A Pool

Let's consider our party.  For those who may be new, this series of posts have been thought experiments on what we can do as DM, given a theoretical action of the party.  As discussed in the linked post, our faux 1st/2nd level party has been attacked by a huge spider, has discovered a warm thermal side passage, and learned it contains at least one additional huge spider.

Let's say the party chooses against this path.  Instead, having come this far in the colder cavern, they decide the alternative is a pointless side quest, meant to draw them away from their primary goal.  That was to find the "demon" within the dungeon (not knowing precisely what it is) and clues to the disappearance of the local noble's father.  For those who have been following along, you can see how the campaign has become this enormously confusing, dense, yet suggestive series of clues and facets ... while you've been here all along as each has been introduced to the party.  As a DM, I know my players will call out things along the way, wanting to know how each thread fits into the whole.  Because I've been running games this way for decades, I'm able to answer in some manner, or explain how because they never did confront a certain threat, or go into a certain part of the dungeon, that mystery remains unsolved.

But I digress.  Obviously, I need to write another post that links all the threads on this series to date.

You may remember that one of the dangers of the dungeon that I mentioned in a post two weeks ago were miniature ochre jellies.  Having eschewed the thermal vent, the party is nonetheless moving adjacent to it and therefore the caves here are slightly warmer to those where the lake witch resides.  I trust you're all following along.  Questions in the comments, please.

I've set up this post to explain an encounter where the party has no warning of what's coming, which is quite different from the fight with the witch's minions.  Imagine a cavern-passage that slopes downwards at about ten degrees.  In shape, it varies between 20 and 30 feet wide, but only five of six feet high ... the floor is crumpled, but made smooth by the dampness.

As it levels out, the ceiling rises a few feet, but then the passage submerges itself into a pool.  This fills the passage from side to side; the water's far edge can't be seen properly with a torch, but a bullseye-lantern hints that the far edge might be 70 feet away.  A good light spell, if the party has one and casts it to illuminate the whole tunnel, confirms the far edge is 75 feet away.  From there, the passage slopes upwards and out of sight.

The water is perfectly still.  It's black in colour.  The players test it with a staff and find that it's but a few inches deep, as far as a staff can reach.  A longer pole shows that it gently deepens to a foot or so.  Actually entering the water reveals that it has a solid rock bottom and never gets more than four feet deep, before climbing out to the other side.  This last, of course, isn't told to the party unless they can somehow learn it by experimentation or with a spell of some kind — the sort of spell a 1st or 2nd level caster wouldn't have.

Throwing stones into the water turns up no visible monsters.  Nothing appears to be swimming around.  The surface reveals no fins nor movement of any kind, other than what the party makes.  But then, a jelly wouldn't, would they?  They don't "swim."  They don't move fast.  Their bodies are so watery, with the upper part soft and flat, that the top of one wouldn't break the surface tension.  As the water is disturbed, they merely float, their bodies matching the ripples.  As they don't produce their own light, they're as black as the water.  But again, we don't explain any of this, not until the party comes into contact, or detects them in some other than disturbing the water.

Is it fair?  The party can't see the monster.  And naturally, they don't trust the water.  Ten thousand cavers a season wade through pools like this without a care, but they don't do it in a D&D world.  It might just be a pool.  But naturally, the players are cautious.

Gygax was extremely fond of the surprise and slash encounter.  Player walks under a part of the ceiling, spider conveniently drops, surprise! and attacks.  This same motif appears everywhere in serial fiction beginning with the 19th century and going right up through Tarzan movies of the 30s and 40s, and into monster films of the 50s and 60s ... just the sort of fare that Gygax watched on creature feature afternoons when he was a little boy.

But constant encounters like this become tiresome, especially from the party's perspective.  All too often, some player is bound to say upon coming to the water, wearily, "I ready my bow for the moment the monster leaps out of the water by surprise."  Cinematically a jump scare might work; but in an imagination-based game structure, it lacks the shock factor.  I can rise slowly in my chair, glare momentously at a member of the party, holding the pose for ten or twelve seconds as the party waits for me to speak — and then SNAP! my hand out and grab a nearby player who I'm not staring at, with a shout, or suddenly slap my hands on the table.  Done right, it's effective.  It's an old carnival trick.

Sadly, this shit only works once with the same audience.

Myself, I like a slow build.  The players are at odds with themselves here.  On the one hand, the party could just refuse to enter or explore it, but it's a bit humiliating to be turned back by what might be an empty pool of water.  This I've also done.  It can be satisfying for the DM, with parties who do turn back, to remark, "Yes, brave Sir Robin turned about, and gallantly he chickened out ..."  Even if it doesn't convince the party to try the pool after all, things like that get in their heads and primes them for the next time they run away from "danger" that hasn't been proved to exist.

So, what's it to be?  These are little bitty ochre jellies, one-sixth the size of the traditional monster (p.75), so they're not going to "jump out" — and let's clear it up further by saying these things don't swim fast.  So the danger is ...?

There's a goodly number, though they're not all in the same spot, they're scattered all over.  If a character moves slowly and cautiously, the jellies will slowly congregate until he or she is surrounded by eight of the devils.  I would run it thusly:  for the first 3 rounds, the slow moving player encounters nothing.  Then, one attacks and if it hits, the player feels a sting for 1-2 damage.  No idea what caused it.  If the creature misses, the player merely feels something brush a leg or an arm.  The 2nd round, there are two jellies; the 3rd round, three; the 4th round there are five and in the 5th round, eight.  No surprise is rolled, nor initiative, because until he or she is touched, the player can't see to win initiative.  The jellies get it automatically.

This is what happens if the character, say the thief, moves slowly.  If he then rushes pell-mell back to the party, or through to the other side, the churning of the water protects him, and there are no more attacks.  But if he tries to hit back, slapping his sword at the water, the attacks increase as described above.

There isn't enough damage here to call the encounter "unfair."  It's really more of a puzzle than a monster.  But players get very stupid about monsters, and don't think of them as "puzzles."  To a player, the weapon is a hammer and every monster looks like a nail.  Even if one player does figure out that the secret is to rush through the water, it doesn't mean others in the party will agree.

There's a good chance someone will try to burn the top of the water with oil.  If so, they'll waste their oil but it won't kill the jellies.  It might be effective against a large ochre jelly in a dry hallway, but these will be insulated against the fire and will merely descend deeper into the water, rising again when the fire subsides.  Burning oil in water is only dangerous to creatures that are partly above the water, where the oil can cling to their bodies and oxygen can feed the flames.

As explained above, situations like these can be made very complicated by a party, who is bound to overthink everything, especially an apparently empty pool of water occurring in an underground cavern where water nearly always pools atop hard stone.  We could make the pool completely empty and the party would still waste a half-hour of game time on it ... and never be completely sure it really was empty.  Because this is what parties are like.

I hope it comes across that our goal is to give it the best description we can, and then let the party waste all the time they want on it.  It's a good time for the DM to get up for a bowl of ice cream.

The DM Helps Those Who Help Themselves

Finishing the trio of posts including this one and this, we've represented the factions of the party and their enemy, and we've lightly addressed the dungeon setting, also notably with this post.  This leaves us the other thingies wandering around, that the party is liable to meet.

But first, unfortunately, we have to address the dungeon, and D&D as a whole, as a game.

Whatever patterns exist within the setting, intended to immerse the players in a fantasy world to give the experience of being there and making decisions, it cannot be forgotten that as a game, success for the players is defined by a set of arbitrary, practical "economic" goals.  Take this from Game Mechanics, Advanced Game Design by Ernest Adams and Joris Dormans, quoting Jesper Juul:

"A game is a rule-based system with a variable and quantifiable outcome, where different outcomes are assigned different values, the player exerts effort in order to influence the outcome, the player feels emotionally attached to the outcome, and the consequences of the activity are negotiable."

Comprehension of how to dungeon master D&D relies upon (a) understanding what the outcome is; (b) how and why it's assigned the value it has; and (c) the steps we have to take as a DM to make certain the players are able to influence the game's outcome, without losing that attachment or the game's negotiable quality.  This is all the harder because D&D has multiple, independent outcomes, each of which can be strived for by the players.

For example, in our scenario, are the players here for the experience, and therefore levels?  To obtain an understanding of this dungeon's nature, and why it's resisted discovery for 25 years?  Is it the treasure?  Is it the sheer emotional process of gaming?

Yes.  To all the above.  But that doesn't mean the above are equal in the party's mind, and especially between players.  As a player, at 1st to 3rd level, I'd be primarily concerned with experience, to a degree of about 90%.  I'm not going to get much from the emotional process, because I've been gaming for so long that I'm jaded.  The mystery carries little weight, as I assume all will become clear once the proper steps are taken, and I can wait for that marshmallow rather easily.  And treasure is a means towards experience.

Later on, when I have sufficient power to carry out my own program, I'll steadily grow less interested in experience and more in the achievement of other outcomes.  I like that D&D is especially flexible in this way.  It's not a one-trick pony, where grinding is the only way to advance the character.  I like that I can choose which outcome, in this moment, matters most to me, while making room to enable other players in the game the opportunity to achieve their desired outcomes, all at the same time.

It's the primary reason why player-vs.-player is game-toxic; it makes it impossible for multiple players to achieve multiple positive outcomes simultaneously.  In essence, it subverts D&D's most powerful achievement as a game.  Want to compete?  Play a competition based game, not D&D.

As a DM, it's part of my responsibility to be aware of, and facilitate to a degree, the player's desired outcomes.  This, however, it a tremendously fluid and indistinct mandate.  I don't want to give the players what they want; but at the same time, I don't want to purposefully withhold it from them, either.  My choices, and my actions, decides how difficult it is for the players to achieve their goals ... and as it happens, making the wrong choice, on either end of the spectrum of "too easy" and "too hard" decides absolutely whether or not the game I'm running is playable.

I've covered "playability" at length, but as it's been some years, let's briefly review, from the words of Ian Bogost:

"And there's something deeply abhorrent about games, something kind of revolting ~ but then out of that revulsion comes sublimity. Occasionally. And it's not just true for games. When you operate a mechanism, like a steering wheel, we sometimes talk about the 'play' that's built into that system, and has a space through which the steering wheel can be turned before the shaft couples with and turns the pinion at its end, or you can find this elsewhere, too. The play of light, the play of the waves, a play on words — and the game designers Katie Salen and Eric Zimmerman have adopted this sense of play in their formal definition of the concept, which is one that I like a lot; free movement within a more rigid structure."

Interestingly, the Game Mechanics book above also quotes Salen and Zimmerman.

It's key here to understand that the right point between "too easy" and "too hard" isn't a single place on the spectrum.  The game wavers from one extreme to the other, so that at some point it has to be easier for the players to achieve their desired outcome, while at others it has to be near impossible.  Comprehending this, and when to make a given situation go this way or that, is absolutely essential to one's skill as a dungeon master.

Though it must be said that nearly everyone associated with D&D and other role-playing games, outside a few game designers and book writers, are woefully ignorant to this.  Magical-thinking idealism tends to reign out there in the wide world: notions like "the DM is always right" ... or "the players are always right" ... or "that the rules have to be played as written" ... are really more dogma than strategies for providing good game play.  No one is "always" right, not even the rules.  The game is far too complicated — and wants to be complicated — for any single fixed policy to meaningfully address all the problems that will arise in a single session, much less an ongoing campaign.  Not only are we forced to gloss over game issues for which no rule has been written, the motives and beliefs of any one human being is a movable feast of the first order.

Yet the endless parade of content makers on youtube, reddit and various podcast platforms continue to espouse arguments based on premises like, "if you do this, everything will be fine."  Um, no.  Nothing will ever be fine.  I've been running games for 44 years (and this blog is just past 15 years old now), and very often emotions get so high that counselling inter-party disputes is as much of a game skill as describing a setting, memorising the rules or rapidly adjudicating combat.

As such, when I see someone arguing that "moving the goalposts" in gameplay is a no-no, it's easy to see a lack of understanding in how D&D calculates outcomes and the player's exertations.  D&D is not football.  One such "goal post" is the monster that has to be killed to achieve the outcome ... but the monster gathers intel during the contest, and the monster moves, and the monster's motives can also change, while the players' mistakes adjust continously how far that particular goal post is from their current location/success.

The pre-made dungeon counters all this, by having the monster located in room 46b, or at the most roaming between rooms 42 and 46.  Whereas I'm describing a game in which the monster's present location in 46b is entirely changeable at any given moment depending on how the players respond post-room 1.  Thus, the program of how the dungeon setting runs is wholly different from that of the games I see being played online, or anywhere, except apparently at my game table.

Suppose, for example, with the last post (linked above), the players do something wholly unlike a typical D&D party when encountering the 15 mooks.  Suppose they run immediately, suspecting the encounter isn't what it appears to be.  Suppose they fight for a round or two, determine the mooks are fairly weak, and then run at that point?  How does that influence the manner in which the Witch below (who is unknown to the players) carries out her campaign?  Does she decide the party isn't a threat, but chooses to keep tabs on them?  Does she realise that these are not dopes, but dangerous people who won't rush in, and can wait for their marshmallows?

In short, the party has meaningful choices they're not smart enough to know exist.  Ask yourself, dear reader: did the possibility of not fighting the mooks even occur?  Or did you, like every other D&D player, just assume that of course you'd fight them.

What the players do in the setting I've described moves the goal posts.  Because logically, they ought to.  Goalposts are desired outcomes, and in D&D those outcomes are not posts in the ground we have to kick a ball past.

While I want to enable the players in their desired goals, I also want to make those goals difficult to achieve.  And, ultimately, that involves creating opportunities to give the players something they need, in order to fight something they're not strong enough to fight now.  Sometimes, yes, that involves having a old soul appear and give the players a sword with which to kill the Jabberwock ... but with this particular dungeon, there are no old souls available and in any case, we can be more subtle.  Thus, finally, (b), encountering things unintelligent living in the caves, that are unable to comprehend beyond their need for food.

Consider: like everyone who dwells in a home, there are always critters we'd like to be rid of, but which are enormously difficult to fully eradicate.  You know, moths and ochre jellyfish and the like.  As much as the Witch, over the last 25 years, might like to root out every miserable creature that flies or crawls in the vast compartments of her lair, there always seem to be a few that escape the hunt, which then birth the next generation of pests.

Ah well, what's a Witch to do?  Anyway, they're something else to feed on the kobalds if they're stupid enough to ever open that blocked door, or upon the party if they wander off the path towards the Witch's personal lake.  So a live-and-let-live policy is forced on her, just as its forced on all of us with regards to cockroaches, spiders, bats in the ceiling and whatnot.

These allow us two game benefits.  First, it's something to harass the players, bleeding their hit points and blocking their desired free movement about the dungeon ... and second, the eradication of these things gives the players experience, working marvelously as an ongoing Nietzschian bargain as the players wander around and purposefully avoid directly following the mook's path back to their lair.

Of course, the players could do that; they might do it.  And thus they'd find themselves at 1st level facing a 7th level monster with dangerously advanced powers, not to mention her undead warrior servant.  Up front, not the way to go.  My best strategy would be to block their way with mooks, which they'd have to fight anyway, in the hopes the players might realise that someone that controls 60 or so nasty squirgeling things is probably too strong to fight right off.  Players can be rather stupidly rigid, sometimes, so there's no assurance they'd get the hint.

On the other hand, a few nasty encounters with relatively less dangerous and unorganised giant cockroaches, spiders and bats can add to the party's experience and self-awareness.  That might lead to a greater desire to understand the setting rather than supposing the setting's been personally designed to supply their wants and desires.  That might, in turn, reveal the back-end passage that would let the players surprise the Witch, who herself has never discovered, because it means a difficult passage through a very warm part of the hydrothermal branches of the tunnel aforementioned — and she hates temperatures above 3 C.

'Course, the backdoor has no value if the Witch is wandering through the dungeon, trying to learn what happened to that group that earlier killed her mooks.

The game's "play" depends greatly on how the players choose to turn that steering wheel.  I can help by leaving coins and mushrooms for the party to gather in their Mario-like wanderings, but they've got to push past the scary-looking bits of those tunnels they're moving through.  This is a place to stop, as that makes a good post to write next.

Intriguing Things

Going from the last post in the ersatz game, as the DM we have three general scenarios to offer: (a) to have them encounter something unintelligent that's been living in the caves, that's unable to comprehend anything beyond it's searching for food; (b) for the party to encounter the first wave of minions approaching from below; and (c) for the party to find something intriguing yet benign, that lends a clue to their situation. None of these are the "right" option; nor are they the only options — they're merely a set of tried and true situations for the party. I've never gone wrong with these; and because they're so varied in type and presentation, our imagination is the only limitation. As such, in three posts, beginning with this one, I'll talk about each, starting with (c).

This is no different than inserting something interesting that happens along the road back and forth from town, or the establishment of clues that draws the players deeper into the adventure.  A vast amount of running D&D is the use of exposition to continuously insert information within the narrative about the setting.  Exposition is best given in drips and drabs ... in as small amount as possible at a time.  The worst kind of exposition is the "infodump," in which the DM or writer gives all the information about something up front, in as thorough a manner as possible.

I assume most people understand what's wrong with this, but if I might provide an example of bad exposition and good, we can get on with the rest of this post.

Let's suppose that for some reason, as the DM we've invented an NPC who happens to know everything about the cavern the players are entering.  We've done this specifically because we knew eventually the players would get to this point, and we've thought to ourselves, "You know what would be great?  If there was someone who could conveniently explain everything to the players at the right time, in full!  Wow, what a terrific idea."

This is bad.  Not only is it boring as hell to shove this NPC's lecture (that is, our lecture) into the player's ears, it also kills the player's sense of mystery and yen to discover.  If the players stumbling across a helmet, they don't puzzle out it's existence, or discuss it, no!  They turn to the NPC, who conveniently provides all the information the players need.  Ugh.  Don't do this.  Just assume that any method that gives exposition "conveniently" sucks the dog.

As the DM, we need to know, but withholding most of this information, for as long as possible, is critical.  As an example, imagine that above the dungeon, within the mountain the dungeon is under, there are a collection of hot springs.  These dribble down through the mountain's interior along channels and chutes, where they provide hydrothermic heat for molds, bioluminescent plants and all sorts of small beasties ... but as it happens, this alternate biome is 99.5% hidden from the bare rock caverns the players can see.  The existing biome perhaps created these caverns a million years ago, but they've been cut off from the original source of water ... except in a few places where the hidden biome bleeds through.

We mustn't say any of this to the players.  On a macro-scale, we want to use this information to divide the dungeon into two parts that are "warm" and "cold."  The lake hag below despises the warmer areas, with her and her mooks preferring the deeper, lifeless areas ... and this acts as a clue to how she can be found.  On a micro-scale, we can use the hidden biome to add spice to the player's experience, without them initially being able to discern the source for what they find.

For example, not long after entering, they encounter a small cloud of fifty large moths, an inch across, turning and swirling around the party, whether or not they're carrying light.  We know these have emerged from some small vent, perhaps an inch wide, too cool to show up with infravision.  Yet the moths themselves carry their own heat, and remain with the party (unless the party can think of a way to obliterate every one).  Then, if the party moves, by chance, towards the hag, the moths turn back and disappear.  But if they move in another direction, they encounter some new strange side effect of the biome.  Say, a trickle of warm water falling from the top of the cavern into a bowl on the floor, where it disappears through cracks.

These things seem incidental, and would be if they were disconnected ideas.  But they're not here.  Deeper down in the warm parts, the players may find themselves blocked by warm pools occupied by miniature, luminescent ochre jelly, just a few hit points each ... but clustered dangerously like clouds of jelly fish.  They may find a small mushroom glade, occupied by a few violet fungi.  They may discover a part of the biome that they can climb down through, only to vacate when they find the bottom is filled with bloodsucking plants.

And later, after more exploration, they may find that getting past those plants is a back door into the lake hag's lair.

Slowly, as we introduce various elements of the biome's influence on the dungeon's "warm side," they may start to put things together.  They may not ... especially if we carefully interweave the warm and cold parts together.

We have to remember throughout, however, that we're describing, not explaining.  Explaining is an easy, bad habit to fall into.  Let the party do the explaining, as they collect evidence.  That's not our job.

The Non-episodic Dungeon

The last post addresses the narrative of a wholistic dungeon, in which the entire space is in play continuously ... and not just when the players open a particular door.  Rather than a series of set pieces, in which various monsters and scenes sit waiting to be discovered by the players, these same monsters move towards the players aggressively, as soon as the players enter their space. 

The more traditional form is passive, being laid out as a series of numbered rooms — we're all familiar with the format.  In structure, the dungeon is a series of "if/then" statements.  The players act and the dungeon reacts.  This may produce a spontaneous interplay, such as a kobald running off the warn other kobalds, who then arrive to turn out the interlopers — but these events are rarely spontaneous.  They're written into the dungeon's description ahead of time, so that "if" the players go here, "then" this happens.

For some years now, I've run my dungeons more loosely than this.  In effect, I've ceased to make a distinction between the interior of a dungeon and the outdoor wilderness.  In the wilderness, as the players travel, they either stumble into events or have events that stumble into them.  A group of wolves spontaneously surround the player's camp, regardless of what the players have done.  They crest a hill and see a large non-human encampment below, just a few hundred yards off.  There's no intention on the player's part ... they're thrown into the situation and forced to grapple with it.

Whereas a dungeon is usually in the form that when the players want to stop, they just can.  Each door and hall form a bubble that, if not popped, remains in perpetuity.  This lets the dungeon function episodically, with each room having its own character, disconnected from other rooms.  In the extreme, these rooms work like the Dungeon! game, created in 1975 by TSR and updated in later years.

I ran in this style for decades, even as I made my own games ... but over the last ten years I've moved progressively towards a battle-campaign style format, in which the characters move through waves of minions directed by a powerful being, who is fully aware of the players throughout.  This being might rush the players early, and then withdraw, or the players might succeed in an early kill.  I tend to avoid the motif of saving "the Big Bad" for the very end.  Rather, an attempt is made on my part to have the central being's perspective, then decide what it would do as the party's actions play out.

There might be pockets of innocent monsters, stuck off in corners of the dungeon — non-intelligent monsters that rush forward to feed, or intelligent monsters who can provide exposition about the Big Bad.  These latter may be too scared to help, or they might throw in with the party if they see that as an opportunity — and this too depends on what the party says, or what entreaties they make.

The flexibility of this format allows the introduction of new ideas from session to session, just as I might do with the wilderness if the players remained in a given valley over a long period of time.  Because the dungeon isn't a carefully fitted-together jigsaw, the removal of some obstacle might reveal an altogether different part of the subterranean milieu that wasn't there or even conceived of last week.  If the party finds cave-in, and takes time to dig through the rubble, they can expose a small hole that can be squeezed through ... and this unburies some other lair that's been divorced from the Big Bad and everything that's come before by perhaps 30 or more years.  It's a part of the tunnels the Big Bad didn't know about, because it's only been here for, say, a decade.

By having the flexibility to add such "side quests" from session to session, I'm freed from the process of building a whole dungeon, top-to-bottom, before I can start running.  I can start with a general idea — kobalds at the front door, a watery lair deep down, waves of amphibian and insect minions rushing the party as they descend — and then lay out the dungeon during the game itself, which I can do with a few minutes of graphic design on my computer.  As the players can see my desktop as I design the room (I used to do this on a whiteboard), they have something to think about and talk about while I'm designing.  Their attention is at a peak, wondering what the final image will be, and I'm not forced to draw rooms in the stale quiet of my room when players aren't there.  This works for both me and the party, as it builds tension and invests both me and the players in the design process.

I recognise this is, for many, simply "wrong."  For others, it seems impractical.  It works for me because I've had so much practice drawing wiggly lines on a computer program that I can do it very quickly ... but in truth I've always disliked the pre-planned dungeon layout, mostly because I'm never sure it's all going to get used, as my parties, freed from the expectation of necessarily finishing a dungeon, often don't.  Thus, I only need to make as many rooms as I'm actually going to use.

Dungeon Foolishness

When last we saw the players, they had been wrestling with their doubts about going back to the dungeon; and here we'll assume that yes, they go back.  As players almost always will.  This once again allows us to insert a little macro-narrative into the player's experience as they travel through the wilderness ... though I don't want to focus on that here, as we've discussed the matter already.  But as we'll come back round to the subject again, in the future, I'll keep this short.

Simply, as the party are heading back for the dungeon, we push the narrative of a future insurrection in the area just a little bit.  As the party travels, they see ahead of them a patrol of thirteen soldiers violently searching an enclosed wagon, or vardo, throwing things out on the road, where the litter of broken pots and furniture can be seen.  A family stands to the side, mother, father, two boys, with one of the boys holding an infant girl; three soldiers with spears pointed at their breasts stand ready.  The soldiers are laughing.  The soldiers are plainly wearing heraldry that give them official status, a point that has to be made clear to the players: get involved and risk crossing the law.  Whether the players intervene and are told to "push off" by the soldiers, or just continue, it should stick in their minds.  As DM, we want to stress the scene as uncommon.  If the players are so foolish as to actually fight the soldiers, be sure that the soldiers scatter if the players kill just three or four of them.  The concern of what might be waiting for the party if they get back should loom large in the players' heads.

Remember: the players have just established a good connection with the town.  They have the ear of a noble; someone they can turn to, someone they have a chance to impress by returning knowledge of the noble's father.  It would be damn stupid for the party to ruin all that for the sake of a few pots and bits of furniture, and the humiliation of one family.  Still, parties DO overreact in such situations.  It makes the game interesting.

With that on their minds, however the scene shapes up, the party gets back to the dungeon.  They return to the place where they suspended their investigations.  Because we need an example for this, let's suppose the players are standing before a large reinforced door that's been barred on this side, and furthermore spiked all around, to ensure the portal is effectively shut against whatever's below.

I've DMed for many years.  It's good to remember that players dwell in a modern technological environment, one in which most of them have never experienced any extensive time underground.  Moreover, they've not experienced a world where electric light is unavailable, or where silence might permeate so intensely as to thicken the air.  Living in a city, like me, there's always some kind of light; there's always some kind of noise.  A rural setting is both darker and quieter, but I've dwelt in such places from time to time and I can reassure the reader that even in a farmyard, there's nearly always some kind of light, as it's needed to locate the outdoor latrine.  It can be very, very quiet, but even in the woods it's possible to lay in one's tent and hear a stream babbling a quarter mile away.

Underground is absolutely devoid of both.  And this has repercussions beyond remembering to employ a torch.  Imagine the acoustic bell that a stone dungeon forms, when cavern walls are replaced with hand-made tunnels and shafts.  As the party deliberates how to get through the door before them, I promise, one thing they won't consider is how much noise it makes for whatever's on the other side.

In such a place, any abrupt noise sounds like a gunshot, even from a great distance.  It's possible that when deliberating on this side of the door, the party could be mistaken for the kobalds that formerly lived here, before the party slaughtered them.  But once that door is touched; once it's adjusted; anything with the slightest intelligence on the other side will know that something's changed; that someone up above is coming.

Yet, to this, the party is certain to be oblivious.  Without hesitation, they'll decide to bash through the door, pulling out sledges they've bought for the purpose ... and for two or three minutes they'll pound on the drum the door makes with abandon.  Imagine the effect.  A mile below, where the snaking tunnels from above meet, some evil entity feels the first blow as though it landed upon it's skin.  It lifts itself, listening, and there comes another blow, and another.  No, the pattern is clearly intelligent in origin.  It can't be mistaken.  Whatever creatures intrude above, they're clearly unafraid.  And foolish.  And so the entity beckons its army of minions — not to investigate, no, as the nature of the invader is made obvious by the sound it makes.  The minions are gathered in a goodly large number.  "Follow the sound," the order is given.  "Kill all you find."  And so tromp a mass of monsters up through the labyrinth, nearly mindless and bloodthirsty, more fearful of the evil entity than of death itself.

None of which the party knows a thing about, of course.  All they've done is knock in a door.  And once they can see past it into a wide, empty, silent cave, the party relaxes.  "Nothing here," they say to themselves, making a joke or two before trudging forward.  Not for a moment do they think anyone knows they're here.

Reversals

To continue, the players have collected their treasure and discovered that there's a dungeon below the kobalds. As the reader can tell, the aftermath of combat produces multiple issues and downstream effects well past the actual fighting. Those who disparage the inclusion of combat in role-playing games conveniently overlook these effects, and their games suffer on account of that.

Here is another.  As the players contemplate the possibility of going forward, they're acutely aware of their hit points.  Without doubt, they'll have lost some to the kobalds, probably quite a few given how many kobalds there were, and what's wanted is sufficient healing to bring them to full.  Only, as the game was originally designed, and as I still play this part of it, healing at lower levels is in extremely short supply in the game world.  There are lots of ways to get healed ... but instant healing any time the party wants, even limited by some number of times in a given day, isn't there.

The so-called value of "super"-healing rules, such as developed in later games, is that the party can continue to move forward, practically indefinitely if need be, so that the whole adventure can be managed as a single narrative ... just like heroes manage in literature, when all the cuts and wounds they receive are either forgotten or fall just short of killing the hero before the quest is finished.  Many players want that exact arrangement.  They don't want to admit they're human, having to give ground to the enemy, having to retreat home with their tails between their legs like cowards.

Yet notice how often in books and films that the characters have to do exactly that — the way that Luke, Han and Leia have to escape so they can fight another day in the original film; or the way that Luke in the 2nd film, having lost his hand, has to escape so he can return in the 3rd.  He didn't spontaneously regenerate all his hit points.  He lost ... but losing isn't the end.

Most D&D battles are set up to give the players a Pyrrhic victory — named after the 3rd century BC general from a tiny country in northern Greece who repeatedly kicked the Roman's ass in fight after fight.  Though Pyrrhus lost less men than the Romans in these fights, he had less men to start with.  He couldn't exploit his victories.  He had to give up the ground he won, eventually having to fall back from Italy altogether, because he never had an army that equalled his genius.

That's the players at a low level.  They're smarter than the enemy, they have more options and powers than the enemy ... but from the beginning of the game, they venture into each lair by themselves, without a proper support network.  Most that I've run won't learn from this, because they think somehow that the rules of the game dictate that to be heroes, they must do it all themselves.  Instead, the players should build friendships and allies, with fortified encampments two or three miles from the dungeon, filled with supplies, replacement equipment, additional soldiers, a friendly cleric or two to give a little more healing and so on.  All these things are possible in the original game, but they're not even considered by a player party ... no doubt because many DMs refuse to reward players for rational thinking, loving to produce a situation where the players return to their camp to find they have to fight their own hirelings for no good reason except the DM is a fucking prick.

So, instead, we invent bottomless magical healing to overcome the "broken" parts of the original game, to spare the problem of players having to think.  And this naturally leads to other designs that circumvent the deadly capacity of the game as designed, assuring that no one's going to get seriously hurt or killed, at least not permanently, just so that dungeon can get finished without anyone ever having to take a break.

Like the fallout after combat, there's fallout after the players retreating back to town that gives the game a greater richness that punching monsters until they're all punched.  Players have to make choices, which gives them reason to consider their priorities.  They're reminded of their humanity, of the simple fact that they can't endlessly wade against an enemy like gods.  This humanity serves them very well when they're much more powerful and they remember the time when they were very soft and vulnerable.  The retreat gives us, the DM, to remind them that there is more to the setting — remember those examples of adventure hooks we set up while the party was on their way to the dungeon, that we can now bring round to fruition.  Other things, too, may happen on the party's way back.  They may get so invested in something else that they don't come back to the dungeon at all, at least not right away.  We give them choices.  We give them perspective.  We assure them that no, the game isn't just an endless cycle of replacing all their hit points and going around again.

Before we get to that, however, we have to talk about the party that decides, "We're just going to camp right here, in the dungeon we just cleared, until we get our hit points back."

This is always fun.  I've allowed it at times.  Online, I had a party moving through a cave that had been sealed off for centuries, sitting literally miles from the inhabited part of the cave, who camped out for four or five days, sending one player character back to town for supplies.  So yes, in the right circumstances, it's possible.

That scenario doesn't really apply to this kobald lair, however.  There are, to begin with, a lot of dead kobalds and dead rats laying around.  Far from being sealed off for centuries, the outer door is open to the world.  Is there a party of 20 kobald hunters on a 2-day trip, about to arrive back?  Have they absolutely killed every last rat?  Won't the bodies of the dead kobalds start to smell?  And — according to the rules of my game — the unburied dead do tend to rise as undead.

Now, it may be that because I grew up reading Weird War Tales as a comic, I'm naturally attuned to this concept ... but resting amidst scores of dead is just begging for a very bad scene.  How long does it take a murdered kobald to decay sufficiently in order to rise as a skeleton, when it's home is being occupied who have such contempt for the dead that they won't even stack them together?  And is it even possible for a human/demi-human cleric to give a proper burial rights to a kobald?  Surely these are questions worth considering.

Even if that's not a thing ... me, the guy writing this, would have quite a lot of trouble resting in the tomb I'd just created.  And I'm a 21st century fellow with a strong knowledge of science and not practicing a religion.  What does a 17th century player character think when he or she's lying, trying to sleep, and some mouse somewhere near by starts chewing on a kobald corpse, causing a random arm 50 feet away to slip and fall to the floor?  "Oh, that's a mouse"?

Remember that healing without magical means (or rule changes) takes a good long time.  A cleric has to get a full sleep, to pray again for a spell that only cures a little bit of the player's total lost hit points.  The party would have to rest for days, hearing every tiny sound gently echoing in hollow stone rooms that would be eerily silent when something wasn't skittering across the floor ... while waiting to see if it's our imaginations or if something real is actually happening.  While all the while being uncertain whether or not we've definitely killed everything that lives in this lair.

Running this is easy.  It's just a choice of picking words.  "You hear something very quiet; it sounds like it might have come from around the corner ahead; but it's probably nothing."  "Falling asleep, you feel like clawed hands are closing around your neck; you feel the boniness of the fingers as you have trouble breathing ..." [pause for effect]  "You wake up, gasping for breath.  It was a dream."

Takes about twenty minutes of this to get a party to smarten up, gather up their shit and head back to town.

The Dungeon's Front Door

Yes, I have a book with this title.  Yet it seems appropos for this post.

Our party has come to the entrance of the dungeon they were seeking in the last post.  This is a critical juncture in the player's experience.  The moment of committing, as it were.  We would do well to examine what happens within the game at this point, and how best to exploit it as a dungeon master.

Let's review first.

1) It's the players' agenda.

2) The player's agency is constrained by what the players have control over, and what they do not.

3) Everyone is bound by the rules, players and DM alike.

The player's are here at the dungeon's door — whatever it looks like, we'll come to that — because they wanted to be.  They asked to be here.  They deliberately sought out the entrance ... and because we've identified this as something that's within their control, they have a right to be here.  Let's give some thought to point 3.

There is a trust issue here.  The players don't know what's behind the door, yet the game dictates they must commit.  The DM ought to know what's behind the door ... and knowing it, properly should respect that content just as rules are respected.  If I imagine the entrance way leads into a forward guardroom of some kind, with three goblins whose role it is to warn the rest of the community if something bad comes in the door, then I'm resolved to put three goblins there and not seven or nine or fourteen.  The only reason to change the number of goblins would be if the players killed them "too easily" — which shouldn't matter a damn to me because I'm impartial.  It shouldn't matter any how, because the other goblins in the lair are also there, anyway, no matter how easily the front three die.  Frankly, I don't understand a lot of what motivates a DM to mess around with dungeon contents, but the principle should be that whatever's there is there.  The party can always find something else to kill later, if they still have an appetite.

To be clear, I'm not saying this dungeon starts with a three-goblin guard room.  That's just an example.

In any case, as long as the party is outside the door, there's another bit of duality at play.  There's a distinction at this point between what the party believes and what's true.

Moving back in time a moment, the party ought to have been given some information about the dungeon before starting out for it.  We tell them there are goblins in these mountains, or that the dungeon was originally sought 25 years ago because a princess was kidnapped and taken there, and though there was a rescue planned, no one of the rescuing party ever came back.  We don't know what became of the princess, or the rescuers.  We know there have been other parties who set out for the dungeon, but the party doesn't know if any of those groups even found the entrance — as none of them came back.  These bits of information are floating in the party's consciousness, and we can be sure some of them are going to be hashed out by the party before they commit themselves to first finding the dungeon, then entering it.

Physically, the dungeon entrance takes space.  This space is exception in the game setting.  It's the doorway between two worlds: civilisation and, as I said in my book The Dungeon's Front Door, a dystopian unreality that exists below ground.  How we depict the crossover matters.  It's best if we don't make it look like a traditional Gygaxian front door, which might just as well have a neon sign out front stating, "Dungeon, -- Vacancy," for all the ambience most modules offer.  And note, giving the entranceway "ambience" matters to the player's mood when they arrive at this point, before crossing the threshold.  We can do better than a plain little hole in the mountain fastness, or a benign little concealed door.  Early D&D subscribed to the "less is more" motif, which may have worked when the game was new and the players unjaded, but we've got to do better.

We can provide ambience by fixing the players' imaginations with intense flavour text, as James Raggi perceived and had a great deal of success with.  Of course, that's become a trope also, and in reality there's only so much that text can accomplish.  It's been over-used since the mid-00s and is apt to leave any experienced party cold ... though it probably will still enjoy effect if the players have only 2 to 3 years of gaming experience.

A better approach is to mix said text in moderate doses with the visual qualities and aesthetics of film.  The strongest benefit of film is that it shows movement ... and so, when introducing the entrance, try to think of some element of the scene that's shifting, stirring or flowing.  Drift snow into the scene, which greater distinguishes the outside, where the players are, from the inside, where it's dry and perceivably warm.  Roll steam or cold air out of the entranceway, with an explanation the players can't know without investigation.  Allow voices, or noise of some kind, to drift out; perhaps unseen creatures are having an argument, or there's an inexplicable rattling or booming sound.  Have a creature actually step out of the entrance, then go back inside before the characters have time to fully react.  Tease, tempt, taunt the players with such.

It's fully in the rules to make the entrance seem more dangerous than it is.  Although the party wanted to be here, it's in our interest to make them question that intention as much as possible ... to force them to push past a legitimate reason to fear and actually BE brave.  In all the various make-believe elements of the game, one thing that is not made up is the real emotion possessed by the players.  It's this real emotion that makes the game so satisfying.

The emotions offered by a video game are frustration, exhaustion, sometimes boredom, occasionally relief and, after a long time, a sense of accomplishment, self-esteem and even triumph.  But rarely can a canned experience, like a pre-programmed game or a movie, or a book, offer an emotion like bravery.  It doesn't make you "brave" to willingly sit through a movie, or buy a game, or finish a book.  Deliberately choosing to enter a dungeon run by another human being who is under no obligation to make that dungeon "popular" to a vast audience IS a legitimately courageous act.  For myself, I carefully cultivate the attributes of a demented, somewhat unwholesome sociopath who could conceivably put something in his dungeon that would be, well, very wrong.  My players sense that about me.  They know that whatever's lurking in there, it doesn't need to sustain itself against a focus group, it doesn't need to answer to a market's whims and it definitely doesn't have to fit a politically correct model.  I might very well include something so upsetting that it takes a few sessions to get over it.  It's one of the reasons why my partner Tamara has decided to stop playing.  She prefers to hear about it, before I unleash such a scene upon the players.

And so, as the players stand outside the entrance, remembering the stories of other parties, remembering how long this dungeon has been "known" without anyone properly cleaning it out ... and knowing that once they commit, getting out might not be so easy — it stands to reason that they'll hesitate.  I want that.  The more time it takes to screw up their courage, the more satisfying it'll be, for them, when they marvel at their boldness to take this chance.

For that's what this part of the game is.  Commitment and chance.  They've invented stories, letting their beliefs get the better of time ... but it's the truth they'll find, that they can't know without first finding it, that really discomforts them.

It's ridiculous when some director like J.J. Abrams makes an argument that what we don't see is the more frightening thing.  What a soft, pampered life Abrams has obviously had.  Numerous film directors have shown us things plainly on camera that shake us to our core ... because we're SHOWN these in their full horror.  Abrams makes scary films for children.  Scary films made for adults are terrifying.

I run adults in my game.  For adults, the yawning maw of a waiting entrance is a passage to hell.  Abandon all hope, ye who enter here.

Counter-duality

Coming back to worldbuilding.

Let's put all the detailed structural work surrounding geography, politics, sociology and such upon a shelf today and talk about "function."  The world is a setting where things happen, but the world is also a tool designed to be practical and useful for game purposes.  The goal is to get the world up and running so that it's applications can serve both the DM and the Players, and other aspects of the game aside.

As a DM, I need the built world to provide me with inspiration at a glance whenever I need it, specifically during the course of the game.  This means that if I look at a map, that map is constructed in a way that upon looking at it, answers to questions and possibilities emerge within a few seconds, leading me to present situations for the players that I may have had no thought of 60 seconds prior ... but which some part of the work and preparation I've done regarding building my world has just provided.

Let's take a familiar example.  A dungeon room is a highly recognisable D&D construct.  Most DMs, having the map of a dungeon at their elbow, can visualise a number of monsters that could be placed on the fly in that dungeon room; what they're doing in the room is fairly evident as well, since there are only a few things that a group of orcs, say, or an ochre jelly, can possibly be doing as the players arrive.  While the structure of the dungeon is limited, that structure provides a clear-cut functional purpose.  The players are here to fight, they're here to find things in the room, they're here to seek passage to the next room.

If I take a random dungeon map, say the sort that Dyson regularly creates, as experienced DMs we can reason easily what ought to be there and, as the game session begins and gets going, the shape of the walls, the small incorporated details, whatever beasties we've inserted and so on sets in motion a fertile degree of invention ... one that we're used to because dungeons are such a large part of our gaming experience.  Plock any dungeon graphic in front of a DM with a mere two or three years of experience and running a single session without notice is not a difficult task.

Unfortunately, DMs tend to rely so much upon dungeons that they gain little experience with other possibly functional locations.  For many, a town encounter fails to produce an adequate experience because there are NO comparable, useful resources on par with what a dungeon layout provides.  Likewise with a wilderness or an occupied rural landscape.  This disconnect occurs because such places are not structurally limited like a dungeon.  A village or a town dictates that virtually anything that's done will be observed by someone, since these places are full of people; and if the act is observed, it's bound to be reported.  Many players, comfortable with doing whatever they want in a dungeon, suddenly become stymied by the lack of "freedom" a town offers.

And that itself is a strange happenstance.  We might suppose that the dungeon is the constrained environment, but it's not.  The consequence provided by a dungeon is whether or not our side can win in a fight ... whereas the consequence in a town is usually the recognition that if the whole town turns out in force, we're definitely going to lose.  Let us fight 20 orcs and not a hundred guards.

Taking the wilderness, it's a question of exposure.  Fight those 20 orcs in a dungeon and afterwards we can spike the doors and make our retreat, confident that once we're outside the orcs will lose our track ... and if the DM isn't the sort to hurl wilderness encounters, or run a strictly wilderness outdoor adventure, we're "safe" as soon as we get outside.  But if, when we're outside, a monster of any size, and particularly of great size because the outdoors gives lots of room, attacks us while we're recovering, um, we're fucked.  Players who lack experience with long-term outdoor situations also lack knowledge of how to protect themselves, while DMs who also lack experience tend to think the only thing one can do is throw large beasties as encounters.

In short, a dungeon seems rich and full of possibilities, while other parts of the setting feel repetitive and the players excessively vulnerable.  In response, many DMs choose to discard any conflict in towns altogether, along with any use of wandering monsters in the wilderness.  It's simply easier to ditch these adventure ideas and get the players out to the dungeon, where the DM is happier and the players as well.

These are setting-function deficiencies.  Because DMs have never been taught or even seen another point of view, they tend to perceive every setting in terms of the dungeon's combat-puzzle-trap model.  Or they turn to the dualism that if the game isn't combat, it's role-playing ... so that town encounters become long drawn-out chat exercises, where most of the party has no idea what to say and everyone is worried that saying one wrong thing will land everyone in loads of trouble.  This IS possible; there's no pretending it isn't.  But likewise, it's just as possible to make one incredibly stupid move in a dungeon and be in loads of trouble.  It's only that most of us have seen what those stupid dungeon moves are ... whereas we have next to no idea what they might be in a town.

Moreover, as DMs want to stress the importance of role-playing in their campaigns (if they're the type), they will up the stakes with every conversation until we're at the baker's, ready to draw our swords while discussing the cost of a dozen biscuits.  Overmuch ballyhooing is the bread-and-butter to many a role-playing DM, struggling to assign tension and difficulty where, in fact, there ought to be very little.  Most conversation isn't, and shouldn't be, stressful.  It should be interesting, full of enlightening details and suggestions for social and status advance ... but my own experience has shown that even if my foot isn't on the gas, the player's will be automatically.  Time and again I've had to explain to players that nothing bad is going to happen if they trust someone.  This is clearly a taught habit, as multiple DMs in the player's past has taken advantage of the player's trust at every opportunity.

The approach to role-playing, again, pushes the players back to the dungeon — where they need only worry about bloodsucking monsters and insidious pendulum traps.  Thank gawd, we're back where things make sense!

Caught between the game being either a dungeon crawl or an uncomfortable chat with townsfolk, it's difficult for most to glance at a map like the one shown and come up with anything except, obviously, to visualise a dungeon located near Reni or Macin or Ibraila.  The rest is a smattering of places and names, without a sense of "D&D" going on.

I've touched on this element before, but it's worth repeating.  The dualism of the dungeon/role-play model is that BOTH place the player firmly in the passive role, waiting for the DM to do something that makes the game go.  The players wait for the next dungeon room.  The players wait for the other side to start the conversation that decides what the players will do next.  And the players, used to this functional structure, dutifully wait to be told what expectations have fallen upon them.  "I need YOU to go to this dungeon;  I need YOU to fetch this object to me; I need YOU to rescue the princess."  The players are a foil in the hands of the DM, who arranges the targets, or obstacles, for the foil to overcome, all the while being in full grip of the foil's handle.  The players get the delicious promise of being the tip, the part that actually touches the enemy; the part whose whirling around in the air feels exciting.  But all the while, the tip possesses no actual control; the control is in the DM's subtle wrist.

Plenty exist to argue this ought to be the case, both DM and Player.  Virtually every player has  experienced only this form of play.  And having had only one sort of experience, it stands to reason that players loving the game would argue against a change.  After all, they don't know what they'd be getting.

When I look at the map, I don't see a passive experience for the player.  Given sufficient wealth, time, freedom from responsibility, freedom from worrying seriously about death — all of which is the player character's perspective — I see a place where I can freeboot around as much as I please, witnessing and personally experiencing the local culture, making arrangements, bartering, intervening as I please against any wrong I see and always confident that my escape is assured ... as planning my escape comes before taking the hazardous step to intervene.  For every enemy I make I'll be sure to make three friends ... and if I am a player in this location where the DM is a fellow like myself, then I have every reason to believe my friends ARE friends, and not a trick of the DM who cannot keep his hand off the foil.  I see half a dozen opportunities in the map for plunder, investment, piracy, trade and settlement ... with plenty of wilderness to clear out mere footsteps from my front door.  That is, wherever I happen to put my front door, which might be anywhere.

Freed of the worry that I might say the wrong thing in a "gotcha" campaign — as these are just people and I've spoken with people all my life — and equally confident of defending myself in a wilderness, I'm not beholden to the DM's instigation of a pre-planned campaign ... nor do I have to approach the DM to beg him to run such and such a module because I'd like to run in it.  I will, thank-you, write my own "module" on the fly, because I think I can do a better job.

And taking the DM's perspective, all the opportunities I've just cited are there:  the little village run by an petty overlord, the lake monster that needs killing so the barges can roll past safely, the beach where the players can lie in wait for a small river craft to seize, the hamlet they can pillage before making haste to another part of the marsh, the dock they can have built, the smuggling in which they can partake, the bay where they can build their lair and so on ... and of course, if it's really necessary, the half-drowned dungeon they'll find somewhere in this part of the world.

The map gives me this; the little symbols on the map as well; the perspective of how a world works; and the readiness to let the players do as they like without arbitrarily restricted their behaviour on any account.  This is worldbuilding to me.  Every part of the human experience laid out, good and bad, with clues baked right into the design that I can see clearly and reveal to the players as they set out to explore.

Observations

There's less and less reason to post on a Saturday; my numbers fall, people who read me are playing D&D, it makes more sense for me to hold off on a post until mid-day Sunday.

But ...

I was just thinking about the favorite explanation of some, that profound treasures are placed at the bottom of dungeons, "to protect them."  Adventuring players must then battle their way through a host of rooms to reach the fabled treasure, that usually turns out to be a profound magical item, or group of items, laying atop a big pile of treasure.

Only, I wonder if any player has ever felt, after getting this item, artefact or whatever, a great need to build a dungeon and fill it with monsters to "protect" this item they've just obtained.  I think not.  Players seem to prefer keeping the items with them, and employing them, flagrantly carrying them along roads and through towns, casually hanging out in taverns with the items in their pocket or hanging on their belts, while spending the treasure on things rather than keeping it in an enormous pile in an isolated place.

So where does this motivation come from that NPCs, having acquired the treasure themselves, and the items, feel such compulsion to sequester such things so far from themselves that when the plunderers break in, the NPCs don't even know the item is gone?  Hm?  Surely, the last room, with all the treasure, should include a magic mouth or something that suddenly shouts, "Oh, so you've found my treasure, that I was keeping here for a week or two while visiting my mother.  Well, I've just cast clairvoyance and clairaudience, so put everything down if you don't want me to kill you in a day or two.  And I'll thank you to round up a bunch of golems, ghouls, kelpies, crayfish and sphinx to repopulate my dungeon. Thank you, have a nice day."

Just saying.

Why is it we never find one of these enormous, underground civilizations
while they're still active?

Exercise: the Space Beyond the Door

Problem the first.

A dungeon generation chart might seem fairly silly to some, but I'd like to pitch its value a moment.  You see on the right a copy of a table that appears on p.170 of the original DM's Guide.  It is plain to see that the table does not attempt to dress the dungeon in any way, but only tries to define the physical space beyond the door ~ whether it is a passage, a room (small space) or a chamber (large space).  The actual nature of the room or chamber, and its contents, is left for another table, which I hope we can discuss another day.

My request for the present is this: can you improve on this table?

Your limits are this: you can't include skins that alter or adjust the walls; you can't add aesthetic features, items or elements that are semi-permanent; you can't add monsters; and you can't create a dead end.  A room/chamber might be a dead end and a hallway might actually turn out to be a dead end, but these things are determined by other tables.  This is strictly about the space behind the door.  The door has to go somewhere.

I've produced my own version.  I built several whole dungeons with the above chart in my first 10 years of D&D, until I felt I'd moved past it ~ and the most annoying thing was always the passage turning back on itself 45 degrees, so I've abandoned that feature.  I've added vertical features.  And I've removed the need for the notes that Gygax had to add for his version.  I prefer to use "ft." instead of a single quote.

I'm certain that Gygax deliberately left stairs out as an option.  His conception was that stairs would always mean the dungeon progressed a level in danger, so his generation made them annoyingly rare, so that a party wouldn't quickly find itself facing monsters out of their capability.  I don't have any problems with a "first-level dungeon" spanning several physical levels.  I feel the cut-off between one level and another ought to be based on something less feature-driven.

My reason for the exercise is merely this: to make the reader think.  We see these massive megadungeons depicted everywhere, but the reader will note they are created of the same functional patterns, like the space beyond the door, again and again.  I'm asking the reader to deal with one small facet of the dungeon creation machine, to see if it can't be improved upon.  Like examining a car's head gasket, to see if it can be made more durable.