Sure. That was my point all along. We drill down to find what really matters to the player, and determine if that is compatible. If it's not--if there are genuinely, utterly irreconcilable conflicts that appear as early as character creation, the group needs to reevaluate. Maybe the player leaves, maybe the GM does, maybe they agree to do a different activity (whether a different setting, a different system, or something entirely unrelated).
Ah, okay. It seemed very odd in context so I understand. My apologies if I was excessively testy.
When it comes to the actual rules of the game, as in the written textual elements, I consider things pretty different compared to the unspoken abstractions or high-altitude things like "setting concept" or the like. Rules, assuming they aren't crap, are clear and relatively well-written. They communicate their function. Hence, by agreeing to a particular game, one is signing up to make use of the rules in question. Some leeway should be allowed since I don't expect anyone to have encyclopedic knowledge of any system, but that leeway should have limits. Hence, the GM (or whomever; could also be other players!) enforcing the rules--and trying to preserve the functionality of the game's rules--is an absolute bedrock element of "playing a game". If we don't have that, we don't have "a game" in the first place, so the activity can't happen. It's like saying that people taking a road trip need to maintain the vehicle they're driving in. If you don't preserve the functionality of the vehicle, your road trip ends. Probably very unpleasantly.
Hence, "permitting" here is a matter of evaluating the game-mechanical fit of certain things. Generally, this entails at least some degree of calculation. You can actually compare quantities, concretely. I prefer game design where players can determine, from their own efforts, that in general (not in specific!), they have several possible and truly distinct paths...and all of them evaluate to having pretty much the same calculated value. When that happens, you get magic: the players must make their mechanical decisions on the basis of qualitative, not quantitative, reasoning. When almost all reasoning occurs quantitatively, the player is thinking about what matters, not about what evaluates. By having multiple distinct paths with comparable arithmetic value, arithmetic value cannot be what decides, and that inherently puts anyone playing--whether "gamist" or not--into the right kind of mindset for great roleplaying. Your head stays in the fiction almost all the time, because worrying overmuch about mechanics makes no difference. (This is why I am so opposed to badly-balanced games. Bad balance encourages pure-mechanics thinking, rather than discouraging it! You get rewarded for thinking mechanics-first.)
As a result, it's very important to build and maintain healthy rules function. For games like D&D, it's usually the GM who is best-equipped to do that. Sometimes that's not entirely true; I know people on this forum who have said that they trust one of their players to be the big brain when it comes to game-rules stuff.
Well, I have some ideas on that front, but they require more brain than I have right now. Perhaps later.
Okay. My concern was only limitedly practical here. More or less, the only practical concern I was voicing was that I see "bright lines" as being a purely personal thing, a "this would upset me, so please don't do that" kind of thing. Not instructions, more like...well, friends being friends with one another and recognizing that there are things it's okay to do and things it's not okay to do, though such things should be talked out rather than left to implication alone.
"Don't cheat and don't try to break the game", on the other hand, does not fit well into the "this is a personal request from me" mold. That's a thing everyone--including the GM--should be on board with from the beginning. While I do think it needs to be said at some point, given how deeply embedded it is, I think there's an argument that it can also be presumed to be true and anyone who pretends that they somehow don't know that you shouldn't cheat at games is probably being a disingenuous jerk.
(This is also, incidentally, a different aspect of why I personally am very opposed to fudging--noting that I define "fudging" slightly more narrowly than some do.)