Shared publicly  - 
Discussion of the Day- Setting books always tend to suffer from massive bloat (IMO). Tons of needless names and places and etc to the point it feels like a chore to read and you end up not remember half the shit in there.

How would you like a setting book to be? What would you like in it?

Myself: I'd like more books to take the Vornheim approach. 30 pages o fluff (and adventures) at the beginning and useful charts and tables in the second half. 
Mike Evans's profile photoBrendan S's profile photoDuncan Young's profile photoZak Sabbath's profile photo
I use a really brief one-paragraph format for each "nation" or major area is my setting.  The first bit is a single line that gives the basic gist of it; the second bit is basically a list of things that is going on there; the last bit is a taste, a song, and an image that sums up the feel of that area.  Anything more than that is overkill for me at the moment.
I like using wiki's instead of campaign books. I can always peruse a wiki or do a search and find lots of interesting stuff. It also makes it more accessible if I'm gaming online or have access to a laptop at the table. The interconnected articles also make it more interesting to read since you're not just reading it cover to cover.
The thing about Wikis is they don't necessarily let you know what's important quickly. With my book I tried to go "Ok, this is the first thing to know, this is the second, this is the third, these other things are not things you need to know to run a game here, don't sweat them"
+Zak Smith which is why, thus far as I have seen, Vornheim is one of the best setting books ever. 
+Jonathan Black Maybe you're right. Wikis confuse the heck outta me. though I do like the "editable" ness.
+Jonathan Black I also think a  wiki is a "scientific" solution. Information organized clearly and accessible. Problem is, not everything in a setting needs to be introduced that way. Some of it does, for reference later after the first time you read it, but in fact, many details are better off being narratively introduced, it makes them more memorable so you never have to reference them again (this is why the most memorable settings come from memorable fictions).
I think the stuff I reference over and over about a setting would do well in a wiki. The stuff I need to know to get oriented I want in order and with as little "choice" as possible.
I'll just point to a few examples of setting books I love.

Dictionary of Mu, its a list of demons with some fluff for Sorcery & Sword, all organized like a dictionary (it cheats with an appendix though) 

The World of Near, it breaks down the setting into different "thematic" parts of the world, along with commentary on changing it and such. 

Blossoms are Falling, two short fiction pieces (which are good), followed by a short summary of Heian period in japan. After that, the book follows up with different campaign frames, rules to help invest into the setting, lifepaths, etc. 
What do you mean by choice? Too many options in a given area, etc? Too many locations, cities, etc?
The book that finally killed setting books for me is Freeport. HOLY FUCK is that thing bogged down by superfluous amounts of needless info. 
Ok, example: first page of +Jonathan Black 's wiki Wiki. I read a little story. Ok. Read. Now where do I go?
"Timeline" ? Do I need a timeline to play the game or run it? Don't know. Don;t know whether to click or not to get oriented.
"Other sites"? umm.. these don't have to do with Heaven's gate so I don't want to read them
"Site tools"? Whuh? Are these for me? How do I know which tool I want?
Aaaaand that;s everything on the page. I'm confused.
+Mike Evans No. I mean choices about narrative paths to find out what I need to know to run the setting. I want all the info I need in a useful order to get oriented presented right away.
Here is a slightly better "wiki" though I perfer the book, since...I'm me, this is the World of Near wiki.

It starts off with a short introduction, then if you go to Front Matter (which HONESTLY should be broken up, since that contains tons of advice and crunch). After that, each "movement" has a short introduction explaining whats in that section. 
+Zak Smith Ok... Gotcha. Yeah, wikis can be a nightmare to navigate. Especially as a world gets developed and gets piled w more info and links. 
A wiki could be written in similar fashion to a book but still be able to be hyperlinked for ease of writing and aimless perusal later.
+Justin Smith Page one:
I am reading a bunch of publication history I do not need to know. Then given a table of contents-like choice.
I go to the only possible meaningful choice "front matter"
Introduction to what it is, then... more publication history...extensive metadescription (no examples or reasons I might be convinced this is a good idea yet)...MORE publication history (acknowledgements)...contents!
1st Movement
consists of the material I think of as the "short Near" ( what? ): it tells of Maldor, the empire, and what became of it after the Year of Shadow ( so: history? Do I need to know it? Don't know yet ) . I've arranged this material with internal coherence in mind: if you're overwhelmed by the extent of the setting in whole, the 1st Movement is actually a sort of mini-setting in its own right, perfectly playable. oh ok
"2nd movement..."
(Why aren't these movements just descriptive chapter titles?)
The kind of organized encyclopedic engineer mind required to organize and be enthralled by RPG-style data and the kind of humanistic "What are people like and what do people need to read in order to grasp things and be motivated to grasp them?"-mind often do not function simultaneously.
Far too many RPG books are like "Classes? They're under the heading 'classes', duh!" yes but we want you to show us the interesting thing about your setting ASAP to decide whether to keep reading, not make a periodic table of setting elements If the classes are what make your setting worth reading, put them or a reference to them up front. Anyone who would call a chapter "Third Movement" instead of "What's in the Southern Continent" is thinking like a programmer, not a reader.
+Zak Smith Ah, ya, as I said its slightly better. I do have some issues with The World Of Near, like yes...drop the movements, give usable chapter titles (and more breakdown during each "movement"). Also I do not like how Front Matter is just...well, a bunch of Front Loaded info. 

But I like how there was an attempt to organize with "reading" in mind, rather than like in an encyclopedic text book fashion. 
+Zak Smith Yes, I also feel that setting books should also be made with a type of system in mind. The needs for a D&D are a bit different than a Sorcerer game, which is fairly different from Heroquest.

Then again, that goes into your point about thinking about what people need, rather than just tradition or what makes sense to the writer.
+Justin Smith I dunno about system. I think a setting book should go (in order):
1. Show me something that makes me want to use this setting (instead of close competitors!).
2. Tell me the bare minimum I have to know to know what's going on
3. Any special ideas/systems/unique places etc you want to introduce
4. Give me any gameable tool I will want to refer to over and over all organized and marked and sectioned in the back, or, on the web, in its own section
Hm, I think it does. Having run different games, as a GM, I need different bits to grab me. Running D&D, I'm looking for cool locations and monsters. Heroquest, give me some cultures and big picture situations.

Though, that could be how I just approach games. Either way, your four-point list is more spot on. 
Personally I don't need a paragraph description of each shop and people and their machinations, sexual fantasies, and bowel movement regularity.... Too many slip into this type of info... Which (IMO) is useless at the table, because I hate fucking stopping to consult a book and have to read paragraphs of shit to find that one golden nugget of info I need.

That's why Vornheim was such a breath of fresh air. A) heres a summary of info on this crazy wackadoodle city; B) here's s few pages of adventures to give you feel of city if you want; C) a plethora of tables that give better feel of city, can be resolved expeditiously, and keep focus on table antics. 
+Mike Evans I make wikis for my campaigns just for me. It's something to pass the time with at work in between shift/busy work. If the players read it and get engaged enough to want to add to it... that's also a benefit of using a wiki.  I don't know that a wiki would ever be useful to market a campaign setting though.
I think it was +Zak Smith who said something very wise about dropping the setting bloat and instead creating tables of events that were full of local flavour. I've not had the opportunity to try it, but it seems sensible. One would imagine that you could tie this approach into the tagging system used in Stars Without Number too.
+Kelvin Green It really works.  Before I actually did it I was running a sandbox style Firefly game and got tired of needing shit on the fly that I didn't have an answer to (such as how long to get from 1 planet to the next) so I said fuck it and did a Vornheim hack for Firefly/other sci-fi games. :)  Works beautifully.  
It all depends on who you're writing it for +Kelvin Green. If you're writing something for yourself and your campaign, not your players or to sell... I still think the wiki is the best option. It puts all the information at your fingertips and since you wrote, you're more likely to remember it. Which will make it easier to find later.
+Jonathan Black, true, but the strength of using tables is that you don't have to remember the details of what the weather is like on the Nanjooby Plains, or what kind of things live there, or what there is to eat.

I also like +Jeff Rients' twenty questions approach; it's simple and direct and answers the questions that the players are most likely going to ask. I think it's easy to forget to make setting stuff useful as game material, and to fall down the rabbit hole of writing your own Silmarillion.
I do use a wiki for myself and for my players to remember details (if they want to bother looking at them) such as NPC's they've met and pissed off and interacted with....  or events that have happened.  I have my map on there and just enough info to jog my memory should I need it (  but largely the wiki grows as my players fuck up my world :)

 Honestly I use a mix of the two- I use my wiki and Vornheim or Firefly or my wilderness hack I'm working on now (depending on what I'm running)...  but the thing that generally sees the most use is the Vornheim style book (that's just for me though).. 
+Kelvin Green - You would have to remember after the first time you roll on the tables though. Players expect consistency. Hot dry weather on the arid Nanjooby Plains would be odd if the last time they visited it was a frigid tundra on the border of the arctic.
Wikis are tools for collaboration. Otherwise, they are just hyperlinked documents. And they tend to rot, because different parts don't get the same amount of attention, so some (often most) pages are out of date (for whatever out of date may mean in a given context).
The best thing I've read about setting books recently is that bit that +Jeff Rients referred to in Mekton Empire. Here is a quote from that sourcebook:

In the vast galaxy that the Bendar Galactic Empire calls home are many mysteries that player characters can explore. Only the Referee should know the Real Truth behind these mysteries, but nothing stops a player from reading a game background book and discovering the real story behind the great mysteries of a game universe. For this reason, we have created a variety of possible answers to the most puzzling questions about the Bendar Galaxy. Each Referee can decide what the real answers are, and how these truths will affect the player characters.
I have no favorite setting books, but I do recall that the best have tons of names and facts. They're usually called guides, gazetters, and histories.

Remember that a setting book is only words until you guide a party through it.
I quite like the structure of the Poor Wizards Almanacs for Mystara - there's a lot I don't need (standing army sizes etc) but it focuses on the politics and plots between nations that can be nice backdrops to whatever location-based adventure you want to run
I found +Rob Conley's Points of Light 1&2 to be Short and concise setting books. They are regional maps with a key; most things get a paragraph or two. NPCs get name, class, level. It's very similar to the Wilderlands of High Fantasy but shorter. I have used both at the table and have been very happy.

His Majestic Wilderlands has more house rules which I don't care for and unusable small maps. This is why I haven't used it. I usually find that browsing the blogs and reading books gives me cool ideas and make me want to add stuff to the campaign. These things don't need to be in a setting book.

I've used the Forgotten Realms book for D&D 3 as a player to help me write a backstory because I felt the DM enjoyed this kind of thing. The setting book provided names and places galore, so it was very useful to me. As I suspected, however, the backstory itself turned out not to be very useful at the table. This is why the kind of games I like to run don't need this sort of setting book (Forgotten Realms, Eberron, Golarion, Glorantha). Generally speaking I need much less religion, history and culture background than is usually offered by the big setting books.

+Alex Schroeder Thanks for the shout out. Sorry you didn't get much use out of Majestic Wilderlands. My upcoming Scourge of the Demon Wolf should be more useful for what you are looking for and has a mini-hexcrawl detailing a barony 
It always bothers me when someone calls something people clearly use or used "unusuable". It's like they don't know there's other people in the world besides them. You say "I found __ unusuable" or "I couldn't use ___ "
+Zak Smith Isn't the "in my case" and "in my opinion" assumed? It bothers me that I have qualify all statements with those notes.
+Zak Smith The great blog of holding..  I swear you've got a post in there for every situation/argument.  
I dunno +Zak Smith. I get some of what you're saying but if someone says to me "X is unusable", I read an additional "...insofar as my experience and examination takes me." on to that. It's up to me to then assess my view on that person as an informed critic whose view is trust. 
Also, one can always ask the asserter for more details.

+Zak Smith If their response is "because it is my opinion dude, and you can't argue with taste" then I believe your post applies 100%.
+Duncan Young Disagree. Because sometimes "x is unusable" is a statement of literal fact. In a medium where half the stuff is homemade or produced by companies that make massive mistakes there are may cases where things are literally unusable: Like the 2 page spread in the 4e PHB is literally unusable as a character generation guide and _literally (not subjectively) says things like "Character name: write your character's name here" and DIY publishers literally can accidentally print things with actual mistakes which make them unable to be used  and some automated CG programs literally do not function as advertised and maps sometimes literally do not match the territories described.
If you are reporting to people not present about things they haven't seen, you have a responsibility to recognize the difference between things that are so mistaken or misconceived they are literally errors and things you, given your own parameters find difficult to use. To do less is to do a disservice to the people who put their hard work into making the thing. And to anyone who is trying to figure out whether to buy a thing or not without spending a lot of their own time doing it.
+Brendan Strejcek No. First:
-that is untrue, sometimes "the asserter" is unavailable for comment, and, second
-many statements are deceptive and readers (people on whose perceptions small projects could fail or fly) will obey them unquestioningly because it doesn't occur to them a reviewer has chosen to use hyperbole. Like for example with many details of Carcosa.
+Zak Smith There should be a different standard based on medium. On my blog, or Google Plus, it is easy to ask me to explain or expand on something. If I'm publishing an article in a scientific journal, less so. In the same way that I don't footnote and source most statements in face to face dialogue, unless someone asks. Social networks are closer to conversation than publication. Part of their vibrancy comes from the fact that people throw things out that are hyperbole or are overgeneralized.

I agree regarding the Carcosa example, but that is sort of a different thing, as much of what was stated was additionally just false. At that point, it is no longer a discussion in good faith anyways, and we can't have the standards for all conversations be assuming that they are not in good faith.
+Brendan Strejcek I agree that on G+ we usually assume we can ask people about their hyperbole and they usually answer. But in nearly every other forum, we can't rely on that and (unlike G+) that hyperbole is on the Google search for the product in question forever . So while I may have unnecessarily jumped down someone's throat here, I think it's a bad habit to get in to.
I do not think hyperbole is a good thing in RPG discussion in general and I think nearly everything people bitch about that shits up RPGnet RPGsite storygames and every other game forum is largely a result of it.
+Zak Smith If I understand you correctly, you are arguing for more nuance? In general I agree. In this concrete case though, I think Alex's mention of the maps in the Majestic Wilderlands book is not problematic.

[Edit: we cross-posted, and your above comment answers this.]
+Brendan Strejcek I think if anyone pressed for time and trying to decide what to buy a thing reads that the maps in that book are "unusable" ever anywhere ever then that is a goddamn tragedy that could've been avoided by the person writing it being more thoughtfu. You can be real mean about a product, but there is no excuse for imprecision in this context
Seems a lot of the issue come from the use of the word "unusable". Printing errors aside, there's plenty I've read in rpg supplements I personally wouldn't use or couldn't imagine using in a typical game - but little I couldn't imagine being usable or useful at sometime or to someone. 
"couldn't imagine using in my typical games"
+Duncan Young but there is some stuff that's literally unusable . Errors are common and need to be pointed out. Confusing that with a subjective judgment about utility is a real thing when you are dealing with a small project like Wilderlands.
Add a comment...