So Andy's actually had to take out the big Moderator Mallet on Story Games for the thread Games and Gaming: Ready made out of the box?. The thread is on 24 hour hold and he suggests that the participants take the definitional question out into the blogosphere. Sounds good to me. Here's my take:
"Out of the Box" is a meaningless term that is based on a faulty assumption -- that playing a roleplaying game consists of following procedures outlined in a book and no others. A game playable "Out of the Box" ostensibly has everything you need to play already included. This is an impossibility. Roleplaying games, by their natures, require player investment. Player investment, in turn, requires the players to have something to invest. That something always comes from outside of the box.
Example the First:
Candyland can be played out of the box. Here's your playing piece, here's the board, here's a deck of cards. A computer could play this game. When real people play this game, they are doing one of two things: (a) developing their color-recognition skills, assuming they are small children, or (b) socializing with the other players and using the board game as a pretext for interaction. This is the stupid-simple case, but Monopoly works the same way. Ever played Monopoly against a computer opponent? Using just the published rules and the tiniest bit of code for bidding on properties, a computer opponent is just as good as a flesh-and-blood person. The decisions that players make do not require any real-world experience or knowledge of real estate or even basic strategy; nothing from outside the box is required by the operation of the game.
Example the Second:
A roleplaying game -- let's say Capes -- cannot be played outside the box. Click-and-locks, index cards, dice, and tokens are insignificant markers of significant elements of the fiction. Those elements of the fiction are provided by the players. Players must bring information, knowledge, experience, and judgement from outside the "box" into the game or else nothing can happen. From creating a conflict to taking an action, player input almost always comes from outside. A computer cannot play this game, because the computer can't formulate, "I'm putting out this conflict because it will make the other players care enough to invest in it because their characters are all keyed into this central theme, so then I can harvest a whole lot of tokens off of it." The game requires human input and human investment.
Now, I pitted two board games against a roleplaying game. By no means am I suggesting that board games are universally playable out of the box. Games like Pit, any trick-taking card game, and a number of poker variants, not to mention Go, all require out-of-game input in order to play, and cannot be played out of the box. However, I can't think of any roleplaying game that can be played out of the box, simply because the foundational act of role-playing fundamentally requires input from outside the box.
Game - people participating in an enjoyable experience with structured (but not necessarily formal) rules of interaction. A game is a social experience.
Book - a published artifact bearing information. In the context of RPGs, the book is what you hold in your hands, read, and reference. You cannot play a book; you can only play a game.
Rule - a principle that describes acceptable behavior. In the context of RPGs, a rule can be suggested as part of a text, which can in turn be presented in a book, but the suggestion only becomes a rule when the players agree to use it. Not all Rules are presented in books — some, like “No PC death” or “Don’t split the party” come from other sources, but they still determine what is and is not acceptable behavior. A collection of rules does not make a game (player interactions make a game), but it can describe how a game might be played.
System - the processes and procedures by which the players agree on the characteristics and development of fictional content. (Which is, yes, pure Lumpley Principle.) A system is always composed of rules (see above) but very rarely are all the rules of a system presented in one text — usually some rules are contributed by the social reality of the players.
But Josh, you may be saying, all of that isn't what the question is about. The question is about the rules of the game! And here we come to the real stumbling block. Are we talking about the rules presented in the book or are we talking about the system used in the game? These are not the same thing. One may be based in part off of the other, but the two are never identical.
Game designers sweat and toil to create a set of rules that work together, creating synergetic effects and guiding play in desired directions. Game players use rules to inspire, structure, and adjudicate the fiction that they create. However, the set of rules that the players use -- the System -- is broader than the set of rules that the designers provide. Every time. The decisions that the players make are rational decisions based on principles that are relatively persistent. You have no idea how many scrappy, cynical, loud-mouthed ideogogues I've played through the years. No game rule in any book ever told me to play such a character, and yet I have done it consistently for a long time. Similarily, good GMs and players have been using flags long before Chris Chinn identified and named the phenomenon. A friend of mine, for a third example, has everybody at the table vote for the best roleplayer at the end of each session, and the winner gets a candybar and bragging rights; they do this every game, regardless of what other rules they are using from other sources. These are rules, and these are parts of the System in use at gaming tables around the world.
The "Out of the Box" proposition suggests that the rules in the book can become the system of a game without any interpretation, modification, or addition. While this may be an admirable goal, it's also impossible to accomplish. All games are individual creations of their players, who take inspiration from multiple sources. One source may be more prominent in a given game than others -- a "Dogs in the Vineyard game" may base most of its System on the rules presented in the book Dogs in the Vineyard, but it cannot and will not take all of its System from that book. Vincent knows this; flip through Dogs sometime and look at how many times he references the play group and their preferences, advises everyone to pay attention to their fellow players, and explicitly recognizes that the interpretation of the rules is the responsibility of the players. It doesn't work any other way.
Nir Shiffer, who started the thread that spawned all of this, said, "My position was that given what I'd like to do with a game, I prefer a system that does this while requiring as little alteration as possible. The ideal system, then, would be one that I wouldn't have to change at all." I do think that he's assuming that system is presented in a book, which I'd quibble with, but on the whole, I can stand behind such a statement in a few different ways.
The ideal System in play does not require changes in the midst of play. Hear, hear! If you can fine-tune your System to do what you and your friends want and do it consistently, then bully for you! If everything that you're going to encounter already has a provision waiting for it, and everybody's on board with those provisions, your play is going to move like greased lightning. I suspect that a lot of gaming groups around the world are already there (Emily's article on the Ennead comes to mind). However, this is system in play, not rules presented in a book.
The ideal gamebook presents a set of rules that can be implemented as the System of a game without any changes at all. I'm perfectly fine with this as an ideal, much like never giving up, but that doesn't mean that it's actually acheivable. Unattainable ideals are good -- it means we never reach the destination and continue on the journey, and I'm all good with that. Further, challenging those ideals is a continual process that lets us further refine our understanding of how games really work -- that is, in fact, what you and I are doing right now in this article. Recognizing that this is an impossible ideal is important, though, else we become fanatics and veer off into crazyland.
The ideal gamebook presents a set of rules that can be implemented, with other rules that come from different sources, in a System-in-play without changing any rules from the book. This works for me, too, and this is a worthy and attainable goal for game designers. This is providing people with tools that don't need to be jury-rigged in order to operate, tools with access ports for the other tools and rules that come from other sources, tools that work. I suspect this is mostly what Nir and others mean when they say "Out of the Box."
So why am I railing against using this "Out of the Box" term? Because it short-changes players. It denigrates their creativity. It does not recognize that players are being creative and taking control of their own fun at the table. It suggests that all they are doing is following rules, that the high-almighty gamebook is the sole source of good times, and if you're not having fun what you need is a better gamebook. That is all utter crap. I'm sure that's not what Nir means -- not at all -- but I think it's a dangerous possibility if we start thinking that we game designers are packaging up experiences and selling them to players.
Games don't come in boxes. Stuff to play games with come in boxes. Games are created by playing them.