Writing as a Tool for Analyzing Problems (part 1)

A big part of game design is problem solving. We have to solve problems to make a game work the way we want it to. But oftentimes, we develop blind spots. We overlook something, and that keeps us from realizing there's a solution in front of us.

I've found that writing about problems is a good tool to solving problems. The process of describing the problem to another person can help put the problem in clear focus. Accurately defining a problem goes a long way towards solving the problem; in fact, in a few cases it outright solves the problem.


I'd like to talk about a weird experience I've had while designing a live-action game for my friends. One night I was staying up writing out the ideas we've had for the game up till that point. My purpose was to make a list of problems to be solved in the next meeting. I tried to describe the problems accurately. I edited it a lot. Every time I wrote a sentence there was something I want to fix. I kept massaging the text over and over, trying to define the problems in clearer language. By the time I was done, I had somehow ended up with a set of working game rules that answered all of the standing problems.

It was weird because I wasn't actually trying to solve any of those problems. As I mentioned, at that stage I just wanted to describe the problems so we could talk about it in the upcoming meeting.

In this article, I'm going to try to recreate that experience by reconstructing my various edit versions. I don't have a copy of the original text with me now (and the different edits on it are lost even if I did) but I'll try to recreate the path that my edits took as best as I can. I'll break it up into short pieces so you can have fun seeing what solutions you come up with at each stage of the process.

....

The time was 1999. I was in college. At that time, I took part in a camping group that liked to do crazy and silly things. While this wasn't a group of gamers, one of things we liked to do every once in a while was to play capture-the-flag type games. 

For those who don't know, capture-the-flag is a kind of game you play out in the field, where you have some kind of live-action combat system using various weapons such as waterguns, slingshots, melee weapons, etc. With the most typical version of these games, each team has a flag planted in the team's territory or base camp. You try to fight or sneak past enemy defenses, steal the flag, and bring it back to your base camp.

At that time, the prequel to the Star Wars saga had just come out: it was Episode One - the Phantom Menace. The sword fights in that movie were so impressive (compared to the first films) that some of my friends and I just had to make a game to play it out first-hand.

I began working on this game with two other people. We brainstormed some ideas using the movie as a motif. There were some crazy ideas - one person had the idea of making a race track run by Jabba the Hut (as in the movie). While the main focus of the game was sword-fighting, the idea was that people could take part in these races or bet on them to make extra cash, then use the cash to upgrade their weapons in the shops. Although in the end we made the game much simpler than that, it always helped that we had a clear motif or theme (like a movie) to provide inspiration for a game.

Now, the basics of the game after the initial brainstorming was that there would be two teams - the Light team and the Dark team. Each team had a leader with a secret identity. The goal of the game was to find out the identity of the enemy leader and execute him/her. To help discover who the enemy team's leader is, we would have each player carry a piece of paper with a clue written on it about the leader of that player's team. Whenever you defeat a member of an opposing team in a swordfight, you get to steal the clue off of that player. When you collect all the clues, you know the identity of the opposing team's leader.

In addition to that, we had the basic combat rules (fight with paper/plastic swords) and rules about what happens when you lose. Basically, when you lose, you get towed away to the enemy base as a prisoner. You are freed when a living team member taps you. 

When you capture a player, you can choose to execute that player if you are sure that it is the enemy team's leader. But if you execute the wrong person, your team loses, so you have to collect clues and make sure first.

As I was writing an e-mail summary of what we had, in prepartion for a meeting between the three of us, I added in some ideas. I wrote, "You know, it would be cool if the leader of the Dark team was actually posing as a member of the Light team, just like in the movies. There might be a few difficulties, but I think it can be done."

One principle of writing is that when we write, we try to be clear. (Especially when we're writing for another person!) So when I wrote the above, it didn't feel right without elaborating what those "few difficulties" were.

So I described the problem. "Let's say the secret leader of the Dark team is this guy named Ken. We tell Ken to move over to the Light team. Then we tell the Light team, 'Hey guys, Ken here will now join your team. Now, try to discover who the imposter in your team is. Good luck!' It would be so obvious who the imposter is!"

Once I described that, the text felt incomplete without suggesting at least one solution as an example, even if it was one that I didn't particularly intend to use. I wrote, "I still think it could be done. For example, we could ask someone in advance to be the Dark leader and make sure he ends up in the Light team. K.S. would be a good candidate. He's a good swordfighter and leader. He's also pretty sensitive about whether everyone is having fun or not, so he would steer the game in a good way."

Having listed that approach, I wrote about the problems of that approach. "The problems with this is that it would be imposing a role onto someone that he or she might not want to play. Also, if the players know we pre-selected a Dark leader, they might be able to guess what kind of person we'd want to select. Finally, the fact that the answer to the mystery is pre-fixed by the 'game masters' is in itself not desirable. For these camp activities, I like it better when the players get to make their own puzzle, because it's much more participatory and creative that way."

I massaged that text a little more and what happened was that the description of the problem basically spelled out the answer. It wasn't the only problem I had to solve, though. I'll talk about what happened in the next part of this.

...

I think it would be fun if I break off this segment here so you can suggest ways to do this in the comment section. The solution to this first part might be obvious, so here's an additional question: If the solution you thought of is the same as mine, that solution would create a couple of new problems. What are those problems?