I've split this off from Generic American Settings
; hope no-one minds me reposting my comment from there.
I'm suggesting looking at written setting content (perhaps I should be calling this colour - the stuff within your game environment) as falling somewhere on two axes. These are the level or depth of detail, and another factor that I'm muddling between calling evocative detail and idiosyncratic detail. Talking will hopefully show what I think I mean.
With these 2 axes, rather than seeing games in terms of generic to detailed - in my opinion too coarse to capture important possibilities - you have four possible poles, with games falling within the limits they offer.
1. Detail Low, Idiosyncracy Low: Generic setting. Setting emerges from what the players bring to the table, but without guidance or a firm vision there may be a risk of tending to the lowest common denominator - safe options that are less likely to clash with the game, and that everyone has a good grasp on (because of their ubiquity in culture). However, a group confident with the game system and with their own vision are going to be able to wrap it around this fairly straightforwardly, as there is little 'getting in the way'.
2. Detail Low, Idiosyncracy High: I would put Dogs in this category: lots is left unsaid, but you have elements (coats, ceremony, mormonism) that are very evocative, and so interact with individual and group schemas to produce an emergent sense of setting, directed to some extent by those elements. But these elements suppress some possible settings just as they offer up others, and this may dismay some who find it difficult to play as these elements (in this instance, more due to mormons than to coats, based on internet chatter) get in the way of where they want to go.
3. Detail High, Idiosyncracy Low: Concerned mainly with the mapping out of relationships between setting elements, in terms of distances, partnerships, and so on, to give some kind of predictable framework; the kind of work that John Kim has said he likes to be provided with by a game setting (hence I imagine one such setting might be Harn).
4. Detail High, Idiosyncracy High: Concerned mainly with throwing crazy ideas at the reader to get the juice flowing, as well as the mapping provided by the previous. Possibly something like Exalted? I dunno. Anything that provided me with the stuff that I got from reading China Mieville, say - a million tit-bits that left me wanting to know more.
I notice now that my poles aren't truly poles; I'm sure that someone could come up with a setting that expressed almost no facts but was incredibly evocative in what it did give you. But hopefully you get the drift.
People will prefer some approaches over others for certain games. I would prefer a generic setting for a supers game because I have a palette of super-stuff that I want to smear this way and that. A spacefaring game, I want some ideas, but once you get me started I'll be happy riffing off of them. The third option is less my taste, but would make sense where the characters were very compelling and I wanted a complex structure for them to explore and alter. The fourth I would probably read and then use the snippets that most jazzed me, zooming into the bits that I thought were cool. My current short-duration gaming taste shapes these preferences to a large extent. Possibly other people will have a single flavour that they always go for.
Does this gel with how other people see things? If you think so, is there a case for one method consistently working better than others, say for a specific creative agenda or type of genre? Is it firmly down to individual differences?
While I think formal setting/situation creation mechanics are really cool (and I'm trying to monkey around with sitch creation for my own design) I'd be really happy to leave them aside for this thread and focus on how the setting as written is contributing to the players - to their imagined space, the urgency of stories, the richness of characters, the space to play.
[edited to tidy up.]