Open-ended games leave the players the freedom to choose their path and provide a rich environment with lots of highly interactive toys lying around. As we have seen, it may sometimes lead to a series of short-term objectives with no connection or consistency. While this is fun indeed, these short-term gameplay cycles reduce the player's anticipation.
As we all know, video game worlds are often extremely static. Merchants wait at the same spot all day, monsters just wander aimlessly until adventurers kill them, and NPC's never loot chests. Though it makes no sense in those game worlds, it is necessary for players to control what happens. When everything is dynamically changing, they cannot anticipate the results of their actions, and this leaves little room for long-term strategies.
Conversely, complex dynamic behaviors decrease the player's anticipation. Unfortunately, they have the same effect on the developers. And when we fail to anticipate how the game will behave, this leads to a larger number of bugs. In fact, that's the definition of a bug: an unforeseen and unwanted behavior that emerges in a piece of software.
Thinking Outside the Box
When complexity leads to unexpected behaviors, most of them are bugs: they are unwanted. But what if those behaviors actually turn out to be interesting and make sense in the game world? Then the game designers turn their frown upside down and start calling it "emergent gameplay" -- lucky bugs, if you will.
Two famous examples of emergent gameplay are the ability to use wall mines as steps to climb up (Ion Storm's Deus Ex) or using the blast of a rocket to "rocket jump" to very high places (various FPS). Now this move has become a classic, and has even been integrated as a perfectly normal technique in Team Fortress 2 (Valve).
It takes a creative mind to find new solutions to gameplay problems, especially when these solutions are emergent: players even need to be more inventive than the designers.
The frontier between emergent gameplay and loopholes is quite thin. In fact, the distinction is purely moral: every player judges if such and such emergent features are a legitimate part of the game. One could argue that mine-climbing makes no sense, and therefore refuse to use it. As in every type of game, cheating is a moral issue.
Still, the act of cheating can indeed be creative when it implies finding a clever workaround. But I would not qualify cheating as gameplay, because it usually removes all obstacles intended by the designers. Without obstacles, there cannot be any gameplay.
We will not discuss hacking here and its creative merits, as it clearly happens outside of the gameplay. Creative gameplay actions can only happen inside the game. We have already seen though that the player's goal may come from "outside the game's box": from the player's imagination, mainly. But it may also come from an achievement or trophy system, for example.
What is the difference between internal and external objectives? Most the time, they are the same. A character asks the player to beat the boss, and an achievement asks the same. Even when an achievement is not consistent with in-game goals, it still is a gameplay objective, and this has no impact on creative potential.
But achievements may sometimes encourage players to guess what the objectives are. They do that by giving no description (or a very cryptic one) or just by being hidden away in a sub-sub-sub-menu.
With 250 achievements such as "die on downward spikes", "look hopeless", or "suspend a corpse for two seconds", Achievement Unlocked 2 (Armor Games) gives you many occasions to guess how silly or surrealist your remaining objectives are.
Is guessing creative? Again, creativity is more about the path than the destination. The way your brain sniffs in all directions to find the solution may be creative. Even if every player in the world will end up with the same answer, each player may or may not have been creative reaching it.
Guessing may happen without an achievement system. It may even work without an objective: guessing about hidden features or secret zones. Again, that may or may not be considered creative. Most seasoned gamers check behind waterfall for bonuses, but that attitude is clearly non-creative.
Guessing gives unique feedbacks to the players. They feel as if the designers are telling them, "Yeah, we thought about that too!" Usually, these features use an exogenous grammar. For example, in the first level of Kingdom Rush (Ironhide), I tried clicking sheep over and over, and I was thrilled to discover it had the same effect as in Warcraft. Similarly, a friend of mine was very excited when he discovered empirically that driving a DeLorean over 88mph in Driver San Francisco (Ubisoft Reflections) unlocked a special challenge.
What We May Learn from Board Games
Looking at non-video games is always very interesting. We, video game developers, are a very young species. The Egyptian game of Senet is said to be one of the first games ever, and apparently dates to around 5,000 BCB (Before Crash Bandicoot).
Video games changed the play paradigm by having computers assist or replace humans. It sounds creepy, but it really isn't: displaying graphics and playing sounds assists the player's imagination; managing the rules means the players do not have to do it; simulating AI characters replaces human opponents . Even the human player's actions are sometimes simulated (real-time demos, tutorial examples, AI vs. AI fights, etc.)
At this point, you may suspect that I am going to rant about how board games are better, about the place of human in society, and the fact that computers may never understand the beauty of a sunset... Not at all: video games have a lot of cool specificities, and most of them come from the use of computers and what that allows for. I do not compare video games and non-video games to determine which is better. But creativity being highly subjective, the presence of computers has a large impact on the game system.
In almost every video game, a computer is in charge of managing the rules. In more traditional games, this has to be done by all the players or by a designated player (the game master, the referee, the banker, etc.) This difference allows for much more complex rule sets in video games: apart from hardcore pen and paper role players, no one would enjoy calculating the chances for a sword swing to hit a monster when it is influenced by a dozen variables. So the computer's enormous calculation power is put to good use. Board games, on the other hand, seldom allow the kind of creativity that emerges from the game system's complexity .
We have seen about Create that a computer is unable to identify or appraise creativity. The same goes for morality, artistic value, negotiation, aesthetics or any other subjective matter. In a board game, the players, who deal with these matters as they see fit, handle rules intelligently .
Board games, it seems, could just treat creativity as the game's goal. But often they don't. Again, this ability is a means, not an end.
When creativity is a major game mechanic of a board game, it serves a purpose such as making someone guess a hidden word by drawing or modeling clay (Cranium, Richard Tait & Whit Alexander), hide a few weird words in a free speech (Nonsense, Véronique Houbaert & Bernard Ralet), coming up with a believable definition for a word (The Dictionary Game), role playing to make a point or get oneself out of trouble (The Werewolves of Miller's Hollow, Philippe des Pallières & Hervé Marly) -- contra this last example, "role playing" may mean a lot of things in a video game: giving your character a name, chatting in a weird way on ye olde public channels, building stats... but it never implies that your role playing has an impact on gameplay.
Dixit (Jean-Louis Roubira) is a gorgeous and subtle board game in which creativity is encouraged and rewarded.
Video games do allow other players to rate another person's creativity, for example giving five stars to another player's custom level. But those decisions are made totally outside of gameplay. In the aforementioned board games, the other players' decisions are gameplay decisions: they have an impact on victory or defeat.
Video games have other players too. Of course, I am not talking about 12 year-old rage-quitting Counter-Strikers. But with the current trend of social networks and asynchronous games, multiplayer is more and more about taking time to play with people you know and like. Could we follow in the footsteps of board games?
Actually, we already do. The very popular Draw Something (OMGPOP) is obviously inspired by the timeless classic Pictionary (Rob Angel). Drawception (Nihildom) uses the same principles as the traditional game of Paper Telephone. A few MMOs allow rating other players' creations to determine their popularity score, such as Mamba Nation (Mimesis Republic).
I believe this trend is very interesting and makes a lot of sense for social games, as do many ideas coming from the world of non-video games. It is up to us to hear what they have to teach.
As this article nears its conclusion, I realize I cannot conclude this article without mentioning two of the greatest video games ever for creative players.
 The genius of a project like Chris Hecker's Spy Party precisely comes from asking people to act like AI (though saying that might be oversimplifying the game's concept).
 Though I have to point out that board games and various toys have used physics as a game mechanic long before Havok came out.
 As a consequence, there sometimes is much more arguing!