D&D General Creativity?

[Manbearcat]

I'm reading the Dungeon World SRD right now (again) just in case I missed something and I just don't see it. Dungeon World moves as written don't allow you to narrate any more than a D&D proposition would. In fact, they are just encoding the most common sorts of D&D propositions.

@EzekielRaiden engaged with this a bit, but I'm going to go a bit further.

The above misses key components of Apocalypse World integrated GMing and system and the systemitized fundamentals of the structured conversation of play in these games. I wrote out a Stonetop excerpt of play above that featured a Know Things move (which is Spout Lore in DW). I can't recall whether or not in this instance if the fiction was conjectured by a player downstream of "ask question and use the answers" or just the typical back-and-forth about speculating over the fiction and then putting it to the test when you trigger a move > resolving the move based on the procedures (eg on a 10+ the insight gleaned must be both interesting and useful; eg clearly and immediately actionable).

But its bog standard in Apocalypse World (and derivatives) for either (a) the GM to ask questions and use the answers such that it amounts to "player making a stipulation about the accumulating fiction/situation contingent upon move outcome." On a 10+ we consult the move and the stipulation resolves (assuming the stipulation both (i) honors the move structure/result completely + (ii) honors the accumulated fiction to-date) or might need to be subtly modified (to observe the structured outcome of the move result typically), on a 7-9 the stipulation resolves with some level of modification (to observe the structured outcome of a 7-9 result), or on a 6- perhaps some of the stipulation is true but also this really bad news is also true (now deal with it) or the GM reorients the player's stipulation "painfully" by making a turn their move back on them move.

This is typical of knowledge/perception moves from Spout Lore to Divination moves like the Fighter's Heirloom Weapon or Through Death's Eyes to the "choose x" result of varying moves where the players either stipulate what is true or forbid the GM from bringing a particular consequence to bear (or sometimes both...depends on the move and the result).

 

[EzekielRaiden]

This is typical of knowledge/perception moves from Spout Lore to Divination moves like the Fighter's Heirloom Weapon or Through Death's Eyes to the "choose x" result of varying moves where the players either stipulate what is true or forbid the GM from bringing a particular consequence to bear (or sometimes both...depends on the move and the result).

Exactly.

@Celebrim As an example, Bend Bars, Lift Gates explicitly includes the option to declare that, whatever damage you did, you can fix the thing again without a lot of effort. This recontextualizes any "breaking" you might be doing into something more like bending, dislodging, or detaching, which is highly useful for stealth or intrigue contexts (e.g. leaving no evidence you were there.) Or you can declare that nothing of value is damaged, which has all kinds of recontextualization effects: "nothing of value" also refers to things you value, in addition to things others value, so it's got self-protection elements. These are WAY above and beyond what any version of D&D (except, possibly, 4e!) has allowed players to do with a single skill. Declaring you can fix it again easily, nothing of value was damaged, and it doesn't make an inordinate amount of nose? That's easily three skill checks back to back (strength check, likely either some kind of saving throw or some kind of Wisdom/Perception check, and a stealth check), and that's from a relatively permissive "trad D&D" DM. Many "trad D&D" DMs fall prey to the "keep rolling stealth to stay hidden" problem--whatever feat of physical prowess is the target of BBLG, it could very easily be something "trad D&D" DMs would expect to require multiple rolls.

Further, as is often the case for DW moves, the things you do not choose automatically become valid playing pieces for the GM to frame the scene with. Say the player rolls a full success on Bend Bars, Lift Gates. They choose the three options already listed: easily fixable, valuables undamaged, not an inordinate amount of noise. That means the player didn't (and couldn't) choose "it doesn't take a long time," which per the DW rules is a "Golden Opportunity" that invites GM moves. Clearly, the player is prioritizing stealth or at least trying not to leave noticeable signs--but needing to take a long time to do something is always a risk, sometimes a big one, when you're trying to go unnoticed. The fact that the player didn't choose "it doesn't take a long time" recontextualizes the scene just as much as the fact that they did choose the other three things.

 

[Pedantic]

My assertion is that, in fact, they are functionally very different. I make this assertion based on decades of play of A&D and Rolemaster, which is - in this respect - not very different from AD&D.

Here's a relevant quote from Vincent Baker:

​

Here's a quick resolution mechanism.​

​

1. We each say what our characters are trying to accomplish. For instance: "My character's trying to get away." "My character's trying to shoot yours."​

​

2. We roll dice or draw cards against one another to see which character or characters accomplish what they're trying to accomplish. For instance: "Oh no! My character doesn't get away." "Hooray! My character shoots yours."​

​

What must we establish before we roll? What our characters intend to accomplish.​

​

What does the roll decide? Whether our characters indeed accomplish what they intend.​

​

What do the rules never, ever, ever require us to say? The details of our characters' actual actions. It's like one minute both our characters are poised to act, and the next minute my character's stuck in the room and your character's shot her, but we never see my character scrambling to open the window and we never hear your character's gun go off.​

The player who is making the climbing check has to tell us what they are actually doing to traverse the surface (at least, assuming a model of action declaration similar to Prince Valiant or Burning Wheel or Rolemaster - declare action, which tells us which skill at what difficulty, then resolve action).

The fact of the dice roll doesn't just introduce uncertainty - it makes the fiction (which underlies likely consequences of failure) salient to both declaration and resolution.

Casting a fiat-type spell - Spider Climb can be an example, though sometimes it may not be (depending on further fictional details); but Rolemaster's Leaving or Long Door almost always is, as is Dimension Door - is like Vincent's example. We see the start of the scene, we see the end of the scene, we never actually see the character in action.

You're assuming a completely different model of Climb checks here than I am, and I think the difference is salient. A climb check is a percentage chance you get the ability "Gain a climb speed equal to 1/2 your base movement speed" and it's my general expectation that players invested in climbing should drive that chance up to 100% very quickly. Other abilities available include "gain a climb speed equal to your base movement speed" and the percentage change of activation night get with some features of the wall.

It's a design flaw certainly that your wizard will be flying eventually, but the LFQW problem is an issue of class design, not action resolution.

The only difference between a character resolving a situation using skills vs. spells, is that the skill user has a percentage chance of failure on some action declarations, but does not spend consumable resources to declare them. That isn't an intrinsic point of differentiation, characters can still use should they've broken the RNG on to declare actions that skill allows for.

 

[Celebrim]

As an example, Bend Bars, Lift Gates explicitly includes the option to declare that, whatever damage you did, you can fix the thing again without a lot of effort. This recontextualizes any "breaking" you might be doing into something more like bending, dislodging, or detaching, which is highly useful for stealth or intrigue contexts

I think you are really confused. And I'm beginning to think that word "recontextualizes" is meaningless, or at the very least grandiloquent. The way you are using the word, all propositions recontextualize the fiction.

To begin with, let's talk about the process of play in a Trad game. In a Trad game, you analyze an inanimate object in the fiction and you make a proposition to interact with it. This could include breaking, bending, dislodging, detaching, or whatever any piece or part of that object. That is to say in Trad play what you propose to do is specific, concrete, interaction with the fiction. You do not (usually) propose Moves in Trad play. You can but the proposition of a move is often strictly secondary to a natural language proposition or as an alternative to a natural language proposition. For example, when interacting with the lock, you might want to say, "I visually inspect the lock mechanism, being careful not to touch anything." prior to declaring a move like, "I check for traps." (Implicitly declaring the intent to use a "Find Traps" skill.)

A proposition like, "Using a prybar, I want to remove the bolts from the lock mechanism quietly so that I can disassemble it without damaging it permanently" is perfectly valid Trad proposition. The main differences between that and DW's Fighter Move is that the Fighter Move is abstract and has priority and it's up to the player and the GM to talk out what that abstract move means in terms of concrete fiction. But I can equally imagine as a process of play in Trad gaming, a player asking the GM, "Is there a way I can remove the lock without permanently damaging it and without making too much noise?" and a discussion going on about whether that was possible and what that meant before the GM decided whether or not a fortune roll was needed (or was it such a straightforward plan it just happened) and if a fortune roll was needed what would it be.

The only real advantage here is that DW has kind of outlined this process of play formally and the only real mechanical change is that by emphasizing the abstract over the concrete, the player is put into a position where less often the GM is going to say, "That particular outcome is not possible." And if you want to say, "No, the GM can't say no.", then I suggest you should consider what happens when the player calls "Bend Bars, Lift Gates" as a move on The One Ring, Castle Greyskull, or Mount Everest. It's clear that the talking it out phase in Dungeon World includes some degree of negotiation about how much time and strength is needed to break something. The abstraction works only if the player is not trying to use the Move abusively and is using it against ordinary dungeon features and items of a suitable size (and possibly with suitable tools available, if only a trusty battle-axe).

But yes, I concede it is in some ways a better written skill than many typical D&D skills. But I don't concede that it is different in kind from D&D skills. It's not any different than doing something like declaring you are tumbling through a threatened area without drawing an AoO. You aren't recontextualizing the scene. You are only proposing to do something within the established fiction.

Declaring you can fix it again easily, nothing of value was damaged, and it doesn't make an inordinate amount of nose? That's easily three skill checks back to back (strength check, likely either some kind of saving throw or some kind of Wisdom/Perception check, and a stealth check), and that's from a relatively permissive "trad D&D" DM.

This is a strawman. I can see a lot of different skills being potentially applied to sabotage an inanimate object, but I see no reason whatsoever to think it requires more than one roll.

Many "trad D&D" DMs fall prey to the "keep rolling stealth to stay hidden" problem--whatever feat of physical prowess is the target of BBLG, it could very easily be something "trad D&D" DMs would expect to require multiple rolls.

Sure, there are bad DMs. Whether or not it's a good idea to roll once and roll with it depends on the circumstances. You are imagining a hypothetical bad process of play and then burning that down.

Further, as is often the case for DW moves, the things you do not choose automatically become valid playing pieces for the GM to frame the scene with.

How is that any different than trad play. Anything not set by the stakes is always valid things for the GM to use to frame the resolution of the scene. Taking a long time doing something in Trad play always invites wandering monster checks or monsters making additional preparations. That's straight out of the 1e AD&D DMG. Breaking something causing noise, say forcing a barred door whether by brute strength or the Knock spell, always has the potential to alert nearby monsters. Are you saying though that the GM should metagame based on what the PC didn't pick? Like are you saying that a monster should be nearby to be alerted because the player didn't pick, "Doesn't cause noise."?

The fact that the player didn't choose "it doesn't take a long time" recontextualizes the scene just as much as the fact that they did choose the other three things.

No, it just changes the adjudication of the scene. That's very different than recontexualizing the scene. If all you mean by "recontextualizes" is "The shared fiction is changed after a successful proposition" (something that normally happens after a successful proposition), you really need to find a penny word to express that and stop using a 50-cent version. If you mean the outcome of the fortune recontexualizes the proposition itself, then that's just shifting from fortune in the middle to fortune at the beginning, which is a real shift because you have to backfill what the proposition was in the first place, but that's not really doing what you've been focused on hitherto.

 

[Lanefan]

Alright, with that out of the way, let me piece together my sense of the important constituent, converging parts here.

* At level 3, both AD&D and B/X put the M/U at 3 spells; 2/1.

* At this point in play, the significant bulk (exclusive in Basic) will be Dungeon Delves; contained adventuring sites with particular dynamics. One of those important dynamics is # of Obstacles/Problem Areas in a dungeon. Let us just establish a number here and say the average dungeon, before Wandering Monsters, will contain roughly 6-8 Obstacles/Problem Areas (not the same as # of Rooms).

* With 3 spells, and only 1 x level 2, the M/U's spell loadout and deployment decisions will be intensive and consequential as both their repertoire and their "gas tank" is thin, particularly with respect to # of Obstacles/Problem Areas faced; see my (2) above.

* Name Level for an M/U in Expert is 9, while its 11 in AD&D. For Expert there isn't much play left, but when you extend to CMI, there is a lot of play left. For AD&D, there is a ton of play left by level 11. Lets work off level 11. The spell loadouts are subtly different here and, imo, more advantageous to the AD&D (now) Wizard, despite the fact that the B/X (now) Wizard gets access to a level 6 spell while the AD&D counterpart does not. This is because the AD&D Wizard has 18 total spells usable by this point with a huge 12 between 1-3 levels and a chunky 6 at 4-5. Meanwhile, the B/X Wizard will have 2 less spells usable total (16), with those 2 less being in that sweet level 2-3 range (where there is an abundance of "answers to problems").

* So the problem is 4-fold. The first issue is the bullet point directly above about (i) spell proliferation/potency. The next 3 work in concert to create a positive feedback loop that amplifies the situation. Those are other 3 are:

(ii) The # of Obstacles/Problem Areas of a dungeon/adventure does not scale with either # spells usable or spellbook growth.

The number of Obstacles/Problem Areas might not scale with level but I posit the difficulty of each one very much does; meaning that where each O/PA at low level might only require zero or one spell from the caster (the rest of the party deals with the rest), each O/PA at high level might require two or three or even more spells; meaning spells then become just one more attrition-able resource similar to hit points etc., and for this model to work the game needs to throw more O/PAs at the party between rests than most parties (if they're wise) would want to allow.

This creates conflict between the game (and-or its DM) and the players/PCs, in that the game/DM is seeking an attrition-based model and the players/PCs are seeking to avoid just that. It's the 5-minute workday issue again, only approached from a different direction.

At most, you may see either (a) a small increase in total # of Obstacles/Problem Areas on a per dungeon/adventure basis or (b) an increase in "Megadungeons." But this doesn't remotely do the necessary work. Wizard players will suddenly have a considerable increase in their 2c (their prospective move-space/"answers") above while simultaneously dealing with a decrease of downward pressure on their loadout and resource management concerns.

(iii) Dungeon delves at this point notoriously break down for a myriad of reasons.

IME around level 11-ish the whole system kinda starts to fall apart, not just dungeon delves.

Consequently, the constraints/pressures on Wizards reduce further while their prospective move-space expands even further when out of that artificial "obstacle course" environment of the dungeon.

(iv) To compound matters, Wizards develop more means to both "refresh the Adventuring Day" and "bigger guns" to allowing them more potent means to "recontextualize situations" (or reframe scenes). These further reduce pressures on curation of loadout/spell deployment because refresh = your allocation of loadout per Obstacles/Problem Areas is now completely undone while Big Guns can mean obviating multiple Obstacles/Problem Areas (or even just fundamentally "re-wiring a scenario" in a single spell deployment.

I'd throw Clerics in here as well: by 12th-ish level they too can stand a whole lot of O/PAs on their ear with a single spell use.

So this is where I'm at with your "recontextualizing situations" paradigm. The game doesn't scale for M/U's turned Wizard. And that scaling problem isn't univariate. There are multiple issues with lack of scaling, the host of which converge to create Wizards who recontextualize situations at a rate that is far, far beyond their martial counterparts, therefore disproportionately controlling the trajectory of play.

Which, depending on one's long-term view of things, may not be all bad.

At very low levels the mages need the martials to keep them upright. At high levels the martials need the mages to keep them upright - the dependency has flipped around. It's a strange but IMO viable form of long-term balance, in which the zero-to-hero path is significantly steeper for one group of classes (casters) than the other (non-casters).

And how does that massive advantage get "offset?" By GM intervention via Force and Calvinball buggery that typically gets justified by either/or/both "simulation pretensions/veneer" or "storytelling imperatives."

In part, perhaps, but I think you're ignoring one other major factor that helps martials keep up at least to some extent: gear and magic items. And here the DM can make a big difference (if s/he thinks it matters), by placing more martial-oriented treasure and less caster-oriented treasure.

On a broader scale, I'm not sure how easy it would be to fix this in a 0e-BX-1e-2e system while at the same time avoiding either a lot of power creep or a lot of nerfing.

 

[bloodtide]

So this is where I'm at with your "recontextualizing situations" paradigm. The game doesn't scale for M/U's turned Wizard. And that scaling problem isn't univariate. There are multiple issues with lack of scaling, the host of which converge to create Wizards who recontextualize situations at a rate that is far, far beyond their martial counterparts, therefore disproportionately controlling the trajectory of play.

And how does that massive advantage get "offset?" By GM intervention via Force and Calvinball buggery that typically gets justified by either/or/both "simulation pretensions/veneer" or "storytelling imperatives."

D&D, by the vague rules really does not scale up the setting by the numbers, but it does sort of suggest it. Though the rules really only do it for monsters: at level one you fight goblins and at level 12 you fight fiends. And you get a hint or two about dangerous terrain and traps, and then the rules just drop off. And D&D, even more so the "core" of D&D has always been like that. Not that D&D really describes any basic setting either, you are just left on your own to do the "kinda like Earth long ago".

Of course this is already a problem as most gamers don't really know much about Earth That Was in the Past. And maybe worse, many only know history from TV shows and movies, so they think a typical castle is a square made of plywood painted gray like stone with a door, for example. Lots of castles have lots of mundane tricks, raps and desgine features to slow down, block and even kill attackers. Like a castle gatehouse: The gatehouse was filled with obstacles – multiple metal portcullis gateways; arrow-slits to fire at intruders; many different gates, doors and drawbridges; and even the infamous ‘murder holes’. So at low levels places have normal doors, and at medium level they have arcane locked doors.....but after that: no more ideas. The 15th level demon lord still just has a arcane locked door....that the lower level wizard can crack.

The next step beyond historical stuff is to use your imagination, but D&D really drops the ball here. Core D&D has never bothered much with world building rules that scale up, except traps and monsters. There are a couple of bits here and there, but little in the core rules. This leaves a lot of DMs lost: They don't know what to do. Even if they have a vague idea from their imagination, they are not sure how to do the crunch side. You do see this sort of thing in the rare high level adventure, but again not in the core rules.

Though having the DM create things has the problem where the DM feels like they can't say "no" to the players magic. Even using some of the things in the rules that do effect magic, like blocking teleportation, the DM will refuse to use as the magic using players would get sad.

On a broader scale, I'm not sure how easy it would be to fix this in a 0e-BX-1e-2e system while at the same time avoiding either a lot of power creep or a lot of nerfing.

The way to do it, is to have a great imagination, and not all for all the roadblocks like Putting Magic on a Pedestal. And drop the ideas that you "can't" do the power creep and "can't" nerf things. Just last weekend, one of my 12th ish level groups encountered a tunnel full of ectoplasim. No spell or magic any of the spellcasters had was of any help, so they were in the same boat as the mundanes in the group. So, now they are stuck, trying to figure out a way past this with no magic easy button.

Though I get a bit lost in all the "other" game discussions as all I see is the Easy GM: The players CO-GM so the GM has to do less work is what I see. And that does sound great for a more casual GM.

 

[Manbearcat]

One thing real quick as we now seem to be haggling over "recontextualizing a situation." I don't love the usage here, because it seems like its trying to default to the context indexing thematic relationships rather than game artifact relationships. I'm not interested in the former (at least for the purposes of my involvement in this thread), so I'm basically using it as "reorienting the constituent parts of the imagined space such that the way they index each other has now changed significantly." So not ret-conned and not thematic recontextualization. But as actually parts of an imagined space. Are you a problem to my goals? Now you're not. Are we over here? Nope...we're over there. Is this a thing at all? Nope, not a thing because I've rendered it irrelevant.

The number of Obstacles/Problem Areas might not scale with level but I posit the difficulty of each one very much does; meaning that where each O/PA at low level might only require zero or one spell from the caster (the rest of the party deals with the rest), each O/PA at high level might require two or three or even more spells; meaning spells then become just one more attrition-able resource similar to hit points etc., and for this model to work the game needs to throw more O/PAs at the party between rests than most parties (if they're wise) would want to allow.

This creates conflict between the game (and-or its DM) and the players/PCs, in that the game/DM is seeking an attrition-based model and the players/PCs are seeking to avoid just that. It's the 5-minute workday issue again, only approached from a different direction.

IME around level 11-ish the whole system kinda starts to fall apart, not just dungeon delves.

I'd throw Clerics in here as well: by 12th-ish level they too can stand a whole lot of O/PAs on their ear with a single spell use.

Which, depending on one's long-term view of things, may not be all bad.

At very low levels the mages need the martials to keep them upright. At high levels the martials need the mages to keep them upright - the dependency has flipped around. It's a strange but IMO viable form of long-term balance, in which the zero-to-hero path is significantly steeper for one group of classes (casters) than the other (non-casters).

In part, perhaps, but I think you're ignoring one other major factor that helps martials keep up at least to some extent: gear and magic items. And here the DM can make a big difference (if s/he thinks it matters), by placing more martial-oriented treasure and less caster-oriented treasure.

On a broader scale, I'm not sure how easy it would be to fix this in a 0e-BX-1e-2e system while at the same time avoiding either a lot of power creep or a lot of nerfing.

Really good post Lanefan!

I think we actually have a fair amount of agreement here (eg Clerics should get folded in with M/Us...particularly in 2e...and it gets worse in 3.x and 5e) while the disagreements are pretty nuanced. So to get to that:

* I see where you're going with the "Obstacles/Problem Areas generally need to become increasingly multidimensional as you go up in level." And I agree (generally). The issue with this approach is "contact with the enemy." The "enemy" here is two-fold; very clever Wizard players and the potency of higher level spells (or a single “big gun” matched with a level 1 supplementing spell to achieve a particular gambit) and their ability to neutralize that multi-dimensionality by fundamentally rewiring/reorienting the fiction.

* The problem with "the magic item solve" for the disparity between martials and casters is that the use-cases of sudden jumps in utility prowess are quite limited and focused (like Flying Boots) and its a rather fine line when balancing via magic item usage. The game is extremely sensitive to it...and the game is sufficiently fragile such that the other side of that fine line often equals broken.

The other issue with that "the magic item solve" is the default game already involves too much (and increasingly so as levels pile on) cognitive workload and handling time for GMs. Adding more isn't a great answer!

 

[pemerton]

It's a design flaw certainly that your wizard will be flying eventually, but the LFQW problem is an issue of class design, not action resolution.

I am going to borrow this post, by a participant in this thread, from another active thread:

Situation: a few characters hit a teleport trap and found themselves inside a small cell with transparent walls of force, in the middle of an otherwise big open area. They could see there were no immediate threats outside the cell, but had no way to escape (it's intended as a pure death trap; you get stuck in there and unless you've got some sort of dimension door or teleport or similar available, you eventually starve).

Nobody has any of those effects available but a caster does have Rope Trick. So, she casts Rope Trick and everyone climbs in. Rope is pulled up and then (and here's the creative bit) because there's nothing saying the rope always has to be lowered out of the middle of the "floor" of the little demi-plane it's instead lowered out of a corner, outside the cell wall! Everyone climbs down, and then they just have to worry about getting back to where they came from.

There we have the issue in microcosm.

In the scenario Lanefan describes, engagement by the players with the fiction is extremely attenuated. All the action turns on parsing spell descriptions: how does Wall of Force interact with extra-dimensional spaces? what does Rope Trick permit when the rope is lowered down?

Contrast how (say) MHRP/Cortex+ Heroic would resolve it: the player of the caster makes a roll based around Sorcery, which would be opposed by the Doom Pool augmented by an appropriate descriptor from the Wall of Force trap. Or contrast the DW Bend Bars/Lift Gates that @EzekielRaiden has been describing - another way of getting through obstacles - which puts the state of the fiction ("Is it broken?" "Did you make noise?" "Have you taken a long time?") at the core of the situation.

In @Manbearcat's recent post 230 he has focused on the integrity of the "competitive"/"gamist" play space. The point I'm making in this post might be relevant to that, or might be orthogonal to it. I'm still articulating my thoughts in response to @Campbell upthread (post 195), and Manbearcat's initial riff on that (post 198): that a traditional D&D MU/wizard has many abilities, particularly somewhere around 7th level and up, that permit the core unit of play to be reconceptualised while it is still ostensibly being resolved.

Campbell said "The reason to be disciplined about this is to maintain the integrity of the play space." I'm not sure if he meant the sort of competitive integrity Manbearcat is now focusing on, or something else. I'm not particularly focused on competitive integrity. Based on what Lanefan posted, it seems unlikely that escaping the Wall of Force trap by using Rope Trick damaged competitive integrity.

I'm focused on the difference between The players engaging, via play, the situation that has been framed and in which their PCs find themselves and Reframing that situation not via play, but via the deployment of fiat abilities that don't actually require engaging the fiction via action declaration. And the Rope Trick example illustrates what I have in mind - the players engage the spell rules, not the framed situation. As I posted upthread, the fact that we cover that in a fig leaf of fiction - that the wizard PC is consulting their knowledge of the arcane and is casting a spell - doesn't change how things are actually unfolding at the table.

 

[Pedantic]

I'm focused on the difference between The players engaging, via play, the situation that has been framed and in which their PCs find themselves and Reframing that situation not via play, but via the deployment of fiat abilities that don't actually require engaging the fiction via action declaration. And the Rope Trick example illustrates what I have in mind - the players engage the spell rules, not the framed situation. As I posted upthread, the fact that we cover that in a fig leaf of fiction - that the wizard PC is consulting their knowledge of the arcane and is casting a spell - doesn't change how things are actually unfolding at the table.

I just don't see this distinction. The whole point of play is to find some action you have available that interacts will with the situation to change it as you want. All abilities are "fiat" abilities. They do what they say they do, and you pick from the list of them to alter the world as appropriate. I'm not going to argue that Rope Trick doesn't have more narrative power than making a Balance check in most cases, but they're still both actions that are resolved in the same way. Making a balance check lets you move across a slippery surface of Y kind at X speed, and failing it means you can't, possibly with some associated failure penalty. The interesting bit there was the declaration of the action, the players choice to balance on the slippery beam and run to the point they wanted to get to, not "whether or not it worked." That part is just there because some mechanics involve gambling.

Like, you could rewrite the skill system to not involve rolling. Imagine a game where players just took 10 on every roll: they would still have mundane abilities to deploy. The idea of engaging some other rules than the "situation" is spurious. You are always picking rules to bring to bear against a described scene.

 

[pemerton]

I just don't see this distinction.

<snip>

The idea of engaging some other rules than the "situation" is spurious.

The first quoted sentence doesn't entail the second.

You are always picking rules to bring to bear against a described scene.

Here's some more from Vincent Baker - in my view perhaps the best single thing he's ever said about RPG design:

Roleplaying is negotiated imagination. In order for any thing to be true in game, all the participants in the game (players and GMs, if you've even got such things) have to understand and assent to it. When you're roleplaying, what you're doing is a) suggesting things that might be true in the game and then b) negotiating with the other participants to determine whether they're actually true or not.

<snip>

Mechanics might model the stuff of the game world, that's another topic, but they don't exist to do so. They exist to ease and constrain real-world social negotiation between the players at the table. That's their sole and crucial function.​

There are at least two ways the mechanics can facilitate resolution of, or even flat-out resolve, the question of who gets to add new truths to the fiction. One is by brining the participants closer in to the fiction, focusing on its details, making imaginative action declarations that play on those details and thus lead to everyone at the table accepting new truths about them.

The other is by bringing the participants to focus on the mechanics themselves as those operate independently of the fiction, to work out the mathematical, logical, conceptual or similar sorts of relationships between them, and from that to deduce a new state of the fiction.

The Rope Trick/Wall of Force scenario is an example of the second thing. All the recent climb checks that I've adjudicated (actually Dungeoneer checks in Torchbearer) have been examples of the first. Some ways the 3E D&D skill system operates - the ones I'm thinking of a are Diplomacy and Perception - operate in the second way. This is in my view a weakness of the 3E skill system. The ability to "take 10" I think doesn't bear on this at all. It's about whether or not the resolution process brings play into sustained contact with the shared fiction.