By "living, breathing world" I was referring to the style of play that puzzled [MENTION=6696971]Manbearcat[/MENTION], namely setting out backstory in advance, and statting it all up (either literally or notionally), even though this is not for the purposes of creating a classic Gygaxian/Pulsipherian challenge game. (An example would be where the GM has decided that (say) the mayor of the village has such-and-such stats, even though there is no expectation that the players are meant to learn what those stats are and use that knowledge as a resource for their own clever play.)
I'm not referring particularly to PC backstory, although that can be flexible too. I'm referring primarily to world backstory. Ie, the stuff that [MENTION=6696971]Manbearcat[/MENTION] was talking about in his post.
For instance, and to take an example from my own campaign: why is the Raven Queen on good terms with Kas, even though she hates undead and Kas is a vampire lord? I don't know - although it has been established in play that she is on good terms with him, the reason for this hasn't been established yet. When the time comes for me to make a decision, I will be doing my best to make a decision with maximum dramatic heft, both in terms of shock value and generating momentum.
To give a PC-oriented example: it turned out that the human wizard in my campaign was really a deva who had taken human form, and lost his memories, for one of his incarnations. But when he died fighting an angel of Bane, he was reborn once more as a deva and regained his memory of 1000 lifetimes.
The PC had been in play from 1st level. This particular bit of backstory was brought to light at 16th level.
In general, when I refer to "flexible backstory" I am referring to the technique which holds that no backstory is firmly established until it is actually narrated in play, either by a player or (more often) by the GM. The most hardcore form of flexible backstory is no myth.
An additional feature of flexible backstory is not determining the mechanical representation of a gameworld element until it is needed for play. In 4e, this means using all the monster-building tools that have been discussd in this thread. In Robin Laws' HeroQuest revised, this means using the system for setting DCs, where the DCs are higher the more successes in a row the players have had. In Burning Wheel - which uses "objecive" rather than "level appropriate" DCs - the Adventure Burner (which is that game's equivalent of a GM's guide) suggests not statting up the big bad until the last feasible moment of prep, because you want that NPC, in mechanical terms, to be a suitable challenge for the PCs.
Or to refer again to the village mayor example, there would be no need to stat up the village mayor until those stats were actually going to matter to an action resolution declared by the players. (In 4e this would mostly be combat - because non-combat in 4e is resovled via skill challenges, which is a players-roll-all-the-dice technique in which the you don't need to give stats to the NPCs who are part of the scene in order to work out the DCs - you just need narative descriptions like "not very brave" or "as a matter of pride, unwilling to compromise".)
This is not an example of flexible backstory. In fact, it seems to be an example of exactly the opposite. If, in the course of play, I present the players with the opportunity to have their PCs explore the Altan Peaks, why would I then try to discourage them? That just looks like bad GMing to me. If I present them with the Altan Peaks, then the backstory of the world is going to include the Altan Peaks being the sort of place the PCs might try and cross.
To give an example from actual play: in the default "story arc" of 4e, PCs start at heroic tier - able to help villagers, fight orcs and goblins, etc - then advance to paragon tier - able to defend nations and enter the Underdark - and then advance to epic tier - able to travel the planes and challenge the gods and primordials.
When my campaign started at heroic tier I didn't frame the PCs into situations that were obviously incongruous for that tier - eg present them with an invasion by duergar or drow. I framed them into situations that fit with that tier - defending a homestead from goblin marauders, sneaking into a hobgoblin fortress, recovering a stolen elven idol from a haunted island temple, etc.
When the game advanced to paragon tier, the nature of the challenges presented likewise changed. The PCs got the opportunity to take the battle to the hobgoblin army, where the five of them defeated an entire army (but not without losing their wizard, as described in the post linked to upthread). Now that they are epic tier, the challenges have changed yet again: they are on the Feywild trying to end the War of Seasons between Winter and Summer Fey, by defeating the army of frost giants mustered by Lolth in cooperation with the Prince of Frost.
If I were to present my players, with 1st level PCs, with a situation in which the most immediate and pressing threat was a war with frost giants, and then reminded them that, as 1st level PCs their prospects of hurting 17th level giants were pretty slim, I don't think I'd be doing a very good job of GMing. Likewise if I presented them with the Altan Peaks (at least as you describe them).
This makes no sense to me. I don't see any connection between the technique used to establish backstory and mechanical representation of gameworld elements, and combat.
For instance, White Plume Mountain is a pre-authored scenario - both fiction and stats - but any group playing through White Plume Mountain is going to have to engage in quite a bit of combat. Conversely, here is a write up of the design and play of a scenario which was primarily exploration rather than content, but which used relatively little pre-establihsed backstory. It illustrates how players can make decisions based on previus experience and knowledge whether or not the GM has pre-authored and pre-statted the aspects of the gameworld that the PCs are dealing with.
I don't really have any idea what this is about. I don't know what you mean by "intuitive with the world". If the PCs are fighting some creatures, than those creatures are the gameworld. And if the choices that the players make don't have any meaning either for the resolution of the encounter, nor for the broader context of the campaign, then I don't see what the point of the combat is at all.
That's not in dispute (though in 3E wouldn't it be incredibly tedious to play out?). I also noted upthread that, in 5e, bouned accuracy is meant to handle this (though I worry a bit about tediousness there too). The point of the example is to rebut the claim that the 4e method entails an inconsistency of gameworld and of mechanics.
But it's not process sim. It doesn't determine outcomes by modelling the processes that occur in the gameworld. Rather, it uses a large does of fortune-in-the-middle and relies upon both players and GM having a good sense of genre constraints (including the specifications of the differences between tiers) in order to achieve heroic fantasy outcomes.
