The Case Study: Rule-Based Analysis

A fundamental principle of law is that cases are decided based on experience garnered from previous cases. That experience is codified into rules, which are referenced whenever similar cases arise. So, for example, if a person accused of a crime says the police used unfair tactics during the interrogation the judge will decide whether the tactics were unfair by checking the rules from past cases. If the accused says “the police lied to me about having my mother in another interrogation room and I only confessed because I thought my mother might have to spend the night in jail, the police shouldn’t be able to trick me like that,” the judge will look to see whether there are rules about police lying to a suspect during interrogation. If there’s a rule that says that that is OK, the confession was valid and can be used against the suspect; if there is a rule saying the police are not allowed to act that way, the confession will be thrown out.

Making decisions based on rules has a few key advantages that are particularly relevant here. First, it means that likes are treated alike. A crime is a crime no matter who commits it; securities regulations must be followed no matter how big or small the company is; you are protected against certain forms of discrimination no matter where you are. Dealing with each case separately invites mistakes, honest and otherwise. When there’s one carefully thought-out rule for every decision maker to follow, results are more likely to be the result of principled analysis.

(It’s worth noting that in this context “principled” doesn’t necessarily mean “good” in some cosmic sense. Rather, it means something more like “fair.” People who have principles live according to rules that they follow consistently, and principled decisions are made according to even-handedly applied rules.)

Second, rule-based analysis avoids reinventing the wheel. If the Supreme Court has thoroughly considered whether the police should be allowed to trick suspects into confessing, after reading many briefs submitted by interested parties and hearing top-notch attorneys argue for both sides, it probably is not necessary to figure the answer out all over again. After all, if the process is done just as well it will lead to the same result! Better to save that time.

Hence, rule-based decision making is more likely to lead to principled results, is less prone to error, and is more efficient. Those are all traits I want in my game design, just like I want them to be a part of the law. It’s for that reason that I’m importing the technique into my design, and am basing this blog on legal analysis.

When I started work on Over the Next Dune I already had the notion that the law had something to teach me as a designer. Thus, just as a judge would use rules to decide a case, I chose four rules to drive the game’s design:

1. The decisions must be interesting. This rule is derived from . . . well, basically every game I’ve ever enjoyed. I was introduced both to chess and to Chutes and Ladders when I was young. I still play chess, but I haven’t played Chutes and Ladders in years. The former retains its appeal in part because each turn is a fascinating and rewarding puzzle. The latter doesn’t have anything to offer beyond the momentary thrill of seeing what happens next, and I can get much more of that thrill from a good book. I want Over the Next Dune to be more like chess than like Chutes and Ladders.

2. The players must work together. I learned this rule from modern cooperative games, most notably Pandemic and games that have followed it, and by comparing those games to the Star Trek Customizable Card Game. Much of the fun in cooperative games is in working as team. The excitement of finding the best move is amplified by the satisfaction of watching as all those best moves, yours and everyone else’s, snowball to achieve what no one player could. Even if the team loses, the dissatisfaction is mitigated by the fact that the loss is the team’s, and not one player’s alone. Over the Next Dune will be a better game if taps into those dynamics.

By contrast, the Star Trek CCG (at least in its first edition, which is what I played) was often criticized for being “multiplayer solitaire.” Players would start the game by putting up barriers to the opponent’s progress, in the form of “dilemmas” that needed to be resolved before points could be scored (e.g., before restoring an errant moon’s orbit you might have to retrieve a crew member who has run off with a love interest). After that initial process the players usually did not have anything to do with each other; they assembled their crews, faced dilemmas, and scored points entirely independently.

Setting aside the question of whether the Star Trek CCG really was multiplayer solitaire, the criticism points to an important lesson: when multiple players are involved in a game, they expect to be interacting with each other on an ongoing basis. Over the Next Dune needs to meet that expectation, or else it will face the same critiques the Star Trek CCG did.

3. While the players must work together, they cannot do so by talking about their moves. I’ll be honest in saying that this was a badly built rule. It’s too specific; it’s as if the rule about stealing was not “stealing is illegal” but rather “John Doe stealing from Jane Doe is illegal.” The more specific rule might be accurate, but it’s not helpful if Jane steals from John, or if Betsy steals from Jane. A rule should be generally applicable so that it’s useful in the future.

Fortunately, Over the Next Dune is still in development, so we can revisit this a bit. I was trying to solve two problems with this rule. The first was that in many cooperative games, the most experienced player will try to take over everyone else’s turns so that they make the moves that player thinks best. In the end only the most experienced player actually plays while the rest are puppeted about. It’s not fun, and I wanted a rule that would avoid the problem. So perhaps one rule (3a) should have been “it must not be possible for one player to dictate other players’ actions.”

Second, I wanted to capture the theme of people sneaking about under cover of darkness. The soldiers wouldn’t be able to stop and chat for all to hear, and I wanted the players to have to work with isolation and uncertainty just like the soldiers would. Rule 3b probably should have been something like “the rules for player communication must reinforce the theme of the game.” Going forward, let’s put rules 3a and 3b into place.

4. The game must admit of multiple solutions. In other words, there have to be many ways to win; it can’t be that there’s one thing the players always do. This rule came from my time playing fighting games (Street Fighter and the like) in arcades. I enjoyed the games where many characters were competitively viable, such that there was a range of things to try and lots of different opponents to face. When there was a strategically best option (e.g., in King of Fighters 2003 you should really play Duo Lon; in Third Strike you should almost certainly pick Yun) the game could get stale. I don’t want Over the Next Dune to get stale, so it should present lots of different challenges to players and they should be able to meet those challenges in many different ways.

Although development is ongoing, I’d like to stop for a moment and take stock. I have my own opinions about how the game is doing with regard to these rules, but they’re subject to change and I’m always looking for feedback. If you feel that the game is doing a particularly good job of living up to some of these requirements, or that it is falling short somehow, let me know!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s