The problem with revised editions of games is that they frequently revise too much. Most RPG engines -- the crunchy bits of core mechanics which make the game run, hence the term -- start off with a very simple idea:
- "Roll a d20 and beat a number."
- "Add up your stat + skill and roll that many dice."
- "Pick a card."
Of course, an engine by itself is useless. You need wheels to make the car move. You need axles to attach the wheels. You need a transmission to get power to the wheels. You need a frame to hold everything together.
What inevitably happens, though, is that what is needed to make an RPG work is often inundated with optional rules and fiddly bits. Now don't get me wrong, I like fiddly bits in my games, because they usually give me a finer degree of control over my character, either in generation or in play. But they aren't necessary. The fact that Holmes-level D&D is thriving in the OSR while the far more complex Pathfinder and 4e D&D are on shelves is a testament to that fact.
As this complex game grows, it accumulates levels of complexity, much like Katamari Damacy. Whether this is good or bad depends on your philosophy, but it's a fact of life (and marketing) that games which are actively being sold and played experience regular growth of rules. Eventually, the system reaches a point where the core game is lost under the sheer weight of all the expansion and supplements. I call this "splat bloat," but there are other names associated with it.
Once a system reaches splat bloat, the clock starts ticking for a new edition of the game. Sometimes this is because the customer base decides that enough is enough and stops buying the books; sometimes this is because the system is so massive that the writers and editors can't keep track of everything and either put out products which contradict each other, or else spend so much time referring to old material to make sure contradictions don't happen that most of their energy is spent in research instead of writing.
The problem with revisions is that they often revise too much. Sometimes, in the name of simplicity and accessibility for new readers, they revise the soul right out of the game. (Insert snark about your edition wars in your system of choice here. My personal selection is Mutants & Masterminds 2e. )
But instead of endless revisionism, why don't we attack the root of the problem instead of sawing at its branches? To whit: The game is only as complex as you allow it to be.
As BA so succinctly put it, "The first time you read a RPG rulebook, you assume it's full of rules." It is holy writ for you, and you follow it to the letter. It is only much, much later, after years of experience, that you allow yourself the heretical notion that you can "simply look at all the text and sections and subsections and decide right then and there how crunchy you think the game should be."
My proposal is that we help novice GMs and players realize this by using a form of symbolic notation to let them know which parts of the game engine are crucial and which parts are not. Continuing with the car analogy, it would be a bit like having the following notations in the rulebook:
- "This is the carburetor. It won't work without it."
- "This is the automatic transmission. It's nice but not strictly necessary."
- "This is nitrous oxide. Use this only if you know what you're doing."
I envision a simple system, like a symbol of a gear in the margins. The more complex the rule or idea, the more cogs are in the gear. The heart of the engine ("Roll a d20") would have one cog. Character generation bits, such as skills or classes, would have two cogs. Tactical combat options would probably have many cogs.
This would, ideally, result in players and Game Masters realizing that not everything in a game book needs to be used all at once, and that the vast majority of RPG rules are optional.