In another thread there has been a discussion of map-and-key RPGing, and how it is a distinctive way of handling framing and resolution in RPGing.
Consider the following situation, not uncommon in fantasy RPGing: The PCs are at place A, on the edge of a desert. They wish to get to place B, on the other side of the desert.
How has it come to be that everyone at the table agrees that that is the fictional situation? One possibility is because there is a "world map", and the position of the PCs, and places A and B, are all marked on it, and by looking at the map we can see that there is a desert between A, where the PCs are, and B.
That is an example of using a map-and-key to frame a situation.
First off, since we've had so many discussions about terminology, I like map and key. I think that captures a distinction in gaming. I also think even for those of us who embrace map and key you point to something that can arise as a potential issue with it in this post.
The PCs decide to cross the desert. How long does it take, and do they make it? One way to resolve that is to use the map to determine the distance, to divide that through by a movement rate for the PCs, to look at how much food and water will be needed per unit of time, etc. This is roughly what is presented in (eg) Cook/Marsh Expert D&D and Gygax's DMG. I get the impression it's also an approach used in at least some 5e D&D (eg the wilderness travel in Tomb of Annihilation, if I've properly understood what I've read about it).
That is an example of using a map-and-key to resolve an action declaration.
Consider this different situation: The PCs are lost in the under city sewers and catacombs, and find themselves looking up through a grille in the street where an enemy is looking down at them, gloating. The reason why the PCs were sneaking through the catacombs is because they were trying to make it to a wizard's tower in the middle of the city, before their enemy (whom they'd drugged to give themselves a head start). Now they have to race through the undercity while their enemy proceeds above ground!
How has it come to be that everyone at the table agrees that that is the fictional situation? One possibility is because there is an undercity map, and the GM has been providing descriptions to the players (eg "You come to a place where the tunnel stops, and the water drops over a 'falls' to a cross-wise tunnel 20 feet below you - what do you do?") and then the players have been describing which way their PCs go (eg "We use a rope to lower ourselves down alongside the 'falls' and then proceed down the cross-wise tunnel to the right"). Eventually, the result is that the PCs find themselves at a place where there is a grille to the street above.
This points to the challenge of the map and key approach, which is at granular levels, you don't have a map. This can vary obviously, a GM might map out a whole city, including the undercut, maybe even major structures in the city as well, but it is rare to see a GM map every single building in that city and to map every acre of countryside. I think a crucial thing to have in a map and key approach that keeps it from becoming too much like a board game or video game is to always remember there are blank spots on the map that are implied by the map and that you can always zoom the lens closer past the details of the map you have (and here I am thinking the video games from my childhood, as I am sure newer video games are advanced enough that this isn't an issue). In any map and key game there has to be a point where the GM is improvising, extrapolating, randomly rolling, or using alternative techniques like you describe below.
That would be an example of using map-and-key first to resolve the action declaration "We travel through the undercity catacombs and sewers" and then to help establish the framing "You are looking up through a grille in the street, where your enemy is looking down at you." How the presence of the enemy is established, on this approach, is a further thing - perhaps its a random encounter roll, perhaps the GM has just decided to add it as a type of "spice" to the situation dictated by the map-and-key.
For me most encounters are either
When the situation I've described happened in my Burning Wheel game, though, it didn't arise via map-and-key play. There is no map of the city or its catacombs and sewers. After the PCs drugged their enemy, their players decided that they would sneak through the catacombs to the wizard's tower. I called for a test on Catacombs-wise, at an appropriate difficulty. The test failed, and so I framed the PCs into the situation described: they had become lost, and found themselves at the grille where their enemy - now recovered from having been drugged - was looking down at them, taunting them. No map-and-key resolution of moving through the catacombs; no map-and-key contribution to the framing of the situation.
In my game, the race to the wizard's tower was not resolved via map-and-key either. It was resolved via opposed Speed checks.
I often resolve this type of situation by opposed speed checks myself (sometimes I throw in other skills too, very much by feel, depending on the situation). One thing I do occasionally do is bring in spur of the moment map and key, to introduce a sense of objective choice to the players (i.e. I pin down a detail about the layout on the fly that is meaningful and makes a difference depending on whether the players decide to do different things or take different approaches: climb on roof tops, go through a series of nearby alleyways, etc). But I do need to be on the ball for the latter to work and I will often not do it if I am not feeling particularly alert that session.
In the same game, when the PCs needed to cross a desert to get from A to B, there was no map-and-key resolution. Rather, there was an Orienteering check made against an appropriate difficulty.
In this circumstance for me it again tends to be a combination of the two things. There is likely to be a desert on the map (again not every acre is going to be detailed so Q&A is likely to produce new information and material). But once I know what route the players are taking, what plans they have in place for the journey, it comes down to Survival rolls for that terrain based on how many hexes on the map they are traveling through (and so the map is still important in terms of knowing distance, but also matters for encounters because which hex they are in can shape why the encounter is happening and what table they would roll on for an encounter).
Another approach to resolving these sorts of action declarations is from Marvel Heroic RP - I've used it in my fantasy adaptation of that system. The goal or threat is framed as a Scene Distinction (say "In pursuit of the Orcs" or "The Giants are gaining on you") and then the players can declare actions for their PCs which - in the fiction - help the PCs in their goal (eg in one of our MERP/LotR sessions, Gandalf used his magic to create squabbling among the Orcs whom the PCs were pursuing, and who were carrying a palantir that they might well squabble over - this slowed the Orcs down) and which - mechanically - ablate the Scene Distinction. If the Scene Distinction is removed before something else happens to bring the scene to an end, then the PCs achieve their goal (eg they've caught the Orcs, or escaped the Giants).
I am not sure I follow this one entirely so I might be misunderstanding, but I don't think I've done this. That said, it brings to mind some approaches I have taken, especially when we are doing things that involve a lot of PC planning. Just as an example, I have a session coming up where the players have been forming alliances with various sects in the region and are planning an attack on the HQ of a sect called Long Ma Hall that is the main ally of their big enemy in the campaign. The map is important here, they need to know where the HQ is, for instance, but for this situation I am on the fence about whether I need to map out the sect HQ itself or not. It is something that could easily get glossed over in the broader strokes of the battle and tactics. However if it becomes relevant because of something the players want to do, I may have to ad lib the map of the place on the spot. I will probably just draw on the HQ of a sect from an old shaw brothers movie in that case and stick with it. But in terms of where the whole thing starts, that is very much going to be a product of what the players plans are. And I am not going to break out miniatures and play out every single battle (likely we will roll based on the numbers and power levels of each sides men---maybe dividing it into groups of rolls based on individual battles going on, while I allow the PCs to pick individual 'named' opponents to face off against during the battle (always giving them the option to go after larger groups if they want). The point is it is fluid, somewhat cinematic but also grounded in a map and key and in objective numbers (the sects each have all their members listed, complete with stat blocks and Qi ranks, but rather than play out those battles individually most will be abstracted into d10 dice pool roll offs).
In non map-and-key approaches a map might still be fun or even helpful to use, as a source of descriptions and flavour and so on (eg in the MERP/LotR Cortex+ example, we referred to the map when we described the Orcs as having begun their journey in Angmar, and heading south towards the Gap of Rohan). But the map is not being used as a constraining device for framing or for resolution.
This post has focused on the use of map-and-key techniques to establish and resolve actions to do with where the PCs are and where they go. Map-and-key approaches can also be used to determine what the PCs find but I'll hold off on that until I learn whether or not there is any interest in this topic.
One potential issue with the map and key approach is that it can become too static. This is how I often played when I first started (the orc chief is sitting in room 6 the whole time waiting for the party). The way I have come to conceptualize the map is more as a snap shot in time, and I try to incorporate rules to have elements on the map evolve and be more like moving parts or living entities (again here I find the living adventure or wandering major encounter concept from Ravenloft to be quite useful for handling major NPCs and important groups on a map). To use the above example here, a static approach would assume that when the players go to Long Ma Hall, all of Long Ma Hall is there waiting for them (at least all the remaining members as they are specifically attacking now because they know half the organization is in a neighboring city on a mission). But I prefer to take an approach where just as the players are gathering intelligence, so are their enemies, just as the players are making clever moves on the map, so are the enemies. With an individual NPC it is usually easy to intuit where that character is and what they are doing. With a group I find rolling randomly can be helpful. But in both instances a roll is still probably going to be required to figure out the result of whatever the NPC or group are attempting to do or to learn. So in this case I would make an information network roll (a d10 dice pool based on how large a network they have) to see if the sect catches any sign of the players' impending attack before hand (which could result in unexpected tactics or ambushes that the players might face). Generally though, with this sort of thing, my emphasis is on being fair to the players. So I try not to get invested in an idea like setting up a cool and unexpected ambush, rather I give the dice reasonable chances of failure or success and let them fall where they may)