Manbearcat
Legend
I'm curious about a few things Manbearcat, in theory I get how this should work but I am more concerned with practical application...
1. We can suppose a vast number of skill challenges will be undertaken throughout the advancement of a single character, do you write down all of the fictional tidbits that may ever be interacted with at a later date? If not how do you keep track of numerous spur of the moment inventions of fiction that may or may not be relevant at a later time... or if so, doesn't this get tedious? How do you keep it from becoming a logistical nightmare?
2. With the abstract nature of skill challenges (variable time frame, distance covered, non-exact measurements etc.) how do you decide exactly where the gorge is, say if the PC's decided that once they found their bearings they wanted to map where the gorge was so they wouldn't stumble into it again?
Gotcha. Ok. Here is procedurally how it functions at my table.
The clerical side of things is managed similar to the way Aspects/Tags/Scene Distinctions are handled in other games. We have flash cards at the table and each scene will have its own flash card. One side has the scene stakes established at the outsite. On the reverse side is relevent scene information via a pithy descriptor; "Treacherous Badlands, "Nigh-Impassable Gorge". This then evolves as play continues, with new scene "tags" scrawled by myself and the players whose turns may introduce them. I use these to generate complications and the players use these to riff off of themselves. By the time we're finished, we'll have anywhere between 5 - 15 tags (or so) that serve to illuminate "what happened" and/or "what is there."
After the game, I'll generally reflect on what happened (both inwardly and visually by way of the scene prop that we created) and, if relevant to future play, scrawl something on the side with the stakes to illuminate how they were resolved. If I feel something is particularly of consequence as it may be relevant to a potential future conflict, then I'll make note of it in some way (either tangibly by keeping the scene flash card handy and highlighting something or just by making a mental note...I have a pretty good memory).
How do we/I decide where the gorge is and/or any notable marker for future reference? Couple moving parts here:
1) Mostly its just a product of knowing a few abstract vectors. With respect to the gorge, the temple is maybe a day and change of travel through harsh badlands from the forest's border. In the fiction, terrain was much worse on the first leg of the journey from the forest and the gorge manifested at generally that same point on the way back. It appears at the tail end on the flash card, right before the resolution of the challenge. As such, any reference that need be made to it in the future would set it much nearer the forest than the temple.
2) As you've surmised, the default temporal and spatial resolution of affairs is pretty lean. Not even in the same neighborhood of robustness with respect to simulatory sandbox games where setting exploration is paramount. If something needs to have time zeroed in on (such as a conflict that must be resolved before dusk or "something happens"), then we'll zoom in on it there and make sure the table is of the same mind. Otherwise, the default is low resolution. As such, logistics becomes less important (but certainly not irrelevant) to play.
And finally, if I think that something needs to be squared away, I'll certainly ask my players their thoughts and run with their position if it differs from my own and its relevant in any way. This very, very rarely comes to pass though due to 1 and 2. Although I have done a full-on "schrodinger's dungeon" before, in the case where a dungeon/complex scenario makes its way into play (where setting material may be recurrent), I will have an assortment of generic maps that I can make-do with and use as reference to guide navigation (for them and myself) of the exploration Skill Challenge (marking the map with corresponding numbers as we go).
I would say that, in terms of GM workload, the main difference between simulatory sandboxing where setting exploration is the primary agenda and games where conflict-charged scenes are the primary agenda is that the former is pre-session prep and planning intensive while the latter is in-session improvisation and theme/genre-focus./pacing intensive. Both require post-game reflection in order to get internal consistency (of what, and to what degree, you are interested in internally consistency as a major focus; setting and/or thematic conflict) and consider where future sessions may go so you are prepared mentally to engage your players' cues and the content of their characters.
As far as difficulty or tedium (specifically clerical tedium), I suspect that would vary greatly from GM to GM.