Improvisation vs "code-breaking" in D&D

[pemerton]

This post is a response to some posts in this thread.

How can you not remember D&D before the 90s? OD&D was 10-15 years of the most hard-nosed, number-crunching, rule tweaking, nerd-filled gamers as ever existed. (Well, barring some 1970s wargamers.) But we were not well-formed, socially fit, artistic, nuanced expressers! Creating a narrative held no game challenge for us. You can't lose telling a story! (which is why those that can never be a game). We were gamers! Everyone of us.

As much as Howandwhy99 vs. the world is an entertaining spectator sport, I must still intrude to mention that you're discussing from such different perspectives that any progress is impossible.

Howandwhy99 sees the game from an idealized gamist standpoint. The game is a series of maps containing obstacles to overcome. This is is a completely legit approach - extreme perhaps, but completely functional.

His opponents see this as so extreme as to be nonfunctional

Clearly D&D can be played as if it were an offline version of World of Warcraft. I'm pretty sure that [MENTION=4937]Celebrim[/MENTION] agrees with that.

I think it's also clear that D&D can be played without treating the goal of play being to generate a story (in any interesting or rich sense of "story"). I'm pretty sure that [MENTION=4937]Celebrim[/MENTION] agrees with that too - eg when you look at modules like C1 or G1 the goal of play isn't to generate a story; it's for the players to beat the dungeon.

But Howandwhy99's assertions go far beyond this, and that is what makes them false.

First, the fact that you can play D&D without aiming at producing a story doesn't mean that before the 1990s no one played the game with the aim of producing a story. I know, myself, that when I was GMing Oriental Adventures in 1986 an important goal of play was creating a story (roughly along the lines of a B samurai movie). And clearly there were FRPGers much before then playing in the same way: Lewis Pulsipher talks a lot about different styles, including story-oriented styles, in articles written for White Dwarf in the late 70s/early 80s. (Here is a thread from last year discussing some of Pulsipher's ideas.)

Second, the fact that D&D can be played as an offline version of WoW doesn't mean the GM doesn't need to improvise. The need to improvise - by coming up with content, by inventing the backstories and motivations of randomly encountered monster/NPCs, etc is so self-evident it's bizarre that it needs to be pointed out. Celebrim pointed to the example of play in Gygax's DMG. I will point to the example of play in Moldvay Basic, pp B60 and B28:

DM (rolling for wandering monsters): "OK. . . . A secret door . . . in the south wall opens, and two hobgoblins stroll in . . . ."

. . .

Silverleaf steps forward with both hands empty, in a token of friendship, and says "Greetings, noble dwellers of the deep caverns, can we help you?" . . .

The DM decides that Silverleaf's open hands and words in the hobgoblins' language are worth +1 when checking for reaction. Unfortunately the DM rolls a 4 (on 2d6) which, even adjusted to 5, is not a good reaction. The hobgoblins draw their weapons, but don not attack. . . .

The largest of the hobgoblins shouts, in his language, "Go away! You're not allowed in this room!"

"It's okay; Gary sent us," Silverleaf answers.

"Huh?" the hobgoblin wittily responds.

The DM rolls a new reaction with no adjustments. The roll is a 3; the hobgoblins charge.​

There are at least three occurrences of GM improvisation in this example of play.

(A) The GM decides to allow a +1 bonus to the reaction roll. This is not stated in the rules on p B24 ("The DM can always choose the monster's reaction to fit the dungeon, but if he decides not to do this, a DM may use the reaction table below to determine the monster's reaction").

(B) The GM has to twice interpret the meaning of the phrase "Hostile, possible attack" which is the entry on the reaction table for a result of 3 to 5 - the first time the hobgoblins parley, the second time they attack.

(C) When the hobgoblins are parleying but with a hostile orientation, the GM has to make up what they say - in this case, the instruction to the PCs to leave the room.​

And here is some discussion on content improvisation from Roger Musson's essay on dungeon-building in a relatively early White Dwarf (no 27, 1981):

[A]ssuming that you have the corridors and rooms already mapped, there is a very good alternative to improvisation: the Emergency Room Register . . . If players move into an area that you haven't populated, and open a room, select a room randomly from the appropriate list in the Emergency Room Register. . . . It is true that if they had opened the door three down on the right instead of the door they were at, it would have made no different to what they would have found, but as long as they don't know that, it won't hurt them. . . .

I used to keep a goodies bag of unlocated odds and ends, which I would dip into in two sorts of circumstances: one, if players were having such a sad time of it that I actually felt sorry for them; two, if a player searched in a hiding place which was so clever that I wished I'd thought of it myself. Should you follow this practice, never admit to ti. Now that I've admitted it, I shall abandon it. In D&D it isn't necessary to play by the book, but it is essential that the players shall always think you are.​

Basically, Musson is happy with the GM placing a pre-written but as-yet unlocated challenge in front of which ever door the PCs happen to open; but regards GM-improvised rewards as a type of rules-breaking that shouldn't be admitted to. Whether on thinks he is right or wrong in these judgments, it clearly shows that various types of improvisation were taking place before 1990, as well as discussion about the dynamics and proper scope of GM improvisation.

Third, even when playing gamist, dungeon crawling D&D, quite often the players have to engage in reasoning which is not "code-breaking" reasoning, but reasoning about the internal logic of the shared fiction - for example, thinking through motivations, relationships etc of the NPCs and monsters.

An example of this sort of thing figuring in GMing advice is found in another Roger Musson's essay (White Dwarf no 25, 1981):

What are the ingredients that the DM should provide to make his dungeon interesting? If the game is to generate the same interest as a novel, it must have the same ingredients: characters and plot. . . .

The characters generate plots, into which the players may step. The characters and plots together generate the contents of the dungeon. . . .

The true NPC should be as active as the player-characters. If NPCs are to appear credible, they ought to be doing something . . . And it is when the plans and activities of NPCs and those of the player-characters interact that the best games of D&D result.​

This is not advocating storytelling as a goal of play. But it is advocating the use of fictional characters, and their motivations and their deeds, as not just a backdrop to play ("This dungeon was built by a crazed wizard") but as the actual subject matter of play - PCs interacting with NPCs, which means both GM and players reasoning about the imagined lives of these imagined beings. That is obviously not code-breaking and not algorithmic.

Even if a module has no NPCs of note (as is mostly true of C1) the players might still have to reason about the motivations of NPCs to help beat the module - for instance, in C1 it is helpful for players to get a sense of the logic of the place as a tomb/shrine built by an ancient culture. This sort of reasoning isn't purely code-breaking either.

And even if a module has not even a hint of story - eg S2, White Plume Mountain - the players might still have to reason about the fiction in a non-algorithmic/code-breaking way and the GM adjudicate that. In S2 this can include things like ways of flooding the inverted ziggurat room, or surfing doors down the frictionless passage over the super-tetanus pits.

These three elements of D&D - it's potential for story, the need for the GM to improvise, the importance of the fiction as an ingredient of play which is not simply algorithmic in its workings - have been there from the beginning. They didn't just spring into existence when White Wolf started publishing in the late 1980s.

 

[Campbell]

These days I prefer games like Apocalypse World, Circle of Hands, and Burning Wheel over D&D. When I was running D&D 4e I used play techniques taken from those games to drift it towards the type of game I prefer. What I feel often gets missed in the context of these discussions is the fundamental nature of the type of play engendered by these games. I think the way we talk about games like Apocalypse World, and way pemerton, Manbearcat, me, and others played D&D 4e is flawed because it misses the real point of play.

I'll be using Apocalypse World as a stand in here for games that have similar goals of play.

When I'm playing Apocalypse World I am not in it for the story or narrative whatever those things mean. Instead, I'm playing to advocate for my PC who I am required to play as if they were a real person with real wants, real needs, and real goals. When I'm running Apocalypse World I am required to make the Apocalypse World seem real, make the character's lives not boring, and play to find out what happens. Anything I say that is not in service to those three things means I'm cheating the players out of playing the game.

Let's talk about what this actually looks like. As a player, I am playing a character who wants the sort of things that real people would want. That means food, security, emotional connections, a sense of community, whatever. It is my job to go after those things with intent. If I don't I am not playing the game. Now, we wouldn't want these things to be easy. That's where the Master of Ceremonies (MC - read GM) comes in. Their job is to make the world seem real, put obstacles in the way of what my character wants, and have the discipline to not have a stake in the way events unfold.

At its heart Apocalypse World is about players playing real people making difficult decisions in difficult situations. Scene framing is simply a tool used to make that happen. We don't play out the boring bits because that's not what the game is about. Within the scope of play (scene, conflict, whatever) I'm making decisions for my character to get them what they want. Where Apocalypse World differs from AD&D is in scope and goals of play. Rather than the dungeon where I try to get treasure while not being killed by monsters, play exists in emotionally charged situations where I'm trying to get what my character wants. Whatever the differences might be I am still advocating for my PC. From my viewpoint, it is not a fundamentally different enterprise.

The fundamental desire that sparked games like Apocalypse World was the desire to make games that were about something, games that reflected human experience in a meaningful way. They still wanted their games to be games. The basic argument is that you can create emotionally compelling games that are still fun to play.

I'll close with this quote from Apocalypse World that feel sums up the goal of play:

Everything you say, you should do it to accomplish these three, and no other. It’s not, for instance, your agenda to make the players lose, or to deny them what they want, or to punish them, or to control them, or to get them through your pre-planned storyline (DO NOT pre-plan a storyline, and I’m not ing around). It’s not your job to put their characters in double-binds or dead ends, or to yank the rug out from under their feet. Go chasing after any of those, you’ll wind up with a boring game that makes Apocalypse World seem contrived, and you’ll be pre-deciding what happens by yourself, not playing to find out.

Play to find out: there’s a certain discipline you need in order to MC Apocalypse World. You have to commit yourself to the game’s fiction’s own internal logic and causality, driven by the players’ characters. You have to open yourself to caring what happens, but when it comes time to say what happens, you have to set what you hope for aside.

The reward for MCing, for this kind of GMing, comes with the discipline. When you find something you genuinely care about — a question about what will happen that you genuinely want to find out — letting the game’s fiction decide it is uniquely satisfying.

 

[Balesir]

I'm glad this thread has been separated from the previous one, where I think it was essentially threadcrapping somewhat. I worry slightly that the same, time-worn rabbit holes will be plummeted down as we have seen before, but it's an intriguing topic, so let's have another go...

Basically, Musson is happy with the GM placing a pre-written but as-yet unlocated challenge in front of which ever door the PCs happen to open; but regards GM-improvised rewards as a type of rules-breaking that shouldn't be admitted to.

I'm going to nitpick a bit, here , because I think there is an essence to what Musson was trying to get at that is often not recognised - in fact, I'm not at all sure it was recognised at the time, either.

Placing pre-planned challenges before the characters has an important difference from placing pre-planned rewards that relates to player behaviour. A more analogous pairing might be placing (prepared) rewards for things you, as GM, like and placing additional challenges or increasing difficulty for things you don't like. As soon as the players realise that either of these is going on, it may change their behaviour from engaging with the game-world to engaging with the GM in a toadying and sycophantic manner (in ther sense that they start trying to please the GM by whatever route works, not that they begin making comic-book addresses of adulation).

I think this is the essence of what Musson was getting at with the "never let the players find out" comment. What is important is that the players continue to engage with the game world, not with the GM. Whatever techniques are used must not compromise that position.

Third, even when playing gamist, dungeon crawling D&D, quite often the players have to engage in reasoning which is not "code-breaking" reasoning, but reasoning about the internal logic of the shared fiction - for example, thinking through motivations, relationships etc of the NPCs and monsters.

I am rather more sympathetic to [MENTION=3192]howandwhy99[/MENTION]'s general point on this one. Code-breaking, at least in the pre-computer age, was about divining what the user of the code was trying to communicate in some language. Informed reasoning about their intent and situation was thus part of the "code-breaking". The entire enterprise of science is, in a similar sense, "code-breaking"; it is an attempt to work out what is going on "behind the screen". If certain assumptions about (NPC) motivations and intentions cause other observations in the game to "make sense", then it is not unlike looking at a coded message and finding that assuming a certian sequence of letters to represent a particular port makes other things in the code fall into place. "Decoding" people's motivations is indeed, I would say, a form of "code-breaking". The problem may arise, as I mentioned above, when it ceases to be the NPC's motivations that are being decoded and becomes the GM's motivations that are the topic of investigation.

Even if a module has no NPCs of note (as is mostly true of C1) the players might still have to reason about the motivations of NPCs to help beat the module - for instance, in C1 it is helpful for players to get a sense of the logic of the place as a tomb/shrine built by an ancient culture. This sort of reasoning isn't purely code-breaking either.

And even if a module has not even a hint of story - eg S2, White Plume Mountain - the players might still have to reason about the fiction in a non-algorithmic/code-breaking way and the GM adjudicate that. In S2 this can include things like ways of flooding the inverted ziggurat room, or surfing doors down the frictionless passage over the super-tetanus pits.

I left these together since I think they are both, at root, examples of the same failing in RPG setup. That failing is cultural code assumtion. In other words, they assume that particular trigger terms or concepts hold specific meanings to those trying to decode the situation; specifically, they expect (or even require) that the GM and players have the same cultural associations and assumptions about particular concepts.

The first assumes that the players have a specific set of assumptions about what "a tomb/shrine built by an ancient culture" might be expected to include. Why such assumptions, based as they will almost certainly be on the history of this human-inhabited world, should predetermine the valid assumptions in a world with a myriad of fantasy races is anyone's guess.

The second assumes that "friction" (and, indeed, "frictionless") have implications understood by those educated in the modern western scientific paradigm - again, in a fantasy world in which "magic" is assumed to be a valid part of the workings of the universe.

These are a bit like puzzles commonly referred to as "cryptic crosswords". They only really work as valid puzzles if you happen to know a cute little set of code phrases and assumptions concerning how these particular puzzles work. They are, in a sense, "code-breaking", but it is assumed that the basic ruleset underlying the code is known to the person attempting the breaking. If it is not, then they will find the patterns of the code meaningless (not to mention, most likely, totally uninteresting).

 

[Manbearcat]

But Howandwhy99's assertions go far beyond this, and that is what makes them false.

<snip>

Third, even when playing gamist, dungeon crawling D&D, quite often the players have to engage in reasoning which is not "code-breaking" reasoning, but reasoning about the internal logic of the shared fiction - for example, thinking through motivations, relationships etc of the NPCs and monsters.

<snip>

These three elements of D&D - it's potential for story, the need for the GM to improvise, the importance of the fiction as an ingredient of play which is not simply algorithmic in its workings - have been there from the beginning. They didn't just spring into existence when White Wolf started publishing in the late 1980s.

The problem may arise, as I mentioned above, when it ceases to be the NPC's motivations that are being decoded and becomes the GM's motivations that are the topic of investigation.

Putting these together as they dovetail with my position on this.

I am understanding of (but not sympathetic to) the purity (with respect to decoding the causal logic of the system and then making informed extrapolations) that howandwhy has been advocating for some time. However, my problem with his thesis is the disproportionate signal to noise ratio of a classic D&D game. It makes consistently sussing out information and then using it in a wieldy fashion (to make predictions/extrapolations with a high level of confidence) a losing proposition a fair portion of the time. There seems a built-in assumption on his part that a skilled GM is capable of moving those proportions deeply in favor of the signal. I agree that part of the job of a quality GM is (a) maintaining a level of internal consistency (temporally, spatially, continuity, et al) despite the deep abstractions inherent to the system and (b) properly telegraphing/foreshadowing (either at the build stage or impromptu in the moment as required). However, there is a hefty portion of gameplay (both in the content generation/"introducing elements into the shared imaginary space" phase and in the long-term maintenance of continuity phase) that is subject to factors external to the GM (eg various player input and varying degrees of attentiveness by all of the tables' participants).

Finally, (and likely most important) there are a large number of cognitive biases that the GM may unwittingly introduce into the fiction (again either at the build stage or the impromptu stage) that in turn introduces noise which must be sifted through and/or may compound prior established noise. There are a lot of physics and biological anomalies (or utter incoherencies) embedded into the genre of a D&D setting. GMs regularly handwave these or exacerbate them (by introducing further incoherence due to extrapolating from them) due to their complete misunderstanding of how mundane phenomena/interactions work in the real world. How is a player who is bent on tightly constraining causal logic (rather than just going with it and using genre logic) supposed to "break this code?"

When I'm playing Apocalypse World I am not in it for the story or narrative whatever those things mean. Instead, I'm playing to advocate for my PC who I am required to play as if they were a real person with real wants, real needs, and real goals. When I'm running Apocalypse World I am required to make the Apocalypse World seem real, make the character's lives not boring, and play to find out what happens. Anything I say that is not in service to those three things means I'm cheating the players out of playing the game.

Let's talk about what this actually looks like. As a player, I am playing a character who wants the sort of things that real people would want. That means food, security, emotional connections, a sense of community, whatever. It is my job to go after those things with intent. If I don't I am not playing the game. Now, we wouldn't want these things to be easy. That's where the Master of Ceremonies (MC - read GM) comes in. Their job is to make the world seem real, put obstacles in the way of what my character wants, and have the discipline to not have a stake in the way events unfold.

Part of the beauty I have found in recent game design is the honesty and transparency (of which Baker was at the forefront of). Honesty about the implications of a singular GMing technique/resolution mechanic/play procedure on the aesthetics of play and on the participants' disposition toward one another, honesty about the complete requirement of the usage of genre logic and the (sometimes tangled) mesh of it with causal logic, transparency about where and how abstraction aids in the facilitation of functional play, how/when/why to properly telegraph, how simple system coherency can support a paradigm rather than forcing the players to fight against/work around the system.

Most of it puts the RPG issues of "Every GM, regardless of how amazingly proficient they might be, is fallible so how can we help reduce their cognitive workload during play?" and "how do we design games so we can consistently say that we got what we were looking for/expecting after an RPG session" at the forefront. Those are admirable and proper design goals in my opinion.

 

[howandwhy99]

[MENTION=42582]pemerton[/MENTION], for the sake of example, I'll use Tic-Tac-Toe again to describe the differences between games, puzzles, and stories.

Tic-Tac-Toe is a game when someone seeks to accomplish an objective within its design.
It is a puzzle when someone seeks to solve its underlying pattern by deciphering it.
It is a story when someone uses it to communicate a message.
--

First, all games are pattern designs. This isn't under debate. It is the base entry point into game culture making it different from others like narrative culture.

As I explain in the example above, to play a pattern as a game is to actually attempt to achieve objectives within it. In the case of D&D this is done by proxy. In order for this to happen successfully a player must attempt to decipher the design which is a consequence of taking actions. However, most game play happens for the player in their mind as they decipher out their own perception of the pattern before moving the game. This is called strategy, the act all games are designed to engender.

Second, improvising a story cannot occur during the playing of a game because improvised storytelling is an act of invention, while game playing is the act of discovery through interaction. Storytelling will always be limited to resulting in a pattern after the fact. Playing a game means deciphering a pattern in the moment. Therefore, the act of reading is like playing a game, while expressing interferes with it.

Third, all games must have an actual existing pattern in place prior to play. This is the game people will be playing. This need becomes obvious when we remember a game cannot be played without rules in use throughout, the mental pattern accompanying that of the field of play.

As I believe you probably already know, a pattern is only a game for us because we are understanding it as such. This is exactly the same as how sequential patterns are often considered narratives because people willfully comprehend them to be, not because they actually are so. We are engaging with things outside ourselves, but limiting ourselves to a finite set of ideas when addressing them as either a game or story (or anything else).

To clarify, story and game cultures are two separate sets of ideas affixed to actualities by people currently understanding them as such. In neither case are these objects (game boards, pieces, mental processes, etc.) actually "truly" a game or story.
--

Now to your points,

1. Since its publication many people have not played D&D as a game, and since the 80s en masse as a functional game. That is well known. That you wanted to tell a story and yet play a game, that you continue to confuse the two acts does not couevidence of your position. On the contrary, it is evidence of your confusion. Treating a game design as not a game is not proof it isn't a game.


2. A referee in D&D moves, measures and relates portions of the current state of the hidden game to those players who are playing it. They are not players themselves as they do not have any playing pieces on the board. Like any referee they are not there to create anything, only to generate out via calculation results. Specific to D&D, these results include, but are not limited to, To Hit dice roll results, Saving Throw dice rolls, Ability Score dice rolls, and yes, map creation dice rolls.

At no point are referees to interfere with the game, as you say "improvise" by moving stuff around, removing or adding pieces as not directed to under the rules. This is paramount for every ref. The result of doing otherwise is akin to gaslighting the players and discounting the game.

In other words, not only does no GM ever need to improvise during a campaign, they are interfering with the game if they do. That Gary left many mechanical necessities of the game up to DMs to determine prior to play doesn't change this fact. After all, most of what he wrote wasn't known game rules, but suggestions for generating the hidden design to be gamed. This covers all the examples you give as neither of us know what mechanics the GM in your examples was using to determine the results.

As to Mr. Musson's bad advice, see my response to 1. Evidence of someone not running D&D as a game is not evidence D&D isn't one. And I'd assume Mr. Musson would be happy with not randomly rolling monsters to populate his dungeon and wandering monster tables too thereby removing the game of players seeking out monsters, avoiding them, etc. as well as playing the dungeon as a maze.


3. First off, let's drop The Forge "narrative theory is dogma" schtick.

--There is no such thing as "gamism". - The last thing anyone needs is more "isms" in their life. Treat a pattern as a game you are playing a game. Treat it as something else and you're engaged in something else.

--"Shared fiction" never occurs in any game. - There is no such thing as actual fiction or nonfiction except as one culture's ideas of how they understand their world. In stark contrast, games require actual designs which must exist throughout the playing of a game. Remove the design and a game can't be played.

--There are never "fictional characters" in games. - What I believe you are thinking of are game constructs meant to be gamed.

So, FYI, NPCs and their behaviors as contained within their statistical design just like every other game component. They can be gamed through code breaking --the act of mastering a game-- and manipulating the game design. These statistics are largely in AD&D books, but mechanics like reaction rolls, alignment charts, racial relations, morale, loyalty, and plenty of other bits and pieces throughout the early published games do exist. But they are limited as to what the cover, I agree. You desire more depth of pattern complexity within the game subsystem these rules cover. That's admirable, so do I. But this does not mean NPCs aren't pieces of games. It simply means not much in the way of mechanical suggestions were published.

The example of reasoning you mention sounds like players attempting to decipher the social situational patterns of your story creating and treat them like a game. As stories being created are non-repeating narrative sequences being formed into a pattern it is impossible to game them. At best, stuff like that should only be done ironically.

And I see Mr. Musson is clearly misguided about D&D. "Make a plot for the players to follow" when refereeing a cooperative strategy game. That's a horrible idea.

So, all your examples about NPCs refer to game components not improvisation. Or with S2 to the generated game board. All these things must be on the GMs map behind the screen and tracked by them just like any referee running any other game. That was in the rules until the deluded people publishing 2e moved the rules in front of the screen and turned D&D into a half-assed broken game.

These three elements of D&D - it's potential for story, the need for the GM to improvise, the importance of the fiction as an ingredient of play which is not simply algorithmic in its workings - have been there from the beginning. They didn't just spring into existence when White Wolf started publishing in the late 1980s.

Well you're just trying to persist a falsehood. Remember, White Wolf declared their product to not be about gaming, but a game about storytelling. It split the hobby through the 90s and since then that crew of storytellers have not only attempted to rewrite history and the other side out of existence, but they are attempting to write games out of existence too. Heck, gaming and roleplaying are practically the opposite act of storytelling. Just read any history book (that hasn't been whitewashed by critical theorists practicing a highly suppressive variety of narrative absolutism).

 

[Zak S]

Apocalypse World.... I'm playing to advocate for my PC who I am required to play as if they were a real person with real wants, real needs, and real goals. ... I am playing a character who wants the sort of things that real people would want. That means food, security, emotional connections, a sense of community, whatever.

That's funny because I tend to like D&D better than AW for exactly the same reason.

I can't imagine a walking around needing "emotional connections" or "a sense of community" (those things are free and always have been, even in the post-Apocalpse) but I know I don't want to be killed--especially by spider priests.

When I play AW, I feel like I'm being placed in a position where I have to pretend I care about things I so don't care about like whether a made up community has an interesting-enough looking food supply. ("We keep snakes in a sealed-off ice cream truck and breed them and eat them!" "Cool! Way to push the story forward!" Dude I could do this all day and still not care.)

In D&D I just don't want to die. Relatable.

At its heart Apocalypse World is about players playing real people making difficult decisions in difficult situations.

(that's every RPG)

The fundamental desire that sparked games like Apocalypse World was the desire to make games that were about something, games that reflected human experience in a meaningful way. They still wanted their games to be games. The basic argument is that you can create emotionally compelling games that are still fun to play.

No, the fundamental desire that sparked games like AW is everybody else was playing games like D&D and felt they already "meant something" and "reflected the human experience" and were "compelling" and the authors of those games just didn't feel that way yet because they didn't relate to the way the resolution system related to these things.

So they made a different game.

 

[pemerton]

I'm going to nitpick a bit, here , because I think there is an essence to what Musson was trying to get at that is often not recognised - in fact, I'm not at all sure it was recognised at the time, either.

Placing pre-planned challenges before the characters has an important difference from placing pre-planned rewards that relates to player behaviour.

<snip>

As soon as the players realise that either of these is going on, it may change their behaviour from engaging with the game-world to engaging with the GM

<snip>

I think this is the essence of what Musson was getting at with the "never let the players find out" comment.

That seems like a plausible reading of his comment.

Code-breaking, at least in the pre-computer age, was about divining what the user of the code was trying to communicate in some language. Informed reasoning about their intent and situation was thus part of the "code-breaking". The entire enterprise of science is, in a similar sense, "code-breaking"; it is an attempt to work out what is going on "behind the screen". If certain assumptions about (NPC) motivations and intentions cause other observations in the game to "make sense", then it is not unlike looking at a coded message and finding that assuming a certian sequence of letters to represent a particular port makes other things in the code fall into place.

<snip>

I left these together since I think they are both, at root, examples of the same failing in RPG setup. That failing is cultural code assumtion. In other words, they assume that particular trigger terms or concepts hold specific meanings to those trying to decode the situation; specifically, they expect (or even require) that the GM and players have the same cultural associations and assumptions about particular concepts.

<snip>

These are a bit like puzzles commonly referred to as "cryptic crosswords". They only really work as valid puzzles if you happen to know a cute little set of code phrases and assumptions concerning how these particular puzzles work. They are, in a sense, "code-breaking", but it is assumed that the basic ruleset underlying the code is known to the person attempting the breaking. If it is not, then they will find the patterns of the code meaningless (not to mention, most likely, totally uninteresting).

With these, on the other hand, I want to counter-nitpick.

At this point, "code-breaking" seems to have just become a synonym for "reasoning".

I agree that the frictionless corridor in WPM assumes that the players have some sort of contextual familiarity with a particular trope/idea. The same is true of all the old chess rooms, the metal/electricity/magnet/lodestone tricks and traps, and so on. Plus many door traps and pit traps (some of the more baroque examples of latter even depend on familiarity with the particular FRPG-ism of the pit trap).

But they still require GM improvisation or non-algorithmic adjudication: for instance, if you write some sort of wacky electricity trap into your dungeon, when the players start trying to circumvent it by wrapping their hands in (hopefully insulating) cloth or using 10' poles or whatever, you are going to have to deal with those efforts.

Some of that involves shared tropes, and to that extent from the player side can degenerate into "playing the GM". But a lot of it is about understanding and adjudicating fictional positioning - eg if the players are going to use a 10' pole wrapped in cloth as an insulated trap-prodder, the GM has to adjudicate how much cloth they can get from tearing up their cloaks, how hard it is to wrap it around a 10' pole etc.

In the thread I forked from, [MENTION=4937]Celebrim[/MENTION] was describing this as the "freeform" aspect of an RPG. The way I think of it is this: in a RPG, unlike in a boardgame, fictional positioning matters. And sometimes - in fact, often - adjudicating the fictional positioning requires non-algorithmic judgments on the part of the GM.

I think an interesting and hugely important feature of the Vincent Baker-influenced games that [MENTION=16586]Campbell[/MENTION] has mentioned (I don't play much "powered by the Apocalypse", but am currently running a Burning Wheel campaign) is that they have found a way to try and regiment the adjudication of fictional positioning: namely, it is factored into the framing of action resolution, as an input into a relatively structured and abstract system of action resolution (eg in BW, it forms part of the negotiation around obstacles and advantage dice as well as intent-and-task), and then the actual outcome is determined purely by rolling the dice to chooses between the player's preferred outcome and the GM's alternative narration of failure.

In my view (and somewhat ironically, given that [MENTION=3192]howandwhy99[/MENTION] derides them as non-game storytelling), this is a huge part of how these games are able to have the sort of focus-on-the-fictional-stakes aspect that Campbell describes, without degenerating into either playing the GM or adversarial GMing. They are much less hostage to that sort of degeneracy than were earlier systems that hadn't benefitted from this design insight.

Classic D&D certainly didn't have this sort of apparatus for regimenting the adjudication of fictional positioning, and (but for a brief attempt with 4e's skill challenges) it still doesn't. So GM adjudication of fictional positioning not just in framing but in actually determining outcomes remains pretty crucial. From the GM side, this is why advice repeatedly stresses the need for the GM to be neutral - the GM should be trying to reason within the fiction without fear or favour towards the players. From the player side, this isn't code-breaking in any special sense. It's just reasoning about the fiction as if it were real.

I am understanding of (but not sympathetic to) the purity (with respect to decoding the causal logic of the system and then making informed extrapolations) that howandwhy has been advocating for some time. However, my problem with his thesis is the disproportionate signal to noise ratio of a classic D&D game.

<snip>

Finally, (and likely most important) there are a large number of cognitive biases that the GM may unwittingly introduce into the fiction (again either at the build stage or the impromptu stage) that in turn introduces noise which must be sifted through and/or may compound prior established noise. There are a lot of physics and biological anomalies (or utter incoherencies) embedded into the genre of a D&D setting. GMs regularly handwave these or exacerbate them (by introducing further incoherence due to extrapolating from them) due to their complete misunderstanding of how mundane phenomena/interactions work in the real world. How is a player who is bent on tightly constraining causal logic (rather than just going with it and using genre logic) supposed to "break this code?"

All true.

For present purposes, what I want to point out is that extrapolating from causal logic is not "code-breaking" in any meaningful sense of that term. It is just reasoning - what I have called reasoning about the fiction as if it were real. The idea that this can take place without the GM having to engage in improvisation is completely implausible.

Just to stick to my insulated-by-wrapping-a-10'-pole-in-torn-garments-electricity-trap-prodder: even if we gloss over Manbearcat's point about physical/biological anomalies so as to accept that the GM and players are on the same page about how electricity traps and insulated prodders of them work, there is still improvisation required. Eg the GM might have to decide the % chance that the insulating cloth comes unwrapped, that the whole thing catches fire from a spark resulting from arcing electricity, etc. If the players have declared that they wet their cloth in water so as to dampen the prospects of fire on the pole, then the GM further has to improvise the balance between the fire-dampening properties of wet cloth and its increased electrical conductivity.

The game doesn't, even in principle, have any rules for making these adjudications, and can't - because there is no in-principle limit on the number of moves that the players might make by exploiting the fictional positioning of their PCs.

This is why I regard the invention of abstract scene-focused resolution - which uses fictional positioning only to shape the details of the dice pool and the consequences that the dice choose between - as such a breakthrough in RPG design. It's very hard for me to envisage going back from that.

 

[howandwhy99]

However, my problem with his thesis is the disproportionate signal to noise ratio of a classic D&D game. It makes consistently sussing out information and then using it in a wieldy fashion (to make predictions/extrapolations with a high level of confidence) a losing proposition a fair portion of the time. There seems a built-in assumption on his part that a skilled GM is capable of moving those proportions deeply in favor of the signal.

In D&D, it isn't the skilled GM so much as the design of the game which keeps signal to noise ratio low. The game is built so abstractions are only at the most base part and likely never to be encoutnered. I'm willing to bet everything you understand as an abstraction in D&D isn't so at all, but a score meant to be assigned to an aggregated design which enables simplification of common results just waiting for players to pick it apart through in game ingenuity and find the game inside.

I agree that part of the job of a quality GM is (a) maintaining a level of internal consistency (temporally, spatially, continuity, et al) despite the deep abstractions inherent to the system and (b) properly telegraphing/foreshadowing (either at the build stage or impromptu in the moment as required). However, there is a hefty portion of gameplay (both in the content generation/"introducing elements into the shared imaginary space" phase and in the long-term maintenance of continuity phase) that is subject to factors external to the GM (eg various player input and varying degrees of attentiveness by all of the tables' participants).

Both tracking player additions to the design during a session and game content generation prior to each session can add up. But as long as the code that is the game is both relatively small, yet widespread in its coverage the work is simple math. However, I find generation requirements are primarily kept low because prep and play are all done in small portions at a time. Each session's game generation prep results in more accumulated design for players to play, but each actual play session only covers whatever small portion players can manage to mess with given real world time limits. So not only is the prep amount largely constant (though extraordinary teams could be considered more burdensome), but very highly complex games are made possible to challenge players at higher levels later in the campaign.

Finally, (and likely most important) there are a large number of cognitive biases that the GM may unwittingly introduce into the fiction (again either at the build stage or the impromptu stage) that in turn introduces noise which must be sifted through and/or may compound prior established noise. There are a lot of physics and biological anomalies (or utter incoherencies) embedded into the genre of a D&D setting. GMs regularly handwave these or exacerbate them (by introducing further incoherence due to extrapolating from them) due to their complete misunderstanding of how mundane phenomena/interactions work in the real world. How is a player who is bent on tightly constraining causal logic (rather than just going with it and using genre logic) supposed to "break this code?"

Thank god for Gary and random tables. These tools not only organize everything, but randomly rolling on them removes huge amounts of potential, as you say, cognitive bias and you never know what you're going to get. And really, in depth accounting for physics and biology designs by a multidimensional cellular automata code are bread and butter for the game.

 

[N'raac]

Actually, I see a lot of players, who I will suggest view the game much like howandwhy99, who have no intention of playing a "real person" making difficult decisions in a difficult situation. Rather, they are playing a cipher, a plastic playing pawn which attempts to adopt the most tactically advantageous approach to every situation.

These are the players whose characters, faced with the choice of accepting a d6 roll on which 1 means "gain substantial wealth" and 6 means "die horribly" will keep rolling until one of the two results is attained, as "vast wealth" will provide the player a pawn with more power in the game, while "horrible death" just means he starts out a new pawn. How many "real persons" want to sign up for Russian Roulette? Their pawn is not a "character" - it has no emotions, no human flaws or foibles, no principals or goals, beyond "maximize advantage in game".

No thanks. I can play the Dungeon board game, the D&D board game, or any number of other games where I ruthlessly seek mathematical advantage to win, under the terms of victory set out by the rules. To view D&D purely as a tactical exercise makes it a "game", but D&D is not solely a game. It is a Role Playing Game. Remove the "character" from the character, and it is just a game. A 'G', not an RPG.

The GM does not role play the PC's - that is the role of the players. But it is his role to place the characters in situations where the characters are challenged. If we reduce this to mere 'G', we challenge only their tactical skills, and the powers of the characters. But a good GM will also challenge the players to role play their characters - what is more important to him, protecting the innocent or destroying the evil? Will he sacrifice his principals for power or wealth? What happens when the stereotypical threatened village is brought to life as a PC's childhood home?

In a pure 'game' view, my character (PAWN) should be an orphan with no real principals or objectives beyond personal safety, power and wealth. The GM will use any ethics, principals, personal connections, etc. as leverage to prevent the achievement of 'winning' the game. In an RPG, the GM will use those character flaws, foibles and background elements to bring life to the characters and the game. How much more satisfying to not only achieve victory, but to do so without compromising the character's principals.

In a pure game, the choice will be "lose or sacrifice your principals". In a good RPG, the challenge is often not merely to win, but to win within those principals.

In a game, the constraints of a Paladin's code will disadvantage him constantly. In an RPG, challenges to his principals and ethics will be as, or more, enjoyable and challenging as challenges to his strength at arms, and his sterling reputation and past acts of charity and benevolence will return to his benefit in unexpected ways.

If I wanted a board game, I'd go to the closet and get a board game. I play an RPG for a different purpose.

 

[Zak S]

This is why I regard the invention of abstract scene-focused resolution - which uses fictional positioning only to shape the details of the dice pool and the consequences that the dice choose between - as such a breakthrough in RPG design. It's very hard for me to envisage going back from that.

This is baffling:

You like mechanics that blur out fictional positioning advantages to a die advantage because why? Otherwise the player and GM might have to agree on the physics of the game? Otherwise the player might have to trust the GM they agreed to spend hours of their life playing with?