into your optic nerve, where they travelled to your brain and caused electrical signals to do something with like, neurons or something, and then some stuff happened.

Point being, scientists have a technical term for this. They call it “imagination”. You imagined you were back in the Cretaceous period with baby Casey and the terminal, and through the awesome power of language, you felt as if you were actually there. And that is precisely what the interactive fiction genre was all about.

What’s “interactive fiction”? Well I’m so glad you asked. Nobody asks any more. You have excellent manners and frankly it’s refreshing in this day and age when nobody asks. You’re an active listener, and that’s really rare. Thank you. Thank you for being you.

The answer is that “interactive fiction” was the romantic term applied to games perhaps more accurately described as “text adventures”. These were games where the computer printed out some text describing a situation, and the user typed in some text to say what they wanted to do. Originally, this was probably no one’s grand vision of the future of interactive entertainment, but rather just the most natural thing you could make on the computers of the time which often couldn’t do anything except print out text. In fact, on computers that could display graphics, the first games were often graphical, like the seminal Space War and later Pong. But many computers — especially terminals — had no graphics capabilities whatsoever, so text was the only medium for creating an interactive entertainment experience.

In the flashback that opened this blog post, little toddler me was playing the very first computer game I ever played, which coincidentally was the very first interactive fiction title ever created. It was a game known nowadays as “The Colossal Cave Adventure”, even though I think it was called simply “Adventure” at the time. It was one of the earliest computer games (although by far not the earliest), and it invited players to use simple imperative text commands (“get lamp”, “enter”, “climb”, etc.) to navigate a series of rooms in an underground cave.

In these rooms were all manner of fantastic items and creatures. The player would take the items and use them to overcome “puzzles” placed in their path while collecting as many “treasure” items as they could, their score measured by how many treasures they had thus far collected. In one room, as the computer described it, there was a “fierce green dragon” — an obstacle — that sat atop “the Persian rug” — a treasure. What text you had to type to get past the dragon, and get the treasured rug, was something the player had to figure out by logic, trial and error, or most likely both.

Much like Wolfenstein 3D ushered in the era of the first person shooter, Colossal Cave Adventure marked the beginning of a multi-year heyday for the interactive fiction genre, one that many saw titles sell in the hundreds of thousands of copies, a dramatic figure for computer software in that day and age.

But alas, unlike the FPS, the golden age of interactive fiction would prove to be rather short-lived.

You have died.

In titles from one of the most prolific interactive fiction companies, Infocom, when you made a particularly-wrong move, the game ground to a halt and you were unceremoniously presented with the phrase, “You have died.”

Well, much like in the games themselves, in the mid eighties, the reign of interactive fiction as a prominent genre came to an abrupt halt. The concept that someone would pay $30 for a game featuring only text went from being commonplace to laughable, and the genre itself went from being the primary form of computer entertainment to being an extremely small niche where non-commercial titles were made by diehard fans for other diehard fans.

Interactive fiction, the genre, had died — if not entirely, then at least commercially.

Now ideally, this would not be cause to mourn. Ideally, the reason interactive fiction died would have been because newer, better ways of making games came about, and these new ways did everything interactive fiction did, but better. That would have been nice. And for a certain read of the history, one could convince oneself that’s exactly what happened. But for some of us — and perhaps particularly for those of us whose first experience with computer entertainment was hearing their parent’s voice read glowing green letters describing a magical world full of magical possibility — that’s not at all what happened.

Which brings us to the topic of our new series of Monday posts: Molly Rocket’s soon-to-be-officially-announced interactive fiction project. The aim of this project is, over the course of several games, to bring back all of the things we as players lost when interactive fiction died. And not just bring them back as they were, but bring them back anew, with the full force of today’s technology behind them.

What are those things, you ask? What did we lose when interactive fiction died? Well that is a very good question indeed. So good, in fact, that it will be the entire topic of the next Molly Rocket blog post.

The Promise of the Blinking Cursor

Look, I hate to be the one to tell you this, but if you roll up to the valet in your sports car, and you have women sitting on top of other women in the front seat, and they end up in the fountain, and you end up buying the hotel, at some point someone’s going to look you in the eye and tell you, “Bruce, it’s not who you are inside, but what you do that defines you.” This was made very clear in the critically-acclaimed documentary Batman Begins: https://www.youtube.com/watch?v=6poX_8CT9kM

Do not despair, though, because as that film deftly demonstrated, if you focus on what is in your heart and let it lead you down a path of hand-to-hand combat and justice, eventually you can redeem yourself by lifting a woman and a small boy up from street level to the roof while an old man plays with your car fob: https://www.youtube.com/watch?v=3e6TU9UsA9M

This is a very important service to the community, because people often find themselves confined to street level with no direct roof access. This is due to a part of the fire code in many metropolitan areas that requires locked fire doors between a building’s main stairwell and its roof. Also, old people are usually retired and don’t have much to occupy their time, so something as simple as a car fob can provide hours of enjoyment and really make a difference in their waning years.

Anyway, I don’t want to get bogged down in specifics. The point here is that it’s what you do that defines you. Rachel Dawes knew it, Batman knew it, and now you know it.

But unfortunately, once you know it, you can’t help but apply it to video games. And if you do that, you come to some rather dismal conclusions about their fiction.

We run, we drive, we shoot.

OK, so here’s a question for you: what is, you know, Grand Theft Auto V about? Or Call of Duty: Advanced Warfare?

In a pre-Batman analysis, where we ignore what the player actually does, you might describe those games in a way that sounds like they’re about their fiction: “Oh, it’s about these three guys from different walks of life who…” etc., etc.

But post-Batman, where what the player does defines what the game is about, the descriptions wouldn’t sound anything like fiction at all: “It’s about driving, running, and shooting.” “It’s about running between two prescribed points while shooting.”

This is what most modern games are “about”, and it’s not an accident. We’ve gotten really good at creating experiences where players do things like move around, attack, even build. If you look at something like Grand Theft Auto V vs. the original Grand Theft Auto, it’s obvious that the intervening years have seen massive improvements to just about every aspect of what the player can do. Every aspect except, of course, the part dealing with the fiction. In these games, the player still — often literally — can’t do anything with the fiction.

And this, in my opinion, is the biggest elephant in the gaming room today: despite advances on nearly every front, if we look at the level of interactivity a player can expect from the fiction in a game, things haven’t really evolved since about 1986.

Now, the nice part about something like Grand Theft Auto is it doesn’t really matter that the fiction isn’t interactive. That’s not why you’re playing it. The fiction provides context for the experience, and your desire to interact with the game is satisfied by the fact that physical things in the game are interactive: driving cars, shooting guns, running, etc. Each year, the world simulation underlying games like this improves, providing greater levels of interactivity, so the experience really does become more rich even if the fiction itself remains non-interactive. Batman-wise, the game is still about driving and shooting, but there’s absolutely nothing wrong with that.

However, this poses a very real restriction on the types of games that can be developed using this style. It requires that you have some large set of physical interaction sequences sufficient to satisfy the player’s desire to actually play something, because otherwise you have nothing off of which to hang the fiction. And that is not such a good thing for the medium as a whole.

There’s a reason they’re called “walking simulators.”

In recent years, developers have tried to move past these limitations by skirting them. Games like Dear Esther and Gone Home have explored the possibility that non-interactive fiction can be presented without intervening action. While the success of titles like these is great from the standpoint of opening up more ways to tell stories on computers, it still fails the Batman test just as completely as anything else — in terms of what the player does, Gone Home isn’t about a girl and her family. It’s about walking around.

Which leaves us with only the modern, direct-if-distant descendants of interactive fiction itself — games that might bear the visual novel tag on Steam, or the comparatively small portion of the playtime in RPGs like Fallout or Mass Effect where the player choses dialogue options. These games still try to make it appear as if the fiction is interactive, but by nature of their construction, it is all too obvious to the player that this interaction is very limited, like a glorified choose-your-own-adventure. Even in what people generally consider to be the most interactive game of this flavour, The Walking Dead, the player’s biggest “choices that matter” are actually just binary decision points so obvious that the game actually displays a screen comparing your binary choices to other players’ choices at the end of each play-through.

So truly, at present, we really don’t have any such thing as a game that features real, deep interaction with the fiction. We either have a) games where non-interactive fiction is intercut with interactive action, b) games where non-interactive fiction isn’t really intercut with anything at all, or c) games with coarse, superficial fictional interaction.

Subjectively this might not be considered a problem, because plenty of players enjoy all of these types of games. But in the objective sense, it’s definitely a problem, because it means we have a big huge gaping hole in our space of possible creations. We have no idea what it would even look like to have a game where you can actually interact with the fiction. And we definitely don’t know what kind of wonderful new experiences we might be able to build if we did.

Now at this point you’re thinking, “Casey, that’s all fine, but what does this have to do with the bygone era of so-called interactive fiction? Surely if you’re saying that something like The Walking Dead was barely interactive, then you’d have to say the same thing about the average Infocom game from the eighties.”

And you’re totally right. But there’s one crucial point that this overlooks, and I wouldn’t be doing my job as a random person who writes a blog if I didn’t tell you what it was.

With a blinking cursor, anything could have been possible.

What the player could do in an interactive fiction game era was, almost universally, quite abstract and varied. A few things were common, such as the ability to navigate between “rooms” and manage an inventory of items, but beyond that, almost all other actions were specific to the fiction.

A typical interactive fiction game set up situations that could be about almost anything, and your interactions with those situations would see your character doing just about anything. In one game, you might find yourself handing a magic pigeon perch to the knight on a living chess board, while in another you might find yourself cutting the chalk lines of a pentagram to escape a Lovecraftian summoning ceremony.

Now to first approximation, these actions were nothing more than the great grandfather to today’s quicktime events. Most interactive fiction sequences involved a rote series of steps that you had to do, and no matter what other actions might have seemed logical or plausible, the game was only going to let you do the specific steps the designer had programmed.

But, unlike quicktime events, which specifically tell the player what gamepad combination they must perform to continue the action, the interface to an interactive fiction sequence was literally just a blinking cursor. The instruction to do each step had to come entirely from the player’s own brain, and it was in that specific moment — the moment when the player is considering all the possibilities and choosing to try one of them — that interactive fiction really differentiated itself as an experience as compared to anything we make today.

If you didn’t have the pleasure of living through the era of interactive fiction, you may never have experienced this, but trust me, the feeling of staring at a blinking cursor, thinking about typing your next command, is unlike anything we’ve been able to reproduce since. It was the perfect moment of fictional immersion. It felt like you were choosing what your character did next, even if you really weren’t. Your imagination was running wild and you really did believe you were the character in the story.

And not just a story that had to work within a gamey-action context, but any kind of story. Sure, it could be about gaming staples like magic or futuristic robots or a space marine, but it could also be it about a diver in a run-down seaport, or an Englishwoman kidnapped by pirates or even just a guy trying to change his address with the bank.

Thus the typical interactive fiction game, much more so than anything I’ve played in recent memory, delivered a unique experience where the actual things you did in the game were closely tied to the fiction itself. Blissfully unaware of any requirement that games be about well-defined, repetitive actions, interactive fiction titles were free to set games anywhere, with any kind of character, and involve any kind of activity. They were truly on-par with other mediums in their ability to engage any subject matter directly.

But sadly, the blinking-cursor moment was always short-lived.

However, as anyone who has played a lot of interactive fiction games will attest, at the core of the genre was always a critical flaw that ensured it could never truly fulfil the promise of its blinking cursor. And that, my friends, will be the subject of next week’s post.

The Cat and the Dragon

I’d like to start off with a fable I’m sure you’ll all recognise.

Once upon a time, in a medieval sort of storybook kingdom or something, there lived a village of people and it all got burned down. When the smoke finally cleared, the survivors gathered around and said, “who has done this to our village?” One of the town elders said, “Right before the village burned down, I saw a cat knock over a pot of highly flammable oils!” at which point the town dragon said, “That was almost certainly the problem. I’m glad we got to the bottom of that!” He burped a fiery burp, and they considered the matter settled.

By now of course you recognise this story, the famous episode of “The Cat and the Dragon.” It’s a cautionary tale about the readiness of people to place blame on things which have little or nothing to do with their ultimate problems. The cat, in this case, may have had some small role in exacerbating the fire, but it was almost certainly the dragon who caused the village to burn. After all, that’s what dragons do.

I chose to open the post with this timeless yarn because it fits the topic perfectly once you realise it is a metaphor, a term literary scholars use to refer to an author’s substitution of something they’re trying to describe with either a cat or a dragon, or in this case both. This is not to be confused with the closely related term simile, which is when the author merely says something is like a dragon or a cat (but not both — that special case is called a gerund).

But enough with the advanced grammar lesson. Back to the interactive fiction situation.

SPOILER ALERT: Interactive fiction is the burned village.

In last week’s post, I talked about how interactive fiction had a little bit of magic in it, magic that we’ve never really managed to recapture. But that magic was largely illusory, and it remained that way until the genre faded into obscurity.

So the question naturally becomes, why? Did developers not know what the magic was? Did they not know what was necessary to fully realise it? Did they not have the technology to do so? What?

Well, thanks to 30 years of interceding history, we are at least in a position to make an educated guess by looking at what happened to the genre and its descendants.

First, the presentation of the genre transitioned from textual to graphical. This was typified by games like the Sierra line of graphical adventures, such as the famous King’s Quest series. These were almost exactly the same as the original text adventures, in that you still typed text into a command line to do things, but they now had pictures of the world accompanying them and a character which you could move directionally.

Next, the interface of the genre transitioned from textual to graphical. This shift was exemplified by the LucasArts series of “point-and-click” adventure games, where the command line input of Sierra’s titles was replaced with a cursor-based interaction model.

Finally, the abstract world representation central to the genre was replaced by concrete, rule-based designs, and one might say that at this point there was nothing really remaining that one could call “interactive fiction”. The classic example of this sort of game would be Myst, and of course a more timely example comes out in just a few short weeks.

There is a ton of stuff I’m glossing over here, but for the sake of keeping this blog post at a reasonable length, that’s the high-level summary (for the low-level summary, check out Jimmy Maher’s amazing history of interactive fiction).

SPOILER ALERT: The parser is the cat.

If you listen to people talk about the problems with interactive fiction, one thing gets said over, and over, and over, and over: the parser was awful.

“The parser” — the part of the game code that tries to figure out what you wanted to do based on the words you typed in — has always had an awful reputation, and deservedly so. Natural language processing is a difficult thing to do correctly even with today’s extensive research and extraordinary levels of computing power, never mind thirty years ago on computers with, you know, 64 kilobytes of memory. So even though the later interactive fiction titles (most notably those by Magnetic Scrolls) did a significantly better job understanding what you wanted to do when compared to earlier games like Colossal Cave Adventure, there were still tons of parsing problems. Humans are excellent with language comprehension and computers were (and still are) not.

For those who never played text adventures, a typical “parser problem” might be something as simple as the computer saying, “You are standing atop a tall rock spire, with exits to the east and north,” to which you respond, “look at rock spire,” and it says, “I do not know the word ’spire’.”

I know it sounds stupid, but that’s literally the sort of thing that would happen. If the programmers didn’t actually code recognition for all the words they used in the prose, then merely using those words in a command would immediately cause a failure case. And things got worse from there. If the grammar you used was a pattern it didn’t know, it didn’t matter if it knew the words or not — typing, “I want to look at the rock spire” would completely confound the parser, even though it’s much more natural English than the kind of command it could actually handle, “examine spire”.

So this was awful, and I would never argue otherwise. But if you recall the brief history I gave in the previous section, hopefully you can see why blaming the parser for our failure to develop the magic in interactive fiction amounts to blaming the cat for burning the village: subsequent revisions to the genre removed the parser entirely, but the genre still became commercially unviable.

SPOILER ALERT: The simulation is the dragon.

In truth, though, I wouldn’t actually say that people complaining about the parser are wrong about the main problem. Rather I just think they are complaining about more than just the parser. Because even though “the parser” is a very specific thing that deals only with mapping natural language input to game world operations, the results that people see from their use of the parser are based on the entire round trip.

And that round trip is really where the problem is.

Interactive fiction — especially of the text adventure variety — effectively promised the player a complete world of possibilities limited only by their imagination. When the game said you were “standing on top of a giant rock spire”, you imagined a world with you standing on top of a giant rock spire, and everything that implies. You wanted to know about the rocks, and the moss, and the grasses, and could you take the grasses, might they be medicinal grasses, could you light them on fire, are there bugs in them, what kind of bugs, etc., etc., etc.

The game engaged the power of your imagination and took full advantage of it to create a rich game world, but in terms of the game simulation, almost none of the things you imagined actually existed. There wasn’t any grass on the rock spire because nobody had programmed that. There weren’t any bugs in the grass because nobody had programmed that either. In fact there may have been literally nothing at all beyond just an empty node in a graph with the text, “You are standing on top of a giant rock spire.”

And this created a deep, irreconcilable mismatch between the game you thought you were playing and the game you actually were playing, and you were repeatedly and unceremoniously reminded of that fact not by the parser, which certainly did fail often, but by the lack of simulation, which almost always failed to provide the interactivity your imagination expected from the text with which you were prompted.

This, I believe, is the fundamental reason why interactive fiction titles were so frustrating, and continued to be frustrating through numerous revamps. Changing from textual to graphical made it clearer what the scene was like in detail, sure, but it remained almost entirely non-simulated: most of the doors couldn’t be opened, most of the objects were just pixels that you couldn’t pick up or interact with, etc. And even those objects you could interact with still retained the problematic property that the only interactions that actually worked were the specific ones the designer wanted you to do to progress through the game.

Unfortunately, nobody ever solved — or even attempted to solve — this central problem. There are really only two types of modern descendants to interactive fiction: ones that have exactly the same limitations and problems (e.g., any game from Telltale), and ones that have removed the core game mechanic of objects and world simulation and replaced them with something concrete and easier to directly perceive, like the “clickable state machines” of Myst or the “path tracing” of The Witness.

So where does that leave us?

If you’re sitting there thinking, “Game over, man, because, like, there’s no way we’re going to be able to simulate everything a human can imagine,” don’t despair. We may not need to solve such a hard problem to still solve the part that actually matters to the playing experience. And that part, as you may have guessed, will be the subject of next week’s post.


continued in the comment below

4 Likes

continued from above


From Space Ace to Doom

For any type of game, there’s a point where it goes from feeling non-interactive to feeling interactive. Since we don’t have much in the way of scientific inquiry here, it’s hard to say much beyond that, but we know such a point must exist.

Why? Well, we know there are things that feel interactive, and things that don’t. So if we keep changing one of them until it becomes like the other, at some point in between, the change must occur.

Picture, for example, a continuum of games that feature a guy with a gun who is going around shooting bad guys in an effort to get through all the levels in the game. You can use Space Ace and Doom as your two extreme points.

Err, well, maybe you’re too young to do that, so… well… if you never played those games, maybe go watch them on YouTube and then try to picture it? Here’s Space Ace: https://www.youtube.com/watch?v=wU3weqFVb7o

All you did in this game — and I mean literally all — is when something on the screen flashed white, you had to guess which direction on the joystick to push in order to not die. That’s it. You didn’t get to control anything or do anything.

Now on the other end of the spectrum there’s Doom: https://www.youtube.com/watch?v=wU3weqFVb7o

Unlike Space Ace, in Doom you can move the location of the guy freely, change what direction he’s looking (on only one axis), and shoot your gun in that direction. All of these things happen reliably when you do them — you don’t have to wait for any specific time.

Doom’s basic design can still be seen everywhere in the market today. Space Ace’s is nowhere to be found.

Doom (and it’s predecessor, Wolfenstein 3D) were the ancestors of many of today’s best-selling titles, like Call of Duty or Battlefield. Space Ace (and it’s predecessor, Dragon’s Lair) were the ancestors of — well, nothing we still play today, really.

“Playing” Space Ace never felt interactive at all. It felt like watching a movie that required you to do something periodically in order for it to keep playing. Doom, on the other hand, felt like really being in a weird moon base, where you were deciding what to do and honing your attack skills for fighting the monsters that were repeatedly thrown at you.

In short, Space Ace is clearly on the “non-interactive” side, and Doom is clearly on the “interactive” side.

So what makes the difference?

If we added the ability to control the direction of the camera to Space Ace, would it suddenly feel interactive? How about if we gave you the ability to control the camera and the ability to shoot in the direction you were looking?

Amusingly, although it sounds purely hypothetical, we sort of know the answer here. It’s “no”.

I said Dragon’s Lair and Space Ace weren’t ancestors to anything we play today, and that’s pretty much true. But they were ancestors to one more series of games that also failed to have a long lineage.

Space Ace and Dragon’s Lair were arcade games, and they worked by having a laser disc in the cabinet so they could play back full-motion video. This was in the very early days of gaming, so their graphics looked amazing compared to everything else. When CD-ROM drives came to home computers, the same situation repeated itself, and a new wave of Dragon’s Lair-esque games came down the pipe. But this time, since it was a decade later and there was more computing power to draw upon, game developers tried a few other things to see if they could cross the interaction boundary.

They never succeeded.

They tried everything, including spherical projection of the movie to make it appear as if you could control the view direction, and real-time compositing of the enemies so you could “shoot” them directly:

But apparently, it was never enough. They even tried to make “interactive fiction” games this way, going so far as to cast Sir Robert Schneider in the leading role for the cinematic masterpiece, A Fork in the Tale:

Yes, that’s a real “game” right there.

But I digress. The point is, as far as the history is concerned, we don’t really have much evidence that there’s anything you can take away from Doom without the game starting to feel noticeably less interactive.

All of this is highly subjective.

Thankfully, exactly what does and doesn’t have to be in Doom to make it feel interactive isn’t too important to our project. We don’t need to know the specifics for that genre.

All that actually matters for our purposes is that we see there is some point, somewhere between Space Ace and Doom, where games go from feeling non-interactive to feeling interactive. Even if the point differed for every player, there’s always some threshold there. If the game can cross it, it feels interactive to that player, whereas anything less wouldn’t.

Space Ace doesn’t feel like a game to me, it feels like watching a movie. Doom feels like a game. Somewhere in between, the game design crosses from one to the other.

It’s this point that concerns us if we want to make fiction feel interactive. Again, it’s all subjective, but speaking for myself, nothing you can buy today crosses that point for me in terms of the fiction. Lots of things cross that point in terms of other aspects of their game design, but just not the fiction.

Fiction-wise, games are still Space Ace. We need our Doom.

This is why, in the previous post, I said you needn’t despair: we don’t have to produce any kind of magic reality simulator in order to get fiction that feels interactive. We just have to make a game that crosses the threshold — that gets us far enough along to feel like the fiction equivalent of Doom instead of Space Ace. After all, Doom barely simulates anything about reality, but it completely succeeds at delivering a great interactive experience.

But how do we know where that point is for interactive fiction? Is it impossible to even know what it looks like if we’ve never crossed it?

Thankfully, if we go back about twenty years, there may be an answer waiting for us. And that, of course, will be the topic of next week’s blog post.

Volitional Fiction

It was the end of March, 1997 — a fabled time that saw the triumph of human creativity in all its forms. The methods of modern movie-making had brought cinema to the apex of its storytelling capabilities, and its stars were renowned for their subtle and emotionally moving performances. Movie-goers themselves were all seasoned aficionados in their own right, selecting only the finest movies to patronise, as evidenced by the number-one box office movie at the time, Liar Liar starring Jim Carrey: https://www.youtube.com/watch?v=C1no75lpOiw

Not to be outdone, the recording industry drew primarily from classically-trained musicians and prodigal composers to transform the very concept of what popular music could be. Bold collaborations between complex and storied performers took the “pop single” to the next level, as exemplified by the then-number-one chart-topping single, poetic masterpiece Wannabe by the Spice Girls: https://www.youtube.com/watch?v=gJLIiF15wjQ

With such unprecedented splendour abounding, the world could almost be forgiven for overlooking the proceedings underway at a small hotel in San Jose, California, where a few hundred developers were gathering for the Computer Game Developer’s Conference.

It was there, in a sterile meeting room bearing some of the least provocative beige wallpaper yet discovered by science, one of gaming’s seminal designers, Doug Church, presented for the first time the concept of the Formal Abstract Design Tool.

If you aren’t a game designer, you probably don’t know or care what that is. And lucky for you, you don’t have to. Because the main things I actually need to reference from that talk are the examples he gave of such tools: intention and perceived consequence.

Intention and perceived consequence are what make a game feel like you’re actually playing it, not watching it.

At their core, these concepts are very simple. “Intention” means that the player has the ability to make a plan. They can look at the state of the game, decide something they want to try, determine the steps necessary to try it, and then execute those steps.

“Perceived consequence” means that when a player performs an action in the game, they can tell the action had an effect on the state of the game. They can use this perception to draw conclusions about how the game works, and what is possible within the boundaries of the game system.

As you can see right away, these two concepts work together. When the game has both, the player is able to make plans and attempt them, perceive the results, and use that new knowledge to inform future plans that are more likely to produce the desired outcome. The cycle leads to a satisfying feeling of real interaction where the player chooses both what to do and how to do it, and the reaction of the game to their actions doesn’t feel arbitrary.

There’s a lot more here to unpack if you’re a game designer, certainly, but even just the cursory definitions are enough to see what’s wrong with current interactive fiction (or “narrative-driven”) games.

What game developers (and reviewers) currently mean when they say “choices that matter” is at best only perceived consequence, but never intention.

A lot of narrative-driven games have neither. For example, the class of games players call “walking simulators” are just experiences that you literally walk through. There’s nothing for the player to choose, so obviously there’s no planning (other than what route you walk, I suppose). And since there’s no planning and no action, there’s no perceived consequence.

Some games, such as those from Telltale or Quantic Dream, do claim to allow the player to make “choices that matter”. And at face value, they are correct — there are points in these games where players can choose between two or more possible things, and later events in the game will change depending on which thing is chosen.

But the problem here is that while this does satisfy the most basic need for perceived consequence — the player chooses option A and sees that something happens — it provides no actual intention to the play experience. These choices are merely single, scripted instances in time, presented to the player without their initiation — the player never makes a plan or decides what to do. They have a situation thrust upon them and are asked to decide which to do.

It’s basically Space Ace all over again, the only difference being that instead of choice A leading to more of the game and B leading to death, both A and B lead to more of the game. But the nature of the choice hasn’t really changed.

How about a concrete example?

To pick a classic example from Telltale’s The Walking Dead (which occurs in the first episode, so you can quickly play up to it if you’d like to experience it yourself), at a specific point in the game, no matter what you’ve done up to that point, the game puts you in a situation where you can save only one of your compatriots from the zombies: https://www.youtube.com/watch?v=9AMkmRCau8k

This is the primary thing that is meant by “choices that matter” when used to describe modern narrative-driven games — literally, the writers wrote the remainder of the game with two variants, and depending on which character you chose to save, you get that set of scenes.

Is there perceived consequence? Absolutely. The character you choose to save is the only one that will appear for the rest of game.

Is there intention? Certainly not! The player had literally zero intention to set up that situation. In fact, there was no way they could have avoided the situation, because the game is scripted to go through an exact series of scenes, and no matter what “choices” you may have made earlier in the game, this scene will happen.

The player learns nothing from this about the game. They can’t learn that there was a deficiency in the way they made their plan, or what they did leading up to this scene, because they never made a plan in the first place. They never made “bad choices” that lead to the death of one of these people. They never made any actual choices at all. Like a mine cart ride where all the player can do is frantically throw a switch as a junction point comes up, the player is never given any intention. They are merely along for the inevitable ride, and invited to react to what comes at them in the classic Space Ace tradition.

What we want is an interactive fiction game that has both intention and perceived consequence, working together.

So finally we come to what I believe is the core, as-yet-unsolved problem in interactive fiction: providing real perceived consequence and real intention. If we could make a game that had both of these things working in concert, the player would finally be able to truly assume the role of the main character in a fictional sense, making real decisions about what they wanted to do — and how they wanted to do it. They could attempt to actualise their decisions, see the results, and then make new decisions based on the outcome.

Developers have been able to make games that satisfy these two criteria in almost all game systems except the fiction, and to great success. Modern classics like Minecraft and Counterstrike have these elements in spades, and they are incredibly rewarding experiences for players. Everyone from beginner to expert can play these games and come up with new goals and new plans to accomplish them, use the consequences they perceive as a result trying to execute those plans helps them further refine their understanding of the game.

And these plans needn’t be about winning or losing, success or failure. Sometimes they’re just about accomplishing something interesting, to satisfy curiosity, or to build something unique, or even just to make people laugh by the silliness of the results.

We need to figure out how to do these things with fiction.

To avoid overloading the terminology, maybe we should coin the term “volitional fiction”.

For this entire series of blog posts, I’ve been using the term “interactive fiction” to refer to… well, that’s kind of the problem. I’ve been using that term because it’s generally accepted, but the precise definition really isn’t clear. At face value, it sounds like it’s talking about works of fiction that are interactive. But in reality, it has been used almost exclusively to refer to works where the fiction was not interactive, and which gave only the illusion of requiring the player’s participation. So it’d help to have another term for what we’d have if the fiction itself was actually highly interactive, and satisfied the requirements of both player intention and perceived consequence.

For lack of a better term, I propose using the adjective I have often used personally to describe what I mean when I talk about real player control of fiction: volitional. I think of “volition” as the thing that a player doesn’t really have in today’s narrative-based games. The fiction happens to them, and they only affect it in the most unintentional and unsophisticated ways. Thus we could use “volitional fiction”, a term which Google assures me has almost never been used before, to unambiguously refer to a game where the fiction itself was as interactive as all the other elements of a modern game.

So, why hasn’t anyone been making volitional fiction?

If we’ve known — or should have known — what would make a fiction-centric game feel volitional for almost twenty years, why don’t we have tons of volitional fiction games already? Are the designers and writers of narrative-driven games lazy? Are they ignorant of the history? What’s going on here?

The answer is probably neither. In fact, the answer is likely that the problem has nothing to do with the designers or the writers. The problem lies mainly with the technology.

But based on the distance between this sentence and the footer, I think you know what I’m going to say next: how and why the technology is the problem is the topic for next week’s post.

Wouldn’t You Prefer a Nice Game of Chess?

Have you ever read a Choose Your Own Adventure book? You know the kind of book I mean — where you get to the bottom of a page and it says something like, “If you decide to bargain with the man for an extra pound of frozen peas, turn to page 47. If you decide to try using the trebuchet to get past the mutant fish-people, turn to page 131.” You, the reader, carefully consider these options — clearly the best two options for the situation, and the only two worth considering — then you turn to the page indicated for your choice.

Or, if you’re one of those people who cheats at literally everything, including your entertainment, you read both pages and then back-decide which choice to make. Far be it from me to accuse you directly, but you know who you are. You probably have a bunch of GameFAQs tabs open right now.

Anyway, if you’ve ever read a book like this, then you already know how modern “interactive” story games work under the hood. The methods used in games today are only slightly more advanced than the methods authors were using in “interactive” books before personal computers even existed.

While small variations are easier, major variations are still built as manually-written branches.

There are some widespread improvements that have been made thanks to computers. Most games that feature “interactive” narrative now have easy ways for authors to insert or remove parts of their text based on the state of the game, making small variations easier. But on the whole, the major method of construction has not changed: authors still have to write out all the major plot-lines of the game, in their entirety, by hand.

This is not a huge problem if the goal is to just make lightly interactive fiction, where the player gets a little bit of choice but basically plays the same story as everyone else who plays the game. But it’s a huge problem for those of us who want actual volitional fiction. And it’s not just a practical problem, it’s a theoretical problem.

How about Global Thermonuclear War?

As with most theoretical problems involving interactive narrative, the fundamentals were thoroughly explained in the exemplary 1980s documentary War Games. I would call your attention to the following scene, in which the computer is instructed to play itself at Tic Tac Toe by setting the number of players to zero: https://www.youtube.com/watch?v=s93KC4AGKnY

Now, obviously, since it was filmed in the 80s, there are some differences in how this works on today’s more sophisticated computers. For example, on PCs you cannot always set the number of players to zero and have the computer play itself like you can on the WOPR. But everything else in the scene still applies.

What the computer was doing in this scene — if you didn’t see the movie and thus don’t know the context — was learning about how branching narratives are impossible to write by hand. One might also argue it was learning how good linear narratives are also impossible for certain screenwriters to write by hand, and that possibly the WOPR should fire its agent, but that was sort of a larger realisation that it came to later, after the day’s shoot, when it was drinking away its troubles in its trailer.

But I digress.

The point is that even a game of Tic Tac Toe, as simple as it may seem, turns out to have over two-hundred thousand different possible playthroughs. Over two-hundred thousand. This is a game that has only 9 possible moves on the first turn, 8 on the second, 7 on the third, etc. — hardly anything anyone would consider a good game, let alone a great game. And it’s a game most people wouldn’t even bother playing as adults because the space of possible games is so small that even a completely untrained and unstudied human quickly gets good enough that they can never be beaten, only tied.

Things only get worse in terms of game complexity as you move on from Tic Tac Toe to games that people actually think are worth playing, like Chess. Chess has so many possible playthroughs that we don’t even know how many there are.

That’s not me being funny. We literally just don’t know. Type “Shannon Number” into Google if you don’t believe me.

So now project this back onto the authoring process for a narrative game where we want the player to feel like they have real control and are playing a game with real possibilities. In other words, something more engaging than Tic Tac Toe. Maybe it doesn’t have to be Chess, but it’s gotta be somewhere in between.

Hopefully you can see that solving this problem directly is intractable — writing substantially more than two-hundred thousand different stories isn’t something you can expect any game designer or author to be able to do. And if you wanted the interactivity level to be on par with chess, well then they’d have to write a minimum of ten duodecillion stories — that’s a one with 40 zeroes after it.

Thus we come to the other important thing War Games teaches us about manually-authored branching narratives:

The only winning move is not to play.

Writing volitional fiction by hand isn’t possible. It’s just that simple. There’s literally no point in trying.

That’s why last week I said that the reason nobody has been able to ship a volitional fiction title yet is because we simply don’t have the technology. Whatever game you’re playing right now that’s got some narrative elements to it already is — to a first approximation — as volitional as game fiction is going to get until there are some technological breakthroughs.

Yes, there are plenty of small things games have done to help reduce the writing workload, and yes, there are probably some more of those to be discovered. But small improvements like these do not even come close to providing the kind of amplification necessary to cross from the sub-Tic-Tac-Toe scale to the Chess scale of interactivity.

And at the risk of starting yet another major grumpy grumblefest thread on Hacker News or Reddit or wherever else, I should also mention that academia has produced no practical solutions to these problems, so it’s not like games have been ignoring valuable existing research. No, that’s not me being unaware of your favourite story generation paper — that’s me being aware of it, and saying it fails to solve the problem. You’re welcome to disagree — but of course, if you want to, you’re going to have to go make a great volitional fiction game based on said research and prove me wrong.

So what do we do now?

Well, unsurprisingly, a problem of this magnitude that has resisted easy solutions for so long is not going to be solved overnight. If we want volitional fiction to become a reality, somebody needs to establish a long-term initiative to improve the state of the technology.

A little over a year ago at Molly Rocket, we decided to be just such a somebody, and we’ve been working hard since then to set up a long-term initiative that we believe can make real progress. Just what that will look like over the next few years, if we’re lucky, will be the subject for next week’s post.

Fighting the Dragon

Let’s be realistic. Even if I’m lucky, in about ten years it’ll probably be time to put me out to pasture. I’ll be fifty, and I’m guessing by then my hands will be arthritic and RSI’d to the point where I won’t be able to do much programming. That’s just the sad price you pay for coding a lot throughout your life.

Before choosing Molly’s latest project, I felt it was time to honestly ask myself, “what is the most important project to work on right now?” Since I knew I wouldn’t get the opportunity to work on that many more, choosing judiciously felt incredibly important.

The more I thought about the question, the more I kept coming back to the same answer. For almost thirty years I’ve been thinking about interactive story technology, and throughout that entire time, I’ve never actually worked on an interactive story engine. I’ve worked on lots of difficult engineering problems, but that was never one of them.

So that’s why Molly Rocket’s Volitional Fiction Project — for lack of a catchier title — was started. Its mission is simple: finally solve the technology problem I laid out over the past six blog entries. And not just claim to solve it, or publish some papers, or make some tools that never really pan out, but actually solve it for real, where there’s a game to hold up as unequivocal proof that the technology works and finally makes great stories that are deeply interactive.

Yes, this is an incredibly difficult problem.

We know that’s true because if it was easy, someone would have solved it already. People have definitely tried. But they haven’t made much progress.

However, we don’t think that means the problem isn’t solvable. Hard problems take a long time to solve, and they have to be solved in steps. The first traditional FPS wasn’t Call of Duty, it was Wolfenstein 3D, and the vastness of the technical problem space that lies between the two is hard to overstate.

So what we’ve set out to do is to ship a series of games, each featuring a new set of technological improvements, so that we can study and break down the problems of volitional fiction in much the same way as the industry did during the development of modern 3D rendering.

We think there’s something very important about this process — about the intersection of research and commercial development — that drives the development of real, practical technology that actually works. So our game plan looks like this:

Step 1: Put together a great art team and a great graphics engine.

We don’t want to ship anything that isn’t gorgeous. We want everyone to be excited about playing our games, not just people curious about what the improved fiction technology will play like. And more importantly, we want to be able to clearly see if the tech holds up when it must drive compelling visuals, since that’s the way modern games work, and technology that falls apart when it has to literally show what’s happening simply can’t be considered good enough yet.

Happily, we’re just about done with this, so that part of our project is essentially complete. Which brings us to…

Step 2: Iterate, iterate, iterate on the fiction technology until it’s great.

And that’s what we’re starting on now. We know we can hit a certain level of fiction interactivity with a level of tech we already know works, but we want to push the tech as far as we can with each game. So we’re planning on spending about eight months of solid iteration time improving the fiction technology. That’s what we’ll be doing all this year, right up until we launch the game.

After that, if we’re lucky and the game does well, we’ll start on the next one, where we’ll push the tech even further. And the same with the one after that. And the one after that.

Step 3: Flame out on the last hard problem.

Yes, that’s really Step 3. I have a long list of specific technical problems that I believe need solving in order for volitional technology to be considered complete, and I see pretty clear paths to solving all of them except for one. It’s only required for very specific types of fiction, but it’s a super nasty problem and I’m not sure if it’s solvable or not.

So if we get all the way there, and we’ve solved everything else, we’ll do one last volitional fiction game and try to solve it. And probably totally fail. But, hey, that’d be a great place to be, because it means everything else — which is way more important — got solved already, and that would be totally awesome.

So that’s the situation as it stands.

We’ll be announcing the first game in the project real soon now — you’ll know it’s coming, of course, when you see Humphrey Bogart Pterodactyl peek his brooding face up above the cloud line. If you’re interested in supporting our project, please consider signing up for our mailing list in the footer of this page so we can keep you up-to-date. In order to succeed in the long run, we need lots of help from all you folks out there, for everything from spreading the word about the game, to alpha- and beta-testing the engine so we know it’s rock solid.

It’s a tough road ahead, to be sure, but we’re up for it if you are. Yes, tackling such an apparently insurmountable problem as volitional fiction with a small team and limited resources is much like trying to kill a dragon with your bare hands. But I for one never forgot the lesson of the Colossal Cave I learned back when I was three:

When the little prompt in your head asks you if you really want to try to kill a dragon with your bare hands, the best answer is usually “yes”.


Hope you liked it.

5 Likes

I was curious if Muratori ever actually made any “volitional fiction” games, as discussed in the last post. Based on this subreddit, it seems not… But here’s the archived version of the post announcing what was to be their first game: Announcing 1935!

Edit: Previously discussed on this forum! New "interactive fiction" work in progress by Molly Rocket

3 Likes

To boil this down way, way too far, and to then respond only to the parts that interest me:

The idea of giving players more opportunity to establish their goals and intentions within the framework of the narrative appeals to me. So far, so good.

Muratori concludes that the only way to provide this level of flexibility to the player is through automation—and this is where he loses me. He uses the excuse that “even a game of Tic Tac Toe, as simple as it may seem, turns out to have over two-hundred thousand different possible playthroughs.”

I contend that, for both volitional fiction and for tic tac toe, he has confused “uniqueness” for “meaningfulness”. If we care about meaningfully distinct games of tic tac toe, there are only 14 games of tic tac toe that we need to care about. (Or maybe 27, or 3, or—well, the exact number is up for debate, as outlined in the video I linked, but the point is that it’s a dramatically lower number than six digits that Muratori has provided.)

Sure, it’s possible to conceive of a narrative engine in which the player can make any conceivable choice at any conceivable moment—but the result of this freedom is something closer to AI Dungeon. It’s “let’s play pretend”.

I’m being a little uncharitable by reading his argument this way; from the little bit of information available about his plans for 1935, I think his approach was perhaps closer to this Narrative LEGO concept that Ken Levine described.

Have we ever seen any actual games built around systems like these? Most systems in this vein that I can recall become, at their core, “sometimes you can make temporary or permanent allies out of groups of enemies.†” Or “give the NPC gifts to improve your relationship with them.‡” Basically, the more systemic characters and storylines become, the more emergent they become, the less they feel like stories to me. They feel like toys.

† (Later Far Cry games, Metal Gear Solid 4, perhaps the Nemesis System from Middle-earth: Shadow of Mordor)

‡ (Various dating sims, mainly, I think? This isn’t exactly how it works in the Shin Megami Tensei Persona games, but it’s a similar dynamic)

The conclusion I’m drawing from all this is that, if you want to have a compelling story in the “volitional fiction” mold, this is still a problem of authorial design rather than technology. You can give the player opportunities to set goals, but as with the reactive choices that Muratori finds so unsatisfying, you’re still limited to either making small cosmetic changes to the narrative or large, laborious, hand-crafted branches.

The solution that tic tac toe suggests (and chess, to an even greater degree) is to use both kinds of decisions. If you hide the meaningfully distinct branches among various other choices which are functionally symmetrical, the possibility space explodes in a way that obfuscates just how few meaningful choices actually exist. Provide many choices which are only cosmetically different, several choices which diverge from one another before rejoining later, and (relatively) fewer choices which branch so much that they cannot rejoin. This seems like the only way to write this kind of volitional narrative without making the workload overwhelming.

7 Likes

Hmm, I agree. I also think this is a question not just of technology and design but also of user interface.

Old school turn-based dungeon crawlers where you moved from cell to cell could use four directional buttons on screen that could be pressed by a mouse. Doom (and more importantly Quake) required a new control scheme to allow users to make the decisions available to them with a speed at which that freedom expressed itself. Imagine if you couldn’t hold down the arrow keys to move continuously, but had to tap-tap-tap it. The degrees of freedom that Doom gave would never get explored.

One problem that I see with IF is that the input in “discretised” rather than being “continuous”.What that “continuous” input would look like, I have no idea. Maybe a parser equivalent of Oxenfree’s dialogue system.

3 Likes

When faced with a rhetorical treatise about ‘here’s what’s wrong’, which posits an awesome but incredibly specific lion that’s already in the stone if we could only reveal it, and it’s got a great name ready to have a TM symbol placed after it, I’m inevitably going to disagree an increasing number of times, ultimately falling to contrarianess and disagreeing with everything.

I don’t want to argue at a decade old cross-post of someone not here, so I wish him good luck with solving his problems, or the problems of us all we’ve done nothing about. At least we hadn’t done anything ten years ago. It’s possible we collectively got off our butts at that point and began to make some token efforts, but I’m not sure.

-Wade

2 Likes

To start: I feel (I won’t try to quantify this) that we don’t see a lot of long-form craft writing these days, so I’m always happy to read some. That includes this piece.

The essay is a provocation. I can confirm this, since I myself felt provoked more than once. I think that’s a good thing! I don’t think I ought to disagree with Muratori point-by-point, but since it might be relevant to players and critics in 2025: I’ve never tried to solve the problem of gratifying open-ended responses to player intent because I don’t consider it a problem. I think when players/movie audiences/readers sometimes expect a work to be made for them, i.e. “This is bad because I don’t like word puzzle games,” they disregard or devalue authorial intent. I don’t agree with that critical stance, and I wouldn’t cater to a wide variety (a little is fine) of volitions even if technology made it easy.

I think some players have misunderstood my own comments about authorial intent: I believe that authors should not try to “herd” players into a “correct” interpretation of a game. That doesn’t mean that authorial intent doesn’t matter–all works are informed by it! Still, even if we can’t ever know an author’s intent, we know it exists. Is it worth respecting? I don’t think that Zork’s Adventurer rejecting material wealth or sparing the thief would be interesting or innovative. I don’t think players would get much out of a new version of Repeat the Ending that allows D to blow up the pharmacy, take the orange-eyed woman on a date, take Brad on a date, or pulverize the dogs. Repeat the Ending might (might!) be about living through trauma, or experiencing the death of a family member, or even survival via the redemptive power of art. Intentionally. The freedom of the player, yes, very important, but what about the author?

As another point not necessarily directed at Muratori alone (I see this kind of thing around): I don’t think parser flexibility is the greatest challenge facing… what do I want to call them? Not parser games… text games with a sophisticated, built-in world modelling system. I believe the chief problem is often the parser-as-parser. Design-wise, it’s not so hard to make a hyperlink-only version of an Inform game (I name Inform because it’s all I know). Practically, though, it seems quite hard at the moment for reasons that seem to be technical. My two most recent games attempted–who knows if they succeeded–to ameliorate the challenges new players face when up against the blinking cursor.

As I’ve already said, I’m grateful any time I see a bit of writing about craft, aesthetics, or design. So I am presently grateful! I see no point in dismantling the argument bolt by bolt, but, as a provocation, it will have done its job if it leads us to an interesting discussion.

10 Likes

I see this happening a lot in game dev. Developers fixate a bit too much on the combinatorial magnitude of a game, and don’t really consider often if the player would ever meaningfully recognize any differences, or feel an impact from any of it all.

Huge agreement.

I could make an argument for Dwarf Fortress, but I have a few caveats to bring alongside:

  1. We are extremely unlikely to see another game like Dwarf Fortress. Its design goal is uniquely unrealistic to chase, but its devs seem uniquely driven to chase it, and have been extremely lucky and fortunate in so many ways to have the environment they need to keep working on it in the necessary capacity.
  2. Dwarf Fortress also does not explicitly convey the story to the player. It has extremely deep levels of simulation (down to the individual organs of creatures, and the political movements of nations), but it’s intentionally left to the player to discover events occurring, defocus their perception a bit, and see deeper narratives that these events could imply. It heavily leans on the human habit of seeing stories told by chaotic and distributed events.

I point us back to Dwarf Fortress. Each planet might be different, but the devs did impart a general narrative backdrop for the setting. Every world will have Dwarves, Elves, Goblins, Kobolds, Humans, etc. These groups will each follow certain behavior patterns as nation-states. This is part of the authorial design, even in a game as deeply simulated and procedural as Dwarf Fortress.

This, exactly.

I think when I was younger, I would have chased the idea of a “do-anything game”, but I have to wonder if there’s a meaningful difference between a game where truly anything is possible, and the activity of opening a word processor, or picking up pen and paper. It’s an interesting philosophical idea, but with every passing year, it seems less and less worthwhile to me. I think, by design, games should have a focus. Yes, we have Dwarf Fortress, Minecraft, and ifMUD, but even those have a focus.

I’m embarrassed to bring up my goal of IF-Octane again, but its entire design fundamentally hinges on this idea. I think there’s an assumption that text games with world models have to use a parser for input, and that’s probably because all of our historical examples of text games with world models have had parsers. Personally, I suspect this trend is because the games which established this trend had investigation and puzzle-solving as core mechanical goals.

The counter-argument I intent to make with IF-Octane is: If a text game with a world model explicitly does not have these elements of mystery and discovery attached (by virtue of a different gameplay genre), and instead requires the player to have clear, moment-to-moment awareness of what mechanics are immediately available, then is a parser even necessary at all?

I’m not someone who aims to make puzzles, or situations of mystery-and-discovery, or games with a strong narrative exploration. I aim to make games with clearly-outlined mechanics, with challenges for the player to overcome by using these mechanics.

Something I learned after the release of I Am Prey, is that the game had a parser because I had noticed this tradition, but as I received more and more feedback and transcripts, it really begged the question of if that game was improved by the obfuscating* nature of the parser, or if that was holding it back from cultivating player awareness in the game. I think a world model is completely necessary for I Am Prey, but I have concluded that using a parser for input was a mistake.

* Clarification: The parser’s obfuscating nature is a good thing, when considering the goals of many parser-based games. My point is that it does not help my own gameplay design goals.

6 Likes

There is ‘kobold’ again. It must be an omen…

1 Like

I would argue that Inkle’s studio practice is creating systemic event models and then narrating the emergent story in a way that feels natural.

3 Likes

I agree with the long analysis. I tried to solve this with Chasm. But if anything that’s too open-ended. And it didn’t really get much engagement anyway (though I didn’t really advertise it, and it’s hard to set up). Nevertheless I think the concept of the player as unwitting author in conjunction with a LLM as a kind of dungeonmaster is sound. I want to try a more constrained approach whereby there is some kind of DSL world description language, and the LLM hallucinates objects in it based on narrative and user input. The world should persist. I had a lot of the elements there but world modelling is quite hard. I would use inform 7 but that’s a compiled language, and I’d like it to be generated dynamically, to keep the tight feedback loop between player and world. Intention and consequence, as the posts said.