I feel like my own impulses are raping me.

[1.]

(hit the post cap. o)

For example you still present expressivity as a bunch of option with no internal structure we can only pray players will find their own. I want a system that clearly give the player designed incitive.

For example you say in Mario that's the number of node that present the player with options, he will express itself by going through it. We already know it doesnot work, because there is no pull, no reason to select a node over another. As it's a known design problem I know solution are to use landmark and clear functional gameplay affordance, we know how to structure this down to metrics...

Should've been more verbose .

No praying. What are you talking about?

Ok here are the pulls in Mario. Ok? They come in 3 flavours:
  1. Forced player action.
  2. Player intuition.
  3. "Narrative" guidance (I need better term).

1. Forced player action.

People are guided by the desire to go to the right. We can deconstruct that if you'd like... but later. Let's assume that players want to progress.

Ok, in order to progress they have to overcome a whole bunch of challenges. Each challenge creates a different kind of experience. The player has to run fast, or avoid simple moving enemies, then more fancy-moving enemies. He encounters the lakitu. He gets a starman and runs free. He's introduced to the ocean, and jellyfish. He meets that fish that gobbles you in one bite. He's forced to throw a shell. He gets pushed through a auto-panning level with falling platforms. You've played Mario right... that's why I picked it.

Each level produces an "experience" that has some abstract definition. Not every player will have the same experience. Not every player will have the same experience each time they play a level. Obvious; in fact, current mood, and play-length, and death count (etc) have a huge impact on how we the player experiences a level.

Let's say we put those definitions in a bag. We say that level 1 produces some experience that we've summarized nicely. I don't know: running, small fear of enemies, the freedom of jumping, having to time jumps, having to predict enemy paths. That description is technical. There is also the "emotional", "experiential," definition which just takes more work, but anyone here could come up with some half-decent ones.

Now we can assume that every player gets at least the "gist" of the stated experience per level. _That_ is their training. They are learning all the different things they can feel playing Mario, and are learning the relationship between various kinds of actions/situations and the feelings that they produce.

(Note: brilliant enemy design as far as design goes in Mario, for this, but whatever).

Just from that alone the player is learning a whole lot about what Mario makes the feel.

2. Player intuition.

Given number 1, the player is going to spontaneously have their own insights about how to play. Maybe it's more fun to run quickly through a segment? They don't have to, but they do anyway because it feels more free. They don't even think about it. It just happens. Boom. You've trained the player to express themselves.

Think about art school. You want to give projects and feedback so that students develop the ability to explore on their own. One step at a time.

After a player gains some experience doing things just because, and not because they are forced to, then they may start doing even more complex unnecessary things, because they've learned to explore. On it goes.

3. "Narrative" guidance.

The whole Mario world is constantly beckoning you to play Mario in an easy-going, explorative, way. The design of the enemies, the music, the way Mario moonwalks backwards down an escalator: it all re-enforces the experience the player should have. Shells make a particular sound. If you hold b when you land on an enemy you jump even higher, which is more fun, less controlled, and often provides more maneuverability. Mario swings his arms when he jumps, and the koopa "pops" out of his shell. The whole experience re-enforces the fun, exuberant nature of the world. Everytime the player sees a shell, they are subconsciously reminded of that experience, are slowly pulled towards experimenting with it. Levels train jump control... bouncing off enemies becomes even more interesting... etc.

4. Magic.

The real big-daddy is to get the system to understand, dynamically, what the player knows and doesn't know, then provide encouragement to guide him in the right direction, so that he stays engaged, and not frustrated. Have problems with particular jump control? Here's level to train. Getting tired of fast-pace? Here's something to slow you down. Not playing freely enough - too worried about danger? Here's section that will remedy that.

Living tutorial. Live coach. AI? Whatever. Yeah.

(Note, conversation is very close to this magic 4th).


Summary

Expressive world gives players a lot of options. You have to train them slowly to engage themselves in such a world. You can measure engagement dynamically if you want... but that's a rough topic.

Minecraft's brilliance is in the way it self adjusts, sort of like Flow (read: http://jenovachen.com/flowingames/thesis.htm by Jenova Chen). You have to make trips back your base on a schedule, until you have enough resources to stay out a long time (takes "leveling"). You run out of resources then have to re-stock. The game loop. Minecraft without nighttime (and many other things) would be nothing.

Navigation of caves, the dispersed encounter of particular minerals, the crafting tree: these are all laid out in a way so that players can find their own path through it, but are introduced to greater depth and complexity at a well-controlled rate. The player's creative freedom expands slowly with their "gaming" experience, which slowly stimulates them in the ways that inspire personal creation. That game's design folds in on itself again and again.

The player-traversal graph is more generated in Minecraft, than say in Mario. The player skill-growth path in Bejeweled is sort-of generated too. But it is still designed.

----

We'll be back to philosophy pretty soon. You'll see . But let's go any direction you like. Either I write about games here, or somewhere else.

Spawn a new thread if you'd like. I'm easy.

----

Taste of the future...

Conversation is pinnacle of: player expression, player freedom, dynamic story-line, "living" world, invisible tutorial (dynamic accessibility), [and - haven't covered - good AI, balancing, auto-testing] etc.

Deconstruction of human experience and expression is bread-and-butter of building conversation.

Defining human experience is philosophy. I don't make the rules, I just live by them. ... .


-----

@Paul.

Well said. It's all about what happens to the player. You can consider us (humans) equals, or not.

Dude. Guys. I'm sorry. And I'm tired. I just ran out of steam, not things to say.

....

I wish I could write that much. I can verbalize a lot about something in the right moment, but I can't put it in words that easily. These blocs of text do a great job on showing how passionate toast_trip and Gimmy are about game design. (note: i'm not saying that only those who write walls of texts are passionate about game design)  

It's been my experience that the ludicrous verbosity usually means there's a lot of overanalyzation going on, and a much simpler approach will help show what's really going on.  Especially when looking at older, simpler games .. zillions of pages are almost certianly a sign that the writer is seeing things that only exist in their imagination, or is just rambling and not making a lot of sense.

You guys are awesome. <3.

Over analysis is one of my biggest weaknesses. But we are talking about an actual something. I've been doing full-time work on dynamic story-lines, real AI, sandboxes that are also games (etc) full-time for years. I've gone through a lot of shit so I have to do as little as I can that isn't related somehow, and continue to not die. So forays into that realm come naturally. It's like poking a geyser. There's a lot of built-up pressure.

You should see how much I have written about Mario in my own personal notes.

I write every day about games, just as a part of work. So it becomes easier and easier. I find that writing on forums does a good job of enforcing a brevity that is unique to the experience.

When I was 19 I used to tear apart the combat system in Grandia II, trying to compare all of its strengths to any RPG combat system I was familiar with. It has a lot of weaknesses - I love FF, and a lot of others too - but it was so rich in a way that made me cry. Every so often I'd redefine my break-down of it, so I could best explain, even to myself, where all of its magic came from, how to make it better, and in particular, how to share the depth of that experience with other players. Like with Sacrifice, Grandia II's depth doesn't really come out and greet most people. People play it, move on, or never hear of it, even though they'd probably love it.

That always interested me. Shooters, and far more so, fighting games, have always had a depth that I have to struggle to appreciate. I'm a pretty good shooter player, like MW or Counter-Strike, when I'm in the zone, and, in particular, if I'm in the room with another semi-serious player (like the literal room, not the web), but on my own its tough to get in that zone. Why?

What is shooting? How can a player be trained in it? Without annoying them? Tutorials are fucking awful, almost universally. You can tear apart a particular game, but whatever abstraction you use to measure its qualities and assign meaning to the functions of its significant pieces (... whatever) aren't going to translate well to other games - say for comparison - or even to the same game after you've started making changes. How do you take an experience, talk about it, then ratchet it towards the more ideal experience you have in mind, then budget it, and ensure players will be able to get it? How do you do that without the expensive trial-and-error of just iterating the prototype? Developing ideas on paper, in theory, is much more flexible - several orders of magnitude - cheaper, and more collaborative.

You need an abstract language. I know graphs sound crazy. I spend a lot of time designing AIs that understand the game's design to a comparable degree as the designer, so that it can be interesting in all the ways that I want it to be - AI: other passion. When you get down to it, all AIs are graphs, and traversals of graphs. So that analogy feels very natural to me. (Behaviour trees: directed sub-graphs).

I can't get my Dad to play HL2, even though its on his good PC, and I'm a semi-pro life-long dev, and I sing its features at the dinner table. Why? Because he doesn't get it. There is such an enormous barrier between what regular people perceive and what they need to perceive to love games as much as we do. There are just as many barriers that prevent us from appreciating all genres, and so on.

Games like Tetris and Mario are really that deep. There's a reason there's only one Mario, or why even Nintendo has a hard time keeping Zelda as Zelda without breaking sales growth, or why Miyamoto hasn't launched another huge franchise in some time (best: Pikmin). But that guy could bang my Mom and I'd still have his picture on the wall. Mario is dead-simple on the surface, but insanely rich on the underside. That's what makes it so universally like-able and timeless. It draws you and keeps you there.

I'm not trying to pick a fight. I am a naturally very verbose person. I know a lot of this stuff is meaningless to a lot people who might read it, but it means a lot to me. Even when I'm writing I'm always thinking about how to skip to the end without missing anything critical. Normally I write to summarize ideas that I've been thinking about recently too, so it's partial conversation (this stuff), partial seat-work.

Again, apologies if it's like... "woah."

I wish I could write that much. I can verbalize a lot about something in the right moment, but I can't put it in words that easily. These blocs of text do a great job on showing how passionate toast_trip and Gimmy are about game design. (note: i'm not saying that only those who write walls of texts are passionate about game design)

passionate yeah but they should also be passionate about maintaining clarity yaknow
like i mean, see, when im being analytical i try to ask myself so many questions and when i ask myself so many questions and end up not answering them then that cuts a lot of bullshit id otherwise say!!

i mean look at this -- gimmy is proposing a distinction between two words that are, if he checked internet fucking dictionaries, fucking synonyms and toast is arguing that one can make choices outside of one's fucking head!

what the fuck man! i adore their passion but please more clarity.

Yeah, you have a point. I'm not intentionally trying to be vague. Everything I talk about is of critical relevance to the direction of the conversation - at least in my head. All that definition mumbo-jumbo and whatever is isolating the best way to organize the fundamental constructs in game design, from each relevant angle (to the discussion). I pick apart the small things that seem to be causing confusion. When you have a rigid structure you can build up to something stronger. You have to be very careful when you're analyzing something that is inherently abstract. Point: the definition of God is the start of more wars than any other single thing. Clarity of detail is important. Games aren't religion; I'm just using a simpler example for exaggeration.

Think about the periodic table. A correct detail there can unlock huge sections of chemistry, chemical engineering, biology, medicine, etc. "Choice" and "real choice" and "meaningful choice" and "enforced goal" and "stimulated goal" are like our elements.

If I were to be any simpler I'd miss necessary points. If I were any clearer my posts would be even longer, and likely more random. I already find Gimmy's posts too short. I have to mine through them and extrapolate meaning. Though that may (largely) be a function of his writing style.

If you have a question, like something you disagree with, or something that doesn't seem relevant, or anything like that, just write down whatever you are thinking - whatever - and I'll clarify. I don't want to be too verbose without prompting. So if you, or anybody, ask(s), I'll tell.


Partial aside: http://www.youtube.com/watch?v=OGSeLSmOALU
That's a talk by Jon Blow. In that one he uses a lot of "technical" language to deconstruct games. His language a lot of the time is similar to some of the language I've been using here. He makes references to Braid, obviously, and makes a lot of really great observations. If anyone's got an hour, I recommend it. It might make me seem less crazy.

@toast
I think the main difference with you is that I have learn to look at thing more dryer and cut the philosophical bullshit as fast as possible. Your post don't explain in a dry mathematical way how things relate, you explain vague term with another vague term. For example you link conversation to relating, but what's the mechanics of relating?
I seek the abstract "number" behind loose concept, the math, the structure and process. I'm seeking hard design you have a softer approach, still mixed with emotion. I leave emotion for much later, when I dream the game and broke the hard rules to get where I want.

Ok, I thought I was losing you a little bit.

I'm a programmer. So all of my definitions I've given you have a 1's-and-0's interpretation. Everything I said is programmable: raw math, rigid structure.

Ok, so here's the mistake a lot of designers make. Game design, as an entity, is about the fusing of abstraction and logic. That's what it is. That's why I put a lot of effort into that clarification. Starting with a technical basis won't get you anywhere. That's what most theorists do.

Sometimes I feel like there's this desire to always stay in a safe place, where everything is certain, and then only tackle the tough concepts when the theory is ready. But that is like driving to New York with your eyes closed, and when you don't even know where New York is. This is why I like design.

But, with that said, I will take any approach you are comfortable with. I can take as a hard a line as you'd like. You just have to keep pointing out where my definitions seem unclear. (We can keep a tow-line on firm logic forever if you'd like).

You asked me for a starting point right? I didn't know how not to be vague. Conversation is about an expression and perception of mental state. I'll give you an example.

Let's say I want to communicate something simple to you: I am angry. That's it. Conversations have way more nuance than that, and have a back-and-forth. But I'll show you that even communicating just anger, especially in a game, is very complex.

Anger itself is an abstract concept. There is no mathematical function to represent anger. There just isn't one. I don't know if that is obviously true or not. Anger is a feeling that humans universally can relate on, and share a language for, but there is no provable way to so that my anger is like your anger, or that I even understand your anger when you explain it to me. You cannot, ever prove that your NPC or whatever is communicating anger. You just can't do it. (You can do some stuff with brain scans, but that's even less accurate than an honest expression. Someday, far in the future, when everything is different, then we'll be able to).

Humans are more than just emotions. We are also ideas. But for simplicity let's assume that once we can provide the functions for communicating anger we can communicate a wide range of stuff.

Ok. Anger is a conflict. It's the natural response to stimulus. We get angry when something doesn't make sense. Let's say when we are in danger, or when someone cuts us off, or betrays us - particularly if they're a loser. We also get angry at our parents for not respecting us when we grow up, or not being treated fairly, or having our territories threatened. In animals you see anger when they fight over females, or authority if they exist in a social group (like wolves).

Think about all the times you've been angry? Can you draw a definition out of it? Let's say we want our player to be able to express anger, naturally, through play, without having to select an option, then have an AI pick up on it. (Conversation will come later - more complex).

Mario expresses what? Joy? Exuberance? When I play Poker I can detect a lot through the subtleties of another players face. A lot of people can do that. That's a universal skill, so there is so constant property to identify.

Mario players for example can demonstrate "frustration" (close to anger) by becoming impatient - i.e. going faster through parts they otherwise wouldn't, leaping before they look, taking shortcuts that are unnecessarily dangerous - becoming violent (killing enemies just to kill them), and playing with more energy. None of those particular qualities on their own indicate frustration. Some are dependent on the person. Maybe you're watching a friend play and you pick-up on nuances, and make inferences based on them, that someone else wouldn't be able to notice. Violence in the game isn't always a reflection of frustration. Some players are naturally more violent. Since violence in the game can also be seen as more joyful, or part of some sub-goal (say to kill difficult enemies with style while running through), you can't even consider it to be a "tick-box" towards declaring the player frustrated or not.

Conversation, interpretation, communication, is all about making soft-references between abstractly defined ideas. That's what humans are excellent at. Relative to the rest of the non-living universe, we are extremely abstract things.

You need a consistent way of relating ideas in your game, and you need to know how any combination of actions translates to a mental state (i.e. feeling etc), _then_ you can interpret what a player expresses and guide it. True conversational systems require a deconstruction of the human experience itself. That's why games seem so, so far away from it. They really are that far away.


Ok... that's enough for conversation. I hope you understand what I mean about the necessity for defining abstract terms. Tell me where you're at with that one and we'll continue. I don't want to just run around randomly in the dark.

Note: I've been deconstructing emotions, such as anger, for a long time, so that I have a running "database" of "frameworks" for human experience, and the minimalist representations for their expression. That is an endless rabbit hole, I promise you.

Mmm. The last one I redid was "struggle." Normally I pull together a lot of examples from the media and go at it, looking for the similarities. The more I do, the easier it gets.

Mario is cool because players will cover a relatively large range of a level's mechanical possibility space when playing it (if you want, you can represent a level's possibility space as a weighted directed cyclic graph where each node is a choice and the weight represents probability of traversal to a given node.) This kind of "expression" is facilitated by both aesthetic elements such as character animation and the level design itself (i.e. the way the character interacts with everything else, "expression," etc.). This is an approach which other games such as Prince of Persia also take [and also every half-decent computer, console, or phone videogame ever made -- nvm, decent phone videogames don't exist]. To encourage "expression," the game can slowly replace concrete and obvious goals with goals designed by the player (e.g. Minecraft making you want to mine, then you discover that you can build things, then you get more familiar with elements so you can build different things, etc.), as well as storing a "permanent record" of what happened when you played that you can easily share with others [aka what tons of decent computer programs have done for years, the difference being that my videos don't go viral on Youtube and I can build much cooler stuff in other programs].

Sorry, I kind of lost motivation to go on around here (right after the part where he says "Yeah.") but a 12:1 ratio isn't that bad I guess.

That's not bad. Maybe you should do all of my translations.

I'll make one correction. The Minecraft stuff. Games haven't been doing that for years. Minecraft is immediately relatable; most games are not. Minecraft is clear in a way that goes far beyond most "construction" games. As a clue, think about how every decision is literally represented by a cube. Also, the construction process is naturally managed by the mechanics so even normally un-creative people have a very easy time expressing themselves. I mean independently creative people. You may be creative but lack the understanding to express it on your own. Minecraft changes that, and makes it easy to share the experience. It is a lot easier for a novice to create/share a Minecraft adventure than one with SimCity or an instrument.

Successes are rarely flukes.

Yeah, but he instills them with his influence. He's still an authority figure, thank God.

Woah. I didn't realize it was offensive. I'm sorry.

Yeah, Gimmy takes some practice.

Mario is fucking brilliant. I could write a lot more, no prob.

My questions were mostly answered. Gimmy and I have been moving forward, I think. That's how I feel. I know there were some fast subject switches, but I can say that the ones I was a part of were generally the result of implied conclusions.

Now we're talking about how to design a game who's core engagement is in the player expressing themselves, and a game that supports "sandbox" conversations, for example, if Mass Effect was Mass Effect but offered the kind of personal freedom to do as you please, like how it is in Minecraft: real dynamic story lines.

The choice stuff launched out of discussing the difference between superficial choices and meaningful ones: how to distinguish them and turn the former into the latter. That came from eva decrying choices that aren't really choices (dismissal of sandbox etc).

Gimmy also talked about some ways to deconstruct gameplay experiences into parts that you can study.

I didn't know people were listening in. If anything is interesting to anyone, I'll always explain something that doesn't make sense, or seems pointless - of course, if you care.

It is a little insane.

---
edit: Love Spelunky: pickup-and-play, ramping difficulty, proc-gen, personal touch.

Issac. I haven't played it yet. The game loop is more like thematically relevant - an exercise in minimalist design I think. McMillan said he churned out that game in a heart-beat relative to Meat Boy, and it's been selling at a comparable pace (though still less). Fast gameplay, replayable; yeah, it suits.

It's not personal. Dude just rages.

I've been an asshole many times too.

Windows: motivation for Linux.

Ouya: motivation for Pi-Box? Failures matter.

ok i skimmed the other huge posts and realized theyre talkin about nothing so i'll jsut sit back n not read them  

Sorry: hijack.

Gimmy.

Sometimes its hard to piece through your shit. Reading you is like lesson in game design.

Yeah, I thought something was up with Bejeweled. Hidden depth, obvious depth: I'm easy either way.

You're saying that "expression" is a superfluous concept? In general? I don't think it's terribly difficult to design for, any more so than any other game thing. I think in general all experiences are hard to design for. Expression, you could say, is one that requires design theory to execute on a little more, sort-of, largely because it's not based on conflict/challenge/etc. The theory to support it is just as weak as it is everywhere else; it just so happens that trial-and-error is less productive in that arena than in others: namely, only challenge.

Ok, I'll bust out some examples. Minecraft, Mario, Prince of Persia: all expressive. Minecraft expression is obvious... so obvious. It's obviousness is a testament to its brilliance, success etc. Mario is more subtle, and insane. Mario is very expressive, very much so, and in a way that's obvious to a spectator (that cares). There exists a "field" - shall we say... blergh... - of possible options that a player can exercise in-sequence in any given level in Mario. Maybe every level begins with a point in an n-dimensional graph (abstractly), then "moves" between nodes as the player makes decisions. Each movement represents some decision, however nuanced. The distance between nodes represents the likelihood that a player will take that path, or something.

Through a normal play-through - assuming there is one graph to represent all levels - the player will traverse a large section of the graph, obviously a minority, and generate a "cloud" of points that he traveled to. Say: nodes traveled to more often shine brighter - the rest are turned off.

Ok. The brilliance of Mario is that through normal play-through, any player will create a well dispersed cloud within the graph, likely covering the range of Mario's expressive properties. Also, the animations accentuate the expressive function of each action. The player is introduced to Mario's expressive properties cleanly.

What others things the player does is up to him. Given the basic constructs of play - running fast; taking jumps sporadically, or in panick; sloppily landing jumps; unnecessary twitch maneuvering; unnecessary danger flirtation; purely-tactile maneuvers (killing enemies that don't need to be killed, breaking non-critical blocks) - the player can generate fairly complex tapestries of personal experience. They are introduced to these constructs through the games challenges. Ala... I love Mario; goddamn.

Miyamoto iterated Mario crazy-time. He started with flexible mechanics, then brainstormed a new "experience" to add, simplified it to its core function, threw it into the game with animation/appearance that reflects its function, then repeated.

He says (paraphrase): "I don't think too hard about how it all fits together. I just think about the players, and how they will react to the game. I want to see them engaged and enjoying themselves." Maybe design theory is what you use in conjunction with iteration, and experience. All screenplays are iterative, either in the creator`s mind or on the page. That's creativity. If it were predictable it would be a science.

Prince of Persia: same thing, but with even less choice. Jordan Mechner tricks you into thinking you're a parkour ninja just with well-placed animation, then carefully tying how those animations are activated relative to the level design: woaoh.

Mario expressiveness with Minecraft decision-flexibility and permanence. Hmm....

Best solution to encourage expressiveness: slowly transfer from challenges that introduce the player to expressive concepts, to a state where the player is driven by his own expression. Minecraft for example tempts you to mine. Then through the mining you encounter the constructs of the expressive aspect. You are stimulated to build. You build, then you mine. Each time you build you are stimulated to decide what else you'd like to build, what special item you want; so you mine. ... Then you mine more effectively... build armor, create better defenses, storage... blah. I think a game without some kind of sub-textually or overtly enforced goal-setting, at least to start, is a bad idea; that is a toy, or an instrument, or a pen-and-paper.

Only when you can assume naturally expressive nature of all humanity can purely expressive game really be game.

Minecraft: game + level editor carefully disguised as single experience. The level editor is where the power is, and the "game" is the tutorial.

Also, "smart depth." Minecraft is an example, even much better, example of this. In GDC talk, guy talks about adding "smart depth" above existing depth, or the core gameplay. Minecraft adds the "smart" layer directly into the existing experience. There is no distinction between regular play and creative play. Doing something for mechanical reasons - i.e. protection - is just a personal as everything else (proportional to time spent doing it, sort-of).

This direct integration, of the personally expressive layer (i.e. "smart depth"), in Minecraft, makes it extremely viral - super bonus points. An in-game representation of my play experience is built naturally, is easily understood by the player, and by others. It is very difficult to express the experience of playing Mario, not the real experience. Mario happens in your head, personally. The avatar is just a guy bouncing on the screen. The graphics have short-term memory loss, as does the rest. But in Minecraft, you usually have an actual symbolic record of what you went through mentally. You can then share that experience without committing to exploring what really happened to you while you went through it. Minecraft is like a "talk-about-what-happened-to-you" conversational ice-breaker, all the time. Hence youtube, it's popularity without a friggin manual - despite huge play complexity. Genius, genius game.

Interactivity is about action, the player decide to act, act is the atom of interactivity. But to act there must be a drive or a motive, that's the hidden blood of interactivity, without drive there is no meaningful action. Object framework rely on identification to goals to provide drive, action make sense because they lead to goal, the goal define a problem to solve. Now in an expressive system how do we provide drive or motive to act? Actual pure sandbox game left a lot of player bewildered and confused, only creative mind find solace into those game, but they merely bring in the game something that isn't there, their own motive and drive.

Yeah.

I have primitive elements of answer is some game, mostly in management game or affiliate, they rely on need to fulfill, need is a reversed goal, basically you set goal to fulfill a need, so need might one of the answer, what's the theoretical game framework on need that create drive?

Can you expand here?

What is affiliate game? You mean like Farmville, or some other hardcore example, where you profit from a growing economy, but only play an indirect role in its growth (SimCity)?

I think I can say some stuff here, but you'll have to give me an example that clarifies what you mean by the framework on the need that creates drive.

I can take a weak stab here. ... wait maybe I get it.

You want a way to set up goals in a game with total flexibility. You understand how to do so in say Sims: you build a house in a particular area to make a person happy, or something. So you're triggering the goal setting. Will Wright is another great guy.

There is no direct way. Even Sims is dangerous. It is only successful because people already have an intuitive understanding of a person's needs. The game simplifies the need, and its relationship with the goal-progress that can satisfy it, but is still very relatable. The Sims is on the surface a very casual game - i.e. immediately relatable - but with richer mechanics; hence its popularity/respect. 

So resonating with an audience right away is an easy way out; Guitar Hero is an example. But you don't have to do that. People want to play with cuddly things, or watch gumballs move through a Rube-Goldberg or watch Mario jump and exclaim things. Stories begin with a hook, enough to motivate the audience to invest for a little bit, then leverage that commitment to introduce more significant elements, which ideally create more interest and investment, then repeat... slowly moving to larger concepts.

Expression is really an end-state for a game. Minecraft is a product of the viral age. It wouldn't work pre-youtube (or twitter etc). The first tutorial - investment grabber - for it is another person. The whole wiki is fan-made, the mods etc. So you could just create an expressive game and rely on communities to drive up player interest. Musical instruments are entirely this, or any deep hobby.

I would think like this:
  1. What do you want the player to express?
  2. What are the fundamental (orthogonal) constructs (i.e. player actions) that allow such expression.
  3. Create a game progress path that slowly instructs the player in the use of such constructs.

The important thing is to loop between the controlled gameplay and the expression. Minecraft for example gives very limited freedom in construction at the beginning. You have to get materials to build. You need to mine to get materials. Your first trips out for materials are very short; you don't even need to go underground the first several times. Slowly your adventuring trips deepen, and your creative expression options expand. The player follows this loop:
  1. Game.
  2. Express.
  3. Gain slight freedom in both.
  4. Repeat.

Minecraft is like to that loop (the expressive one) as Tetris is to the spatial geometry implementation of the core gameplay challenge loop.

Mmm. Smash Bros: very expressive. Smash Bros single-player: train in necessary skills to become expressive. Biggest weakness of Smash Bros: single-player only train very rudimentary expressiveness relative to that required in successful multiplayer. The same weakness is predominant in nearly all deep multiplayer competitive games (things like Civ don't count because most of the depth comes from your independent decisions - i.e. single player ones - not from interacting with a human opponent).

Another way is the push and pull mechanics in level design, landmark pull player toward a place, narrative mystery might be another pull, the need for closure drive for more. In minecraft there is a clever push mechanics, basically most player are lost until they meet monster at night which push the need to create a shelter, but once the shelter is done, even the less creative player try to ameliorate it in a way that isn't just optimization but actually they start "decorate" it, ie building it in a way that fit their preference, something creative player just do naturally.

Minecraft problem: stop pushing hard after first night: all downhill from there.

Push-pull: training wheels for player-driven experience. Goal: player-driven experience. Even though FF gives me a narrative the whole way through, it's more my personal investment that drives me over time than the desire for the ending - beginnings always better than endings.

The first leading clue there is that for a conversation there must be an imbalance, the nature of this imbalance is to be generalize for every complex case. What is the purpose of conversation (structurally)? how does it start, when does it end? We only know it depend on and reveal who the participant are, their very self at the moment of conversation.

I can answer this with better examples.

The purpose of conversation is to relate. Humans are information presentation machines. Reality is complex and hard to interpret. Humans interpret life all the time, compress it, then relate it in a way that we all train to understand and produce idea in our whole lives.

Interacting with a human has all the same pulls as doing anything. People have one drawback: they have their own goals, which must be respected or the interaction will end, sour (turn destructive), or damage potential for future interactions. The obvious benefit: humans can present far more abstract ideas (i.e. more directly relatable), and can personalize their presentation - something that happens naturally.

Conversation starts when one agent intentionally presents internal state (mental state) for the purpose of communicating. It ends either when one agent shuts down perception of the other, or both shutdown perception and attempts to communicate.

What do you mean by imbalance? You mean the conflict between people, or within themselves, that produces a goal that drives interaction?

So, conversation is three things (from the perspective of one actor):
  1. An intentional expression of ideas (ala everything from before).
  2. A perception of another agent (say, as a distinct entity).
  3. An ability to interpret the expression of the other agent.
Conversation needs 2+ agents.

Build expressive constructs, teach constructs to player (gradually), create agents with "personalities" (i.e. set of reactions to stimulis), have agents express mental state in same way player expresses his, have agents interpret player expression and react accordingly (respective to their defined personality).

Creating compelling conversation: make agents like people, but interesting - all agent goals must exist within game (ok if "hypothetical" "super-"goal implied by collection or real sub-goals); (optional) have "narrative" AI control world events to react-to and funnel dynamic story in the direction of a modeled plot development "chart"
(huge multi-dimensional graph).

You can have the narrative AI fuck with the NPC personalities, biasing their random elements of expression to suit its goals, but that's dirty; be very careful with that; it will most likely fuck with you in a hard to describe way.

I fucking love space.

Never enough, never enough.

It is cute. The picture definitely adds something.

Thank you very much. I will read all of it.

. The behaviour trees. AI is my forte. But thank you anyway. I read all AI literature anyway, always. For perspective, and to be a jerk: I'm currently build a poker AI, which will play the game very well, and will be able construct training schedules for me to play as well as it, optimally. That's where I'm going to get my game funding from.

You talked about, in another post, predictive complexity, like me predicting an opponent's moves in a fighting game. Poker is all that; mechanics are really an excuse to predict. Everything else is about being one move head. That's the connection between the Poker AI and the game one (i.e. "conversational") - also, both are very "soft" logic i.e. intuitive.

I'll post my thoughts on the articles once I've consumed them.

In order to clear any confusion that might exist: I propose theory and analysis from a designer perspective which consider game structure first aka making informed design decision. They are absolute abstraction, concept of the mind to further manipulate reality, they are not reality themselves, merely signpost. They are just a generalized distillation of current set of design tools. Semantic circling defeat the very purpose of these tools and do not advance the conversation.

No, that's not true. All of philosophy is one giant semantic discussion. Religion, you could say, is one giant discussion about defining God, or the purpose of life.

Theory, because of its malleability, is far more valuable when you have a clear idea of what abstraction you're actually using. More importantly, it fuels communication. The more abstract a concept, the more powerful, however the more likely it is to be misinterpreted when communicated, even to yourself (say from reading old notes, or recalling a thought). So it follows, the more powerful the idea, the greater the importance of semantic clarity.

All of programming is a fight for semantic clarity (technical expression of the solution to a problem). Game development is part art - the inherently non-definable - so game design houses a tough conflict: the need for structure of programming meets the need for the expressiveness of art. Design is in the middle. It needs to tie both structure and the totally abstract together as best as it can, so we can fit ideas together on paper, without relying on trial-and-error.

We already agree, but the terms divide us, hence the importance of semantic clarification (Now I'm being a dick).

------

I don't think it's fair to say Bejeweled has no depth. I understand where you are coming from, but I will give you my reasoning, and why I think it is important.

The choices a player makes in Bejeweled aren't directly expressed by the player in the game. The game has a limited sense of the player's mental state, say relative to what a player can accomplish in Starcraft. But the design still absorbs them.

This is what I mean. Bejeweled is interesting because there are a lot of mental tricks that a person can enjoy mastering to get a high score. Bejelewed was designed with this tricks in mind. The designers wanted something addictive, so they studied other games that were addictive, partially mapped out how the player's experience developed through the course of their relationship with the game, then made acute design decisions that were more likely to lead players to what they considered to be the ideal experience.

You can study Bejeweled and extrapolate the designed player experience. There is an implicit game that is hidden beneath a lot of stuff. Minimalist games often have this quality. Tetris does for example. Mario is a goldmine. It has as much unsaid as Bejewelled (and obviously much more).

Right? Maybe we need a new term. There is a distinction between the implicit depth of Bejeweled and the more apparent depth of a fighting game. But there is also a distinction between the implicit depth of Bejeweled and the non-existent depth of some non-successful game that I never played. A lot of players experience some kind of depth when playing Bejeweled, hence it's success. In fact, the fact that it hides its depth makes it more approachable, and is a significant ingredient to its casual-market success.

hidden-depth, obvious-depth, game-reflected-depth. Whatever. You know what I mean.

As an aside, my current game tackles the "true conversation system" problem, with all the stops removed. And I don't mean that you'll probe me and realize that it's ideology and observation, as conversation systems generally are - I'd also say that they are expression. It has a real conversation system, though obviously more limited than actual human language, by many degrees, but infinitely more expressive than any existing game allows.

So, obviously, I have to dig into theory all the time.

Trial-and-error is terrible: slo-mo design.

What do you mean by Subject frameworks?

You could begin with anything. I have a lot to say on that topic. I'll think up some...
 . the basic constructs of an interesting character (i.e. the bare minimum necessary to be compelling)
 . the basic constructs of expression: the bare minimum to express a "person's" state (mental/physical/"spiritual" (redefined mental)), conceptually
 . the basic constructs for representation: the bare minimum to express any idea to a player, with sound, graphics, motion, etc
 . the basic constructs of an interesting story
 . how actions reflect opinion
 . how actions impact the state of other actors (agents/characters/whatever)
 . the relationship between the basic constructs of game mechanical(/dynamical) systems and their corresponding "narrative interpretation" (i.e. what it means for Mario's character when he makes all the decisions that he does, or maybe in better terms, what it means is happening inside a player's head based on what he/she is doing with the character - this can require heavy interpretation by the system of the player's play)

Generally I strip games (video/otherwise) for parts, and stories (real/movie/book/otherwise) for parts. I keep going until I have enough parts that I think I can construct basic games that I'd like to make, test them out with mock designs, then proceed. On and on.

I've been doing this for years.

(now, I'm not sure if I totally understand your position)

Superstition doesn't matter. If a choice is meaningful to a player, it is meaningful to a player. No one can say that a choice is "true" or "not true." That's why I gave the War example. If kids enjoy Monopoly, then they enjoy Monopoly. That's it. If they learn about their own misguided perceptions from playing Monopoly, then even better.

All that matters is that a player finds a choice meaningful. You can use truth as a way to help guide you to design choices if you'd like, but it doesn't invalidate a choice inherently; what matters is the player. You can say that teaching a player truth is valuable, so superstitious choices are interesting until the truth of their nature becomes revealed. That would provide a good reward - learning - and thus make the choices that lead to it more interesting.

ok so choice is everywhere and when playing games it's undesirable to make choices outside of one's head? lol

we all know that video games exist as we experience them! (John Searle would say they exist in a metaphysically subjective way)

It's not undesirable to make choices outside of one's head. The location of a choice is irrelevant to its value (inherently), where it's reflected etc.

What matters is that the choice is made. Understanding what choice is being made is often a very good thing. Mass Effect, for example, like a lot of games, confused what choices were really being made by applying a superficial understanding of what was happening inside the player's head. It took a page from action games and one from RPGs, then mixed the choices from both without first understanding what they really were. Thus a conflict was created. There were several competing choices occurring simultaneously, that didn't support one another, and were individually isolated from much of the game, as a result of the compromise.

I love every genre that I can recognize as one.

I love games that give me a very clear sense of progress. I don't care how I progress, I just want to do it. RPGs mark transitions with leveling and plot; fighting games with skill (over opponents); strategy... with skill (over opponents); platformers with the ease that I can express my skill, even in a single level.

I want to start small, then grow, and grow, and end up somewhere definitively beyond where I started. I want the game to show me that progression, even if its just by searing it into my brain, making me remember on my own.

Movies always "end up" somewhere. Same thing with books, other linears. But they don't reflect where you went. They can signal it, but its up to you to interpret that. Games have always been in my head - the reason I put them on a pedestal - as things through the playing of that I became different than what I was when I started. Provable growth. I love it.

Most of the time I don't like a game is because I don't understand how to progress steadily, or stay aware of my progression, or stay interested in my progression (etc).

I just do work. The project is too big to sit in my mind. The biggest strength to doing a good job is having a strong instinct. The best way to develop your instinct is to do work you believe in, and make all kinds of mistakes.

I plan and tie ideas together and do formal reviews and create filing systems and draft comparisons. I listen to music, and run, and sit under my desk and cry - or not cry, and sleep. Whatever.

My magic formula:
  1. Be critical about the quality of your output.
  2. Do work (i.e. build experience).
  3. Follow your instincts (i.e. do the work you believe in).

I tend to work on a project from many angles simultaneously, and occasionally stop to judge how they might fit together, then carry on. Sometimes it feels like I'm drilling from two sides of a mountain simultaneously, using my good judgment to tell if the tunnels are on-track to meet in the middle (a ancient-Roman architect did this, through a mile-wide mountain, and was off by 3 inches). I have tunnels branching from tunnels, branching from tunnels. Instinct tells me they're on-track, and if it doesn't, then I do some calculations until it does.

The worst things are not doing your best work (i.e not working), or not doing the work you believe in. Everything else is personal (i.e. any system is secondary).

I plan accordingly.

. It's not bad right.

"resonance" they call it.

Miyamoto says it's his new game design grail (i.e. over the last 5 years).

Holy shit, a five kilometer long back and forth between gimmy and toast. What conclusions will they reach??? Let's not find out.

Hahahah. Well, I don't hide my nature. Theory turns me on.


@Gimmy,

In my head I've always had: an option is a single path towards a goal, a choice is a collection of options (i.e. a single one). For example, I have 3 options that make up a single choice, or in the same example, 3 choices that make up a single choice. I can say, "Benny, you have a tough choice to make; here are your options..." or "Benny, you have 3 choices...."

You mean like:
  . option: walking or driving to the theater
  . choice: driving to the theater or driving to the amusement park.

The difference is a matter of perspective. There is an implied goal given for the "choice" - theater or amusement park - which is: to "have a good time" or something like that, in which case the choice becomes an option by your definition i.e. going to the theater or the amusement part to have a good time.

I feel like there is value in the distinction though.

----

I'm going to spin off about the "execution only" strategies, or effort. Track and field, for Olympians, is riddled with choices. They have to get to a very particular mental place that gives them the edge that they need, and adapt to the changing situation, their mood and the state of their body. To a spectator, their process may seem mechanical, but if those observations were true then the competition wouldn't be interesting to the competitors or the fans.

I like to think about how interesting swinging a Samurai sword in the perfect way is interesting, because I like Japan, and anime, and all the stuff. High-performance competitors have to navigate a rich world that is their minds every time they perform. They must always find the path between their current (mental) state and the goal (mental) state - which itself has a definition that is in a constant development. As the game changes and the performer's emotions change, he/she has to find new ways to navigate around inside his head, often getting into positions that are in some significant way novel to him. Spectators like seeing how the athletes (or whatever) will "make do" this time.

Straight execution can become very deep if the player is motivated, and has some way to relate his decisions to results. The decisions are made in the head, not reflected directly through action, but some nuance is represented through behaviour.

----

In your break-down, the category "potential" I'd call something like "impact."

----

I read a blog post that was linked in your recent Zelda thread, which was about the author's personal opinion about the decline of the series (well, partial decline). There is a similar point there to the one you made about Mass Effect, in which the player is given powers that have one-situation-only uses. I'm reminded of a doll my sister had when we were young: Ernie from Sesame Street. He had a button to button, a zipper to zipper, a shoe-lace to tie, a snap to snap. Each was independent. The joy came from the tactile experience. The ME example is like that doll.

Zelda, as the blog poster said, used to be about discovering within a world, or the player's skill set, or the mechanics of a screen, then became about finding the right key to put in the right hole, pulled from the player's ever-growing key chain. There is some joy in doing that - tactile - but it wears thin. What it does do is take the discovery and "choice-making" of the original titles and make them obvious to the player, though removing much of its depth; that's why it was done.

----

A "pure" sandbox can have depth. Goal setting is inherent to the human mind. Decision is all about comparing possible futures, which allows us to carry on living (and reproducing). Life is a sandbox. The technical division lies in what counts as part of the game.

If I play in sandbox, technically, then the sandbox gives me no goals: agreed. However, I might create my own games. In fact, I have to in-order to grow. I'm making choices because I'm human. Without choice I'm not doing anything, in which case I'm not experiencing the sandbox at all.

So the sandbox doesn't enforce a choice; but no game enforces a choice. Mass Effect may give me a story choice, but whether I perceive that as a choice is not a guarantee. Taking it to the extreme, if I don't understand what's happening, say because I'm a monkey (and pressing buttons randomly), then my action isn't a choice connected to the "choice" the game presented to me.

The choice that a player makes and the "choice" that a game provides are always distinct things. They share an non-definable relationship. You can in no way prove that a game's mechanics gives any given player a choice. All you can make are general statements.

So, I can play a sandbox and be inspired to set my own goals, maybe subconsciously, then make choices towards those goals. Or I can play Mass Effect and make no choices at all, or make choices that are partially related to what I'm given, or the more likely case: make choices that are heavily related to what I'm given, but still entirely independent from other players' choices in some non-trivial way.

Saying that a game "definitively" creates choice for the player or not is an arbitrary distinction. Sandbox doesn't cross a boundary where "no choice" is enforced on the player. No choice is ever enforced. All we have are games that sort-of, generally, create choices that are "something like this." In the end all that matters is whether the player encounters a choice.

But I agree with the sentiment, and nearly everything. I'm circling semantics, which are by definition perspective-based (and thus provide no right answer). But the conversation is useful.

I break down games in the way that you have all the time too. I have a big list. It's something I always want to see more of.