Dave Turner said:
That always makes me ask, though: if I have to make up the rules that I need to enjoy your game, then why did I need you in the first place? Why didn't you get it right?
Because there's a difference between a rule and a mechanic. Good rules can be used in virtually any game because, being guidelines, they're not strictly bound by the game they originated in. There may not be a rule that specifies exactly how to handle a situation, but the mechanics should be robust enough to be able to handle the rule you create and the rules should be broad enough to give an indicator of how to apply the mechanic. If I have a rule that says, "Throwing a knife is resolved by rolling DEX+d20+Throwing skill", I can extrapolate that throwing a rock will use the same mechanic.
It is not the fault of the designer that every iteration of "throwing x" is not covered by a rule. It is the fault of the designer if the rules are not crafted in broad enough manner to not allow extrapolation, or the mechanics are not flexible enough to allow the application, just as it's the fault of the GM for being incapable (or unwilling) to extrapolate rules from what already exists.
EDIT: I just remembered a very apt concept. If you run most games strictly by the mechanics, 9 times out of 10 those mechanics will wind up violating what you know to be true - either as a result of your experience in the real world, or as a result of your expectations from the genre you are playing in, or just as a violation of common sense.
For example, I have a game where PCs have 20 hit points and a dagger does 4 points of damage. Mechanically, someone wielding that dagger can never kill a PC while they are sleeping. That violates our understanding of how the world works.
Rules adjudicate the mechanics, by saying,
"A coup de grace will kill any character regardless of their hit points. In this task, by adjudicating the mechanics to help facilitate
the expected range of possible results, by necessity some rules need to be conceived during play. When you decide, for example, that a 1 in 10 chance of fumbling is too much and implement a rule that when a 1 is rolled, you roll again, you are doing so to abjudicate what you believe is a shortcoming in the current rule.
Once again, this is not a failing on the part of the game designer - it's a difference in opinion. I don't buy games to have definitive rules that shouldn't be changed, although the better designed the system is the less likely I am to change the rules.