I’ve been writing a lot of rule books lately—polishing up the rules for New Bedford, writing and updating rules for submissions to the Greater than Games Metagame contest—and I thought it would be a good opportunity to talk about rulebooks. Back in June of last year, Foxtrot Games talked about the challenges of writing rulebooks. This is a perfect reminder of why it’s so hard to write rules. You can’t just write rules for people who read them. You also have to write rules for people who don’t read them. I’ve also been teaching some new games lately, and so I am forced to admit that I rarely teach a new game without missing at least one important rule, so I must include myself in the latter group.
There are two primary ways people tend to use rulebooks. The first is that they work through it while learning the game. Most rulebooks try to account for this by presenting rules in the order they are encountered as the players are learning. There may be some additional steps providing summaries, overviews, and establishing goals. But the information is presented in a logical order. Rulebooks are fairly easy to write this way because you start with explaining the minimum the player needs to know and elaborate on each aspect as you get farther into the rulebook.
But this style of writing also has drawbacks. When you’re intimately familiar with a design, it’s easy to forget all of the minor details when you teach a game. This extends to writing the rules. You know how to interpret everything, so it doesn’t occur to you to clarify some of the instructions. But by and large, this can be corrected by working through the rules as you teach a new game, and writing down every time a player has a question. Answer those questions in the appropriate section of the rules, and you’ve got a much better rulebook.
However, I’ve glossed over something very important in that last sentence, pertaining to how people use rulebooks. Because people don’t just use them once. Rulebooks are also where players turn to answer their questions, used like a reference. One way this frequently occurs trying to re-learn or teach the game after learning how to play from a third party (say, the designer). There is little reason for players to read rules front to back once they know them, or partially know them, so the rules must be ready to handle this new challenge.
Information in a rule can be categorized in multiple ways, and that makes the “appropriate section of rules” an ill-defined term. In the Foxtrot Games example, the player wanted to know whether multiple trades could be made. The question could be posed in (at least) two ways, “How many times may I perform a given action in a turn?” and “How many times may I perform an Exchange action in a turn?” These are only subtly different, but imply different locations to search. The latter points to the section about the “Exchange” action, while the former points to the section about turns. A third way of phrasing the question would be “How many times may I trade cards using Favor Tokens?” It ignores both the turn and “exchange” action, but points toward the section on Favor Tokens. The article succinctly summarizes the problem:
[The rulebook] does not say it where someone would look for it if they weren’t sure of the answer.
And the real challenge is that where players look for answers depends entirely on the way the question is asked. [And this is not to be taken as criticism of the Lanterns rules or the player, simply a case study of how confusion can arise even from well written rules]
The drawback of this is that it’s redundant. You explain the same rule multiple times, and cover the same information in multiple places. Players have to read the same thing several times, and receive the same information again. It’s also repetitive. This hurts the clarity of the rules a bit, making it harder to read. And repeating yourself can be off-putting to the reader. I’m guilty of occasionally saying “yes, I get it already” when reading rules. However, it also increases the chance that players will remember it if they read it several times, which isn’t a bad thing for the really tricky parts. But if you do this for every rule, you lose the logical clarity by having to explain everything everywhere. Rulebooks are already frequently pressed for space, so this isn’t a feasible option even if it were a good choice for clarity.
There is perhaps a third way of using rules that I find myself doing all too often, halfway between reading for instruction and reading for reference: the quick skim. When you are trying to learn a game in a large group setting, you don’t want someone sitting down and work through the rules from front to back for 30 minutes before you start. So it frequently turns into reading the main points and coming back to fill in the details later on. This makes writing your rules as a reference even more important. But the challenge is that you don’t know exactly what the player will need to find. You can put the key rules in boxes, or with large text size or color, but sometimes that isn’t enough. I’ve missed rules written in large bold letters simply because they were near the bottom of the page.
The approach I’ve taken with New Bedford is to use a sidebar, where all of the rules are listed in detail for learning the game, but the main points are highlighted at the side of the page for reference. Rules are cross-referenced where possible, and certain information, such as scoring, is listed where it first comes up, and reiterated in a final summary, trying to cover all the ways how people use it. The real test for rules is real world use, preferably blind testing, and ideally blind testing with the designer present to observe where people make rules mistakes and how they try to resolve them.