you are a strong and convincing advocate to "rulings not rules" !
I'm not 100% sure if you're being ironic, or if you're referring to my discussion of 4e's p 42/skill challenge stuff.
If the former, disregard what follows as it reflects a failure to follow your rhetoric!
If the latter, the following is an elaboration.
(If neither, just start again from scratch!)
I can think of three general variations on "rulings not rules".
The first two take as their pivot that action declarations are grounded in the fictional positioning of the PCs.
Variant (1): the poster children for this, in my mind, are Tomb of Horrors and White Plume Mountain. Action declaration is almost always expressed by relation to the GM's narrated fiction ("I poke my staff through the devil mouth"; "I take the doors of their hinges so we can surf them down the frictionless corridor").
The players describe their PCs' actions in purely, or almost purely, fictional terms, and the GM narrates what happens. In the Tweet/Edwards scheme of action resolution (fortune, karma, drama) this is resolution by means of drama (= talking). In the Baker/Crane scheme of techniques, it's a subtle and complex application of "say yes".
I almost never see this technique advocated for resolving combat, although in principle why "I stab him in the back when he's not looking" is just as amenable to this sort of resolution as "I hammer pitons into the wall and use them to climb up it." (Perhaps backed up by "Remember how I said I was sharpening my sword after the last fight? So it shouldn't have any trouble penetrating his leather armour.")
Indeed, in threads that debate how central combat is to D&D, those who (i) concede that D&D has a heavy dose of combat rues, but (ii) deny that combat is central to D&D, often argue that combat
needs mechanics for its resolution (ie rules that go beyond narration procedures) in a way that is different from other sorts of action declaration.
Here is my tentative hypothesis about this: in D&D, the most important "stake" is character survival. And combat is, by default, to the death (0 hp). And it would be bad for the game for a GM to fiat a player into losing (by declaring his/her PC dead via drama resolution). Hence, when PC death is on the line, we pull out the dice as mediators. (Hence also why ToH, with so much die-without-a-save, is sometimes seen as unfair.)
This variant of rulings-not-rules doesn't appeal to me very much, for two reasons. First, it only works if all the table are in agreement on the fictional positioning. As I posted upthread, high fantasy (and gonzo genre more generally) can tend to put pressure on that agreement. Second, it treats PC death as the only crucial stake for the players. I don't really like this narrowness of scope, because if you are playing for story (as I prefer to) it is too narrow from the dramatic point of view.
Variant (2): For this I think of Dungeon World (much discussed above), Marvel Heroic, Burning Wheel, HeroWars/Quest, 4e's skill challenges/p 42, and no doubt many other games I don't know as well as I should.
Action resolution starts like variant (1), being expressed in terms of fictional positioning of the PCs, but it isn't resolved simply by drama/GM says yes-or-no. The rules of the game give relatively detailed advice and structured procedures for framing a needed dice roll,
and give the player a relatively rich suite of resources to bring to bear to build up a dice pool or bundle of die roll modifiers or whatever,
and resolution is then determined by roll, with the GM narrating the consequences in accordance with the game's guidelines.
I am a big fan of variant (2). Compared to variant (1) it somewhat de-centres the GM, or at least positions him/her slightly differently, because ultimately it is the dice which decide. And it also makes negotiations between players and GM over the framing easier, because this is all done as part of determining the die roll needed
before consequences are narrated - whereas in variant (1) narration of framing and narration of consequences bleed into one another, with the result that player input into the framing too easily becomes player complaining about the outcome and/or pleading for takebacks/do-overs.
Another strength of variant (2) is that the players get to choose (through various devices, and with more or less freedom at various points in play, depending on game details) how to build up and deploy pools of resources, which means they get to choose when to go all-in (because the stakes are really high, for them) and when to let the odds run against them, thereby increasing the likelihood that the PC will fail and the GM will get his/her nefarious way. It's a nice dynamic for story-focused play, I think.
Now to variant (3) of rulings-not-rules. This is the variant that I think [MENTION=6696971]Manbearcat[/MENTION] has in mind when criticising aspects of 5e's approach. (If not, or if he doesn't like my approach to distinguishing variants, he can of course correct me!) Here the focus is not especially on fictional positioning. Rather, it is understood that resolution will be by way of die roll, but (i) the procedures for generating the target number are very fuzzy, (ii) the procedures for identifying a player's available resources to modify the die roll, and for governing their deployment, are very fuzzy, and (iii) the procedures for narrating consequences of a successful or failed die roll are very fuzzy.
A classic example of this would be ways of interpreting reaction or morale rolls that emerged in the early to mid 80s, post-Gygaxian D&D community: not as binding constraints on GM narration of monster behaviour (which rewards players who build PCs with high CHA, who systematically stage ambushes and take down leaders, etc) but as "guidelines" or "imagination prompts" for GM narration of monster behaviour.
Another example would be ad hoc "called shots": "I cut the guy's head off" - "OK, roll to hit with a -4 penalty". Outcomes become somewhat random relative to player choices about moves within the fiction (eg no longer is finding a vorpal sword necessary to cut someone's head off in combat other than by reducing them to 0 hp) and choices about resource deployment (the player didn't have to choose between, say, using a vorpal sword and using a +5 defender).
A contemporary example, in my view, is the 5e stealth rules, where the interaction between the resolution mechanics and player choices about resource deployment (PC build eg being a wood elf rather than a human; and during play, such as hiding behind a wall vs hiding in dim light etc) are very fuzzy.
I find that variant (3) tends in play towards variant (1), but with the huge detriment that on the way there it potentially burns players for their build choices or play choices, by appearing to give with one hand (mechanical choices, or in-play choices that are presented as having mechanical significance) while taking away with the other (because the rules for mediating those mechanical or mechanically significant choices into a resolution outcome are so fuzzy).
A borderline case, in my view, is the 5e surprise rule that requires the GM to determine what counts as
a threat, and hence a potential source of surprise at the start of an encounter. This is close to variant (2) - because the framing is grounded in the fiction, and takes place prior to resolution - but peculiarities of 5e can push it towards a version of variant (1) - the GM narrating the players into a losing situation - which burns their resource choices on the way through. This is because 5e (like 3E and 4e before it) tends to emphasise PC build as the main place for resource choices. Compare this to a Burning Wheel Steel test, which likewise is triggered by GM determination that some sort of threat is present, but still leaves a player free to deploy various fate-point style resources, and other similar stuff, to try and respond to the thread, making it pretty clearly an instance of variant ([-]1[/-] 2).
Here are the two things I think that would put the 5e surprise rules in my variant (2) and hence would, in my view, make it a good application of rulings-not rules:
First, the GM sticks very rigidly to the rolled Stealth results for creatures (a bit like a Gygaxian GM sticking hard to rolled reaction and morale results).
Second, the GM allows the players to spend Inspiration to get advantage (+5) to their passive Perception so that being surprised isn't just a narrated consequence, but one that truly follows from player choices about stakes and resource deployment.
TL;DR: some thoughts about "rulings not rules".
