Thursday, April 14, 2011

Writing as a Tool for Analyzing Problems (part 2)

(Recap of part I:)
I was making a live-action sword-fighting game based on Star Wars Episode I. There was to be a Light team and a Dark team, each with a secret leader. The goal was to find the identity of the enemy leader using clues, and then execute that leader. The leader of the Dark team would actually be an imposter posing as a Light team member. While working on this, I tried defining the problems of this set-up in clear langauge, and doing so helped me solve the problems. The first problem was: how do we get a Dark team leader to infiltrate the Light team without making it obvious?

While writing an e-mail to the other people who were working on the game with me, I had written down two requirements:

  • If just one person moves between teams, it would be obvious who the imposter is. To avoid this, the imposter must be in the Light team from the beginning.
  • For more player participation, each team should get to choose their own leader instead of having the leaders pre-seeded by a game master.

Then it became obvoius that the Dark team should select their leader from someone who is already on the Light team. Rather than have a Dark member move over to the Light team, the selected Light team member would turn into a Dark team member.

As VanVelding pointed out in the comments of the previous part, this created two problems.

 #1. The teams wouldn't be the same size.

 #2. The Dark leader wouldn't know that he or she is the leader of the Dark team.

Also, since both teams would be selecting their secret leader from the Light team members, we get two additional problems:

 #3. The two teams might each select the same person as leader by accident.

 #4. Since the Dark team leader is among the Light team from the beginning, he would be involved in the selection of the Light team leader. The Dark leader would know who the Light leader is.

The problem with team size was easy to solve - we simply had the Light team start with an extra member. (By the way, we decided that for this game, we would tell the players from the beginning that the Dark leader is lurking somewhere among the Light team members. But with more experienced gamers, it might have been interesting to keep it a complete surprise.)

As for the leader not knowing his/her own identity, I started writing out the problem as before:

We'll need to find a way to notify the chosen Dark leader without giving away the secret. If we hand a note to that person, that would single him out and it would be obvious that the recipient is the leader. We could instead take the leader aside and tell him, but that could be hard to pull off discreetly.

As I mentioned before, I wasn't concerned with solving the problem yet. I just wanted to bring it up for the meeting. But when I wrote the words "that would single him out," it made me think "what is a case where handing a note doesn't single out a person?" and asking the question in that way made the solution obvious. I edited the above to:

We have to give a note to each person on the Light team. The one given to the chosen leader will bear the message, "You are the Dark leader." The others will be dummy messages with words like "You are a brave and loyal Jedi."

The solution to this problem may be easy, and I'm sure many of you were able to make the jump to the solution in a single step.

But the thing that was working for me was, after saying things in the format of "This can't be done, because..." I was rephrasing them in the form of "As long as condition A exists, this won't work." Stating it this way puts me in the mindset of "What are some situations where condition A doesn't exist?" And this helped a lot when I went on to define more complex problems like #3 and #4, which I'll talk about in the next post.

By the way, I'm sure that there could be more than one way to solve problem #3 and #4, and I would be very interested in any solutions you can come up with.


  1. So how did the game play?
    Also, I'm curious about your combat rules.

  2. How do your clues work? If you use the balloting/clues system to subtly tell each side who the leaders are, then it suddenly becomes very important.

    Also, do the leaders actually act as leaders? If so, there's really no way for the Light Side to keep a Dark Side infiltrator from knowing who's in charge.

  3. Ok, I'll cover that in the next post.

  4. VanVelding: The way the game ended up, the leaders didn't actually act as leaders, they were just final targets for the opponents.

    I set up a clues system where the players get to make the clues themselves, except it must follow a particular format to control the level of difficulty.