Last I touched this topic I was over religions and nations, all from a high-level perspective and not being too attached to anything. With me actually working on something more concrete, now is time to begin tying knots and setting some bases in stone. While I am far from being done, first I must decide a few things.
You must decide ahead of time how you want to define your world. No matter which method you choose to use, review is a major factor in the creation. Write it out first, then edit. Don’t do the other way around.
Home group
After setting the basics of the world and the locations of the first few adventures, run the game. Seriously. Just run it. Take notes about what happens in the adventure and build or expand as needed. The home group approach evolves the world with the PCs as they grow in power. The biggest advantage of this is that you do not need to create too much ahead of time.
All that you really need are some of those high-level questions that directly relate to players making their characters. What are the common languages in this country? Who is the ruler? What is it like overall? Are there elves/gnomes/dwarves? For that, the Ten Questions are a great starting framework (see my other post about this).
The "Home Group" approach is a great way get into the setting and to try it out as soon as possible. Instead of thinking of some great over-arching world-spanning plot, you can let it evolve from play.
The problem of this method is that it is that as you work for your group, you might forget some important aspects precisely because of the play focus.
Thought-out
This one differs from the Home Group approach because you write as much as you can about the world, giving it flavor through writing, instead of play. There is more design work to do before playing. The world here is designed around itself instead of a group’s preference and style. Not to say that players will not bring valuable input later, but what you will present to them is a product that will be more elaborate.
Okay, okay I can hear some of you (and you know who you are) scream at me. "When I put together a campaign, I do not do it haphazardly!" You are right. The difference is that here, you try to cover more before taking it to the table.
The main problem with this approach is that it can make a setting become stale and hard to change. Say that you really liked an NPC you created but your players keep hammering home that he makes no sense (is stupid or someone cracks a joke at him that sticks). It’s hard to go back and change some of the core things you have already thought of but you have to.
The Mixed Method
If you follow this blog, you know that I can never be too attached to one method or another. I always mix it up. The two methods above are good, but it is unrealistic to only build from what your PCs give you or to write everything then play. The (obvious) solution is to mix and match.
Where you plan on starting your PCs and the focus of the game should be a lot more elastic than region where you do not think they will be going (at least in the short term). So if the PCs start in one region/province/country. Define it, by giving it basic shape and information, but be ready to tweak a lot more and evolve from the "Home Group" approach. The province next door, unless it plays an important role in your main region, can easily evolve from the Thought-out method.
Do not kid yourself, unless you have absolutely no life, family or job and your friends/players are the same, you will NOT be able to play-test everything. You won’t. Unless your world is VERY small. Very, very small.
JP