"[Interaction] is the Holy Grail... the undiscovered country."
-Chris Crawford, December 11, 2003
Today, curmudgeonly game designer Chris Crawford (Mobygames bio) dropped by to give a talk at my grad program. With a swath of publishing credits and a yen for making challenging statements, Crawford is the sort of person you want to listen to, but also the sort of person you want to disagree with. Crawford hasn't played a videogame in 5 years (except for The Sims, which he played for an hour when Will Wright sent him a copy, looking for feedback), having decided there is nothing more they can teach him. He has tired of game design so much that he is now calling for the founding of a new discipline of interactive entertainment software that will "call itself something else" and have absolutely nothing to do with the games industry. Like I said, challenging statements.
What's Chris Crawford's big problem with modern game design? It's too focused on things, and not enough on people. Crawford asks us to try disecting game interactivity into its base verbs. At his talk, he disected a first-person shooter, but I'll simplify further with an example I used in an essay I wrote for this site a few months ago and analyze Asteroids. Asteroids has three verbs: turn, thrust, shoot. This is a fairly limited vocabulary of action, and not one that is terribly engaging to our humanity. Crawford dares game designers to think of a game's verbs as a job description, and if a game's verbs don't sound like an interesting job, then it's not going to be a worthwhile game. Crawford went on to say that while The Sims is, in his view, a first step in the right direction, it still fails in this regard. The Sims, he says, is about making dinner, cleaning, showering, going to the bathroom, and going to work. Hitchcock once said, "Drama is life with all the boring parts removed." The Sims might be seen, then, as life with all the interesting parts removed.
Interestingly, MMOGs don't fare any better in this regard. MMOGs make up for the fact that we don't possess the algorithmic chops to design really good interactive characters by putting us in a world with hundreds of other people. The problem again is that people are dull. The sorts of stories that we associate with the marvelous fantasy and science fiction worlds that MMOGs invite us to play in cannot happen when interaction comes largely from other people. Characters in movies and books do not act like actual people. Dialog is not just conversation. As was recently quipped, "Star Wars: Galaxies isn't Luke Skywalker's Star Wars, it's Uncle Owen's Star Wars." Crawford likened MMOGs to a window: We can stand on opposite sides of a glass window, and we'll each see each other perfectly, but the window will never give us impressionism or cubism. Even if the game designer's art, interactivity, is crude right now, like all arts, time and the right people will bring us better things than mundane reality.
Crawford exists in a space that many find uncomfortable. He has been a creative force behind many successful game projects, and he was the first person to really write about computer game design, but these days much of what he champions seems at odds with reality. Visionary? kook? Both? In any case, worth paying attention to.
"Visionary? kook? Both?"
You forgot one: egomaniacal prick
Sure he's an expert, and sure he has interesting (if somewhat controversial) things to say, and loads of experience, etc. It's still no excuse to behave like an ass, which he does frequently. The man may very well be brilliant, but I just don't have the time or energy to put up with his BS.
Posted by: Habib | 12/11/2003 at 02:19 PM
Crawford makes an interesting argument, although also a bit too smart for my simplistic tastes. He seems to be saying that games are limited these days because they are, well, games.
However, I believe I can understand where he's coming from, especially in reference to on-line games. I've had experiences ranging from boring conversations about last night's television shows to the always interesting accusations regarding my sexuality, racial background, and the depravity of my mother.
I have rarely, I admit, seen an on-line game promote a large, overreaching story that I felt I was a part of. But, then again, I was also raised on Sierra and Lucasarts Adventure games, so my standards of story in a game may be a bit high.
Anyway, good article.
Posted by: Mike Drucker | 12/11/2003 at 02:21 PM
I like his "verbs" excercise, but criticising games for being overly simplistic is a lot like saying Monopoly is not an accurate financial model for financing a real-estate empire. Well, duh.
I think it's ironic that he thinks games are overly simplistic, when a lot of other veteran game designers (id, I think?) are saying that they're getting too complex.
At least I feel better about my own level of gaming burnout now. I mean, at least I still play games.
Posted by: Bowler | 12/11/2003 at 02:34 PM
On his game credits, there are two that were made within the last five years, but hasn't played either one of them? How does that work exactly, where you get credit for a game you've never played?
"he is now calling for the founding of a new discipline of interactive entertainment software that will "call itself something else" and have absolutely nothing to do with the games industry"
What does this mean? His website focuses mostly on 'interactive storytelling', which makes it seem like he's advocating a return to the godawful "multimedia" era where you put in the CD and watched a story unfold, instead of actually playing within the story as you do with modern games. (You could do a bit more than that with multimedia, but not much.) Games aren't literature, but I'm not exactly sure they should be. Is there a copy of the lecture online somewhere?
My vote is for "kook", unless there is something I'm missing.
Posted by: eli | 12/11/2003 at 02:36 PM
This is a bit premature, but I more or less agree with everything he has to say. I'll type more on this tomorrow, cause I am in the midst of finals. My twist is that the need that the need for a game to be "fun" limits the way the interactive medium can operate, and the potentials for an "interactive narrative" are sharply limited as well.
Posted by: Andrew | 12/11/2003 at 03:24 PM
"a new discipline of interactive entertainment software that will 'call itself something else'"
and how would that work unless it is the interactive storytelling mentioned by eli above?
it seems to me that that would be the only thing to fall into his new discipline. if it has any sort of mechanics or verbs, then it would be a game. if not, it would be boring, or a movie.
and Crawford's problem with games is "It's too focused on things, and not enough on people."
so basically he wants better AI?
he seems to be confusing himself, and me as well, as to what the big problem with games is and why a new discipline that makes games but doesn't call them "games" is needed, other than to be able to market to older people who don't want to play games simply because they're called that.
Posted by: TitusByronicus | 12/11/2003 at 03:27 PM
It seems that indeed what he wants is AI that's good enough to interact with on an emotional level. I think the Turing test is still quite a few years from being passed.
I really wish my car could fly, but I still drive it.
Posted by: Ken | 12/11/2003 at 03:57 PM
Ken, you don't need to pass the Turing test to have a good expressive agent. After all, we don't actually expect Harrison Ford to know anything about archaeology.
You may want to take a look at Michael Mateas' research on believable agents and the work of Andrew Stern and Michael on Facade and other interactive narratives.
Posted by: William | 12/11/2003 at 05:10 PM
Ken, you don't need to pass the Turing test to have a good expressive agent. After all, we don't actually expect Harrison Ford to know anything about archaeology.
You may want to take a look at Michael Mateas' research on believable agents and the work of Andrew Stern and Michael on Facade and other interactive narratives.
Posted by: William | 12/11/2003 at 05:11 PM
Few points based on the feedback so far...
Crawford is currently very interested in the whole "interactive storytelling" thing, although the stuff I've heard him talk about has very little to do with the sort of games we saw in the early days of CD-ROM. It might be more accurate to say that he is extremely interested in "interactivity" as an art, and trying to come up with new and better ways to interact with a scene or a narrative.
Crawford seems less concerned with AI than with the lack of focus on Character (in the Aristotlian sense) in game design. As he put it at one point today, "It's the difference between Shreck and Final Fantasy: The Spirit Within." He's more interested in putting emotion into games... not necessarily in a Turing test sense, but perhaps more in a "good screenplay" sense, which, with a healthy dose of good game design (i.e. limiting the player without them realizing/caring) can often come across as the same thing.
Finally to clarify for bowler, Crawford in fact advocated the simplification of things for gaming purposes, rather he wishes that games would try simplifying something more than the navigation of virtual space. I find this particularly interesting since Janet Murray names the navigation of virtual space as one of the greatest pleasures of interactive narrative. For Crawford, however, this has been taken too far to the extreme. He would prefer a more cinematic model, where all (or at least most) actions have purpose, as opposed to, say, trudging back and forth across an overworld map just to get to the next dungeon or other point of actual interest. Ideally, Crawford would like to see games with the potential to challenge us as human beings, although he acknowledges that such games may not come about until after he is dead.
And I would note that Crawford's views are certainly not necessarily my own.
To answer Andrew as myself:
I'm going to have to disagree with you on semantics here. Games should be fun. If you have made a game and it is not fun, then it is a lousy game. First and foremost, a game's purpose is nothing less than to make the most complex thing in the known universe (the human brain) and make it "happy." This is why game design is hard.
That said, you can have something that is interactive, and entertainment and it need not be a "game." Interactive narrative and game are not synonymous to me. So, if, as I suspsect, when you say games need not be fun you in fact mean that not all interactive entertainment need be "fun," then by all means, I agree. However, I'm afraid that something that seeks to be called a "game" but is not fun will only get degraded by yours truly.
Posted by: ClockworkGrue | 12/11/2003 at 06:46 PM
Could. Should. Would.
I wish Mr. Crawford would actually attempt to create what he is calling for. I also find it extremely arrogant to highly criticize modern games and then confess have not played them in five years. To eschew research and only offer criticisms makes him not unlike a fan boy in some message board.
Does he irk anyone else or am I a grumpy developer?
Posted by: NoGoodMonkey | 12/11/2003 at 07:10 PM
Could. Should. Would.
I wish Mr. Crawford would actually attempt to create what he is calling for. I also find it extremely arrogant to highly criticize modern games and then confess have not played them in five years. To eschew research and only offer criticisms makes him not unlike a fan boy in some message board.
Does he irk anyone else or am I a grumpy developer?
Posted by: NoGoodMonkey | 12/11/2003 at 07:13 PM
The thing that's so great about Crawford is that he sparks discussion. Even if his ideas are insane or whatever, he gets the less insane thinking and talking and responding.
People keep talking about how we aren't able to do a good enough job of simulating human relationships or emotional interaction with games yet. This may be true but I think that it's largely because we really haven't tried yet. Game designers and software people are bright folks and when you throw impossible problems at them they do a good job of hacking a reasonable solution.
I mean, modern computer games don't actually simulate physics or have actual real 3d objects or all kinds of things. But we do a good job of making it seem like the AI in Half-Life is coordinating its attacks or that the units in Shogun: Total War are maintaining coherence or that the guys in Close Combat: a Bridge too Far's morale is suffering.
Step 1 is convincing the nice people in the industry to throw their millions of dollars behind solving the problem.
Posted by: Snowmit | 12/11/2003 at 07:23 PM
Step 2 ?
Step 3 Bags of money!
-sigh- couldn't resist. Sorry. Although in this case, that might almost be true. Almost.
Posted by: ClockworkGrue | 12/11/2003 at 07:43 PM
Could. Should. Would.
I wish Mr. Crawford would actually attempt to create what he is calling for. I also find it extremely arrogant to highly criticize modern games and then confess have not played them in five years. To eschew research and only offer criticisms makes him not unlike a fan boy in some message board.
Does he irk anyone else or am I a grumpy developer?
Posted by: NoGoodMonkey | 12/11/2003 at 07:46 PM
Clockwork: "He would prefer a more cinematic model, where all (or at least most) actions have purpose, as opposed to, say, trudging back and forth across an overworld map just to get to the next dungeon or other point of actual interest. Ideally, Crawford would like to see games with the potential to challenge us as human beings, although he acknowledges that such games may not come about until after he is dead."
Cool. Then it sounds then like he wants the same thing all game developers want. Only current game developers know the limitations of the system they're designing for, and attempt to get "baby step" results towards the big picture with what they can afford in the bandwidth of the medium.
And really, in the end, achieving this interactive wonderland comes down to storage space, or the footprint of your game (because time and money are merely variables; the system you're developing for is a constant and fixed target). You can only fit so much on a DVD (or even multiple DVDs), and as Nintendo learned, on a cartridge. For example, one way to make the experience more "human" is to put real voice in the game, except that real voice puts a huge bite into your overall footprint. Just having type on screen is a good way to allow more interactive options (because it takes up way less space), but then you lose the vacade of the human element.
The funny thing is that the memory footprint issue will always be there (and we'll always be pushing against it, be it disk space or downloadable bandwidth), and at every leap forward people like Crawford will tell those of us who are pushing against it that we need to continue to push against it. For the time being, his level of interactivity is just impossible to achieve on the hardware we have to work with (obviously he realizes this).
The industry has been pushing for "more real" ever since Pong. Be it graphics, or better AI, more interactivity or even just better box art. People like Crawford will never be satisfied (and that's a good thing); with every new development that takes us closer to the next step to achieving this "goal," new paths and new horizons are revealed, prompting further exploration.
Posted by: Bowler | 12/11/2003 at 08:41 PM
I don't think the problem is about the story told by the videogame, it's more about the story you can tell after playing.
The story you can tell after playing Space Invaders: "Yeah there are something like 4 rows of monsters, and you have to shoot them all to collect points." It's not even a personal story!
In adventure games, there is usually a good background story, but it's the same for everyone. It's like telling about a movie or a book you read. (Altough of course I'm forgetting about more non-linear adventures - but then you simply have several parallel stories, don't you?).
I liked the Civilization games so much because I could describe (and I lived while playing) my last game as an entertaining war story. And it was my own. And it was new each time. What about NetHack? I'm not a nostalgic player, but try to tell a non-gamer friend that you had just become a Black Dragon thank to a wand of polymorph you had found in the gnome mine. You were lying eggs to have some tamed baby dragons to help you in your quest, but suddenly a gnome came in and zapped you with a wand of sleep. You had no protections since dragons can't carry any item so you fell asleep and was beaten to death by a leather golem. The common reaction is "what, is this really a game?" In this sense, the 'verbs' mentioned in the article are the ones you can build your story on.
It's not necessary for the game to explicitly tell a story, but it should supply enough material for your hidden storyteller to begin with. Wouldn't it be nice if a game AI (I'm thinking here about strategy games, my favorites) would take decisions aimed not primarily to the defeat of the player, but to maximize his narrative experience? I don't think it is necessary for the AIs to become emotional or particularly intelligent or whatever, just inspiring.
Posted by: PulverMuller | 12/12/2003 at 02:27 AM
Based on ClockworkGrue's clarifications, it doesn't seem like Crawford is calling for a post-game (his 'new discipline' or whatever) so much as better games. A noble cause, but he isn't helping things much. If such a thing is to come about, it will come about naturally, not by fiat.
I wonder if, on the advent of Gutenberg's printing press, a scholar somewhere said, "That's nice but I'll wait for Television."
Posted by: eli | 12/12/2003 at 06:29 AM
Two comments:
(1) "Counting verbs..." Well, okay. But something can be very simple and still be "deep." A reasonably smart man can learn to play chess in an afternoon and then spend the rest of his life perfecting his game.
Also, some of the best games ever have had limited "verbcabularies." How complicated was the user interface of M.U.L.E. or Ultima 4?
(2) Crawfords's point about SW Galaxies was spot-on, but this is a problem with "gaming" any literary or cinema property.
You can't "play as Luke Skywalker" because Luke Skywalker's story is already told; his origins, motives, secrets, suffering, losses and ultimate triumphs. To "play" the Skywalker story you'd either have to follow Lucas' canon step-by-step or deliberately change things (join Palpatine?) in which case YOU'RE NOT LUKE SKYWALKER.
The whole thing reminds me of a Dungeon Master I once knew, who was writing a campaign for very powerful player-characters to fight the Norse Gods.
"Just think," he told me, "With just a bit of luck, you could slay Odin and get great treasure and acclaim!"
"Are you fucking stupid?" I responded.
"Odin is fated to die at Ragnarok, from the fangs of the Fenris Wolf. You can't have a character kill him, I don't care HOW powerful they are. If they do, it's not Odin they killed - so what's the point?"
He started to say something else, forcing me to brain him with the slipcased edition of Advanced Squad Leader, which knocked him cold. Crisis averted. :)
Posted by: BrainFromArous | 12/12/2003 at 11:23 AM
However, I'm afraid that something that seeks to be called a "game" but is not fun will only get degraded by yours truly.
Which is precisely why it shouldn't be called a "game." Which is not to say that the form of a "game" isn't an appropriate mode of "interactive storytelling." I think music is a good analogy; pop music is valued because it is "fun" or enjoyable in a similar (maybe not identical) way that games are. But there are other forms or genres of music that are worth listening to, that are more challenging and less accessible. As a matter of fact, I think this is somewhat present in certain rpg's; I don't know if I'm really having "fun" while I'm playing Xenogears, for instance, but in the end it is still a rewarding experience, that's why I play it.
Now these might not be the best examples (and I recognize certain flaws and incrongruities in them) but I hope my point might be better elucidated. Let me be clear that I don't think games shouldn't be fun; but that they should be considered as a subset or a subgenre of what we call "interactive storytelling" or "interactive entertainment" (though I think these terms are inadequate; this is why I agree with Crawford that we need a better name than "games" because it limits and in some way cheapens the types of experiences one can have.) I wonder if these are his reasons for this also?
Posted by: Andrew | 12/12/2003 at 11:41 AM
NoGoodMonkey: "I wish Mr. Crawford would actually attempt to create what he is calling for."
He is attempting just that. He has taken several stabs at it, including the Erasmatron (at www.erasmatazz.com). He also mentioned that he's working on another tool as well, based on a linguistic interface in which players can input actions via a simple grammar of actions (verbs), subjects, objects, and modifiers. I can't say I was entirely convinced that this is the wave of the future, but he is definitely trying to work these things out in a real application sense.
Bowler: "The industry has been pushing for 'more real' ever since Pong."
Indeed it has, but the point is that "real" is sort of a misleading goal. What Crawford argues for is not "more real" but "more human," at least in an artistic sense. Cubism isn't "real," but it is human. Likewise, game designers are artists, and their human interpretations of reality are instantiated in their art (games). Crawford wants us to recognize and celebrate that.
This relates also to the idea of simplicity. All art is a simplification of our experience (on some level) for the purposes of expression. By simplifying bits of reality, we are able to place emphasis on certain elements, expressing ideas which would otherwise be lost in the mix of everyday life. It is through that simplification that an artist achieves expression. In the art of games, the designer has a number of points of impact in which to express herself. The designer necessarily must choose a simplified set of interactions. The set of interactions which is allowed to the player says something about what the game is about (this is Crawford's point about verbs). Likewise, the designer has control over what the player can interact with. Again these choices are artistic and expressive.
Crawford clearly prefers storytelling as the form of expression he would like to see in games, but I think that his ideas (if not he himself) allow for other kinds of expression without completely leaving the realm of games.
Posted by: Clubberjack | 12/12/2003 at 11:43 AM
addendum:
I think Silent Hill might be a better example, or even the entire survival horror genre. The entire concept of "gameplay" is subverted to enchance the experience: the core gameplay is clumsy and awkward, and the mechanics are somewhat shallow and forced (ie find key A, bring to door B). The puzzles are sometimes fun and fulfilling, but they are not the core of what the game is "about." As such, the game is not very "fun" to play; yet it was still fairly popular and garnered much critical praise.
So maybe survival horror games shouldn't be called "games", or, though you seem to resist this, we should rethink our core notion of a "game."
Posted by: Andrew | 12/12/2003 at 12:05 PM
i completely agree with the idea that some games should reflect humanity better. i would love for a push in this direction, though obviously not with all games (after all, how could you really put much emotion into a game like ChuChu Rocket, other than that derived from completing a level?).
think of this: when was the last time a video game made you cry? the last game i can think of that nearly brought me to tears was ICO (now THERE is a great example of the humanization of a video game, even without understandable dialog!)
the trick it seems is to get the right balance of gameplay and inspiring/emotional script.
-Titus
Posted by: TitusByronicus | 12/12/2003 at 02:13 PM
>> "The Sims, he says, is about making dinner, cleaning, showering, going to the bathroom, and going to work."
Those verbs only describe what the Sims themselves do, rather than what the player does. Among other things, Sim players build, watch, experiment, replay, create narratives and share them with other players.
Exploring gaming 'verbs' is certainly a simple and useful way to begin the critique. And I agree that (some sort of "true") interactivity is the Holy Grail - but a fair analysis of the *current* status of interactivity requires that we look beyond the surface actions of the characters on the screen.
Posted by: Jason | 12/12/2003 at 02:20 PM
Sorry to be late to the party, but I thought it was interesting that Greg Costikyan wrote on 11/9 about how maybe we need to move away from story. He then proceeds to break down Chess, which has none.
Posted by: crankyuser | 12/22/2003 at 10:54 AM