I'm in a project with other people where some of us, me included, prefer to go from general aspects to more specific details (top-down). But others do not share the same view and because of that, I feel the project is crawling to a halt.

I don't think any one approach can be considered objectively better than the other, but having an overview of the up and down sides of each approach seems helpful in trying to figure out which way we do this project. For example I think top-down is better for consistency as I want to avoid neighboring cultures having nothing in common.

What are the advantages and disadvantages of each technique? And could the two possibly be combined to form a middle ground?

  • $\begingroup$ I'm not sure if this question has an answerable interpretation that most people will understand the same way. I didn't vote for close (yet), but this may be much broader than intended. General and specific are relative and can be applied to a lot of concepts on a lot of different levels of abstraction. Worldbuilding refers to all kinds of things from landscapes over magic, sciences and species, up to politics and the like. I'd guess this problem should have something to do with limiting discarding due to conflicting constraints, but I might be on a totally different page. $\endgroup$
    – Vandroiy
    Nov 4, 2014 at 23:08

3 Answers 3


This choice can be thought of as more driven by personality and preferences on how to think about a subject, than by a specific need to do one thing or another.

The question is not directly about personality theory, so I will just quickly reference the Big Five Personality traits (where this particular trait is hard to pick out) and Myers Briggs where it is clearly on the Sensing vs Intuition axis.

So my answer is that you need both approaches, they are both valid ways of dealing with building knowledge. They are also two equally valid ways that consumers of your world will approach understanding it - focussing on just one angle is making it harder for 50% of your audience to appreciate the work. Your task is not to choose one or the other approach, but to create a setup and team structure which helps all of you engage with the project.

This is not a trivial endeavour, because building separately top-down and bottom-up is not guaranteed to join up. However, you need each other. The big ideas need to support interesting details, not lead to blandness, or turn out to contradict themselves when thought through in detail. Conversely, complex details that don't fit into a theme or structure risk being a homeless mess of ideas.

The answer in my opinion is checking output and willingness to adjust towards the end goal. That means peer review, respecting each others' angle on how the world works. Where possible build new things that make the details possible, or things that explain the details in the context of the wider picture. This is just the improvisor's advice of "Yes, and . . ." re-worded. Where not possible, discuss suitable edits and revise.

I also suggest repeated rounds of brainstorming how things could and should fit together, as a team effort, as part of a routine for the project. Depending on how quickly the team is producing content for the world, this might be something you would do weekly, once per month, or maybe case-by-case as a peer review on draft content before the group agrees to accept the design and consequences of any piece of work as "canon" of the project world.

This approach works better when contributors can create drafts and plans of their work, presenting them to the group before committing hours of effort to polish them. It is much harder, and the producer less likely to be willing, to significantly modify a large almost-complete piece of content, as opposed to making the same adjustments to an initial draft.


I think it depends on what goals you might have. It makes sense to start generally and work your way to specifics if you only have a general idea of what you are looking for and want as many things to fit together as possible.

Start specific and working to general if you have some very specific ideas you want incorporated or are key features of the world, then it can be easier to build from them to better explain those features and how they came about.

In a group project, if well organized, it can work well if you have people working from both ends toward a middle, but it can also work against you. If you can work out some general parameters for someone and then give them the go ahead to work their idea to fit you might be surprised how well that works. If they need/want to change some of the assumptions you've given them, then they need to come back to the group to discuss the change.

(Unfortunately this is entirely my opinion, though some experience dealing with designing software went into this)


Top down yields very strong consistency when you look at the world as a whole. However, when you look at the details, it often starts to feel artificial, because the details all seem to be written into the stars.

Bottom up yields a more "realistic" word in the sense that we (currently) believe the rules of the world govern the small things at the bottom and the larger constructs are just results of those bottom features. However, it often proves very difficult to build a "good" world from the bottom up to support your needs. If you "cheat" to get a plot going, bottom up either spirals out of control or turns to useless mud.

Another way to word the same thing is that top-down is good at creating beautiful but inanimate worlds, while bottom up is good at creating living breathing worlds. I like to think along this organic/inorganic line because it gives some hints as to how to answer your third question: yes, you can blend the two. However, I do not have reason to believe there is just one way of doing it, so you have to go out and find your own personal way! To suggest directions to look, consider how we (organic beings) manipulate inorganic tools. HISTORY TO THE RESCUE!

  • A swordsman is an organic being that uses an inorganic sword to cut into other swordsmen. One can use a top-down portion of the world to create a tool to allow segments of the bottom-up behaviors that you like to pin down the out of control segments nearby
  • A staff can be used to strike or push an opponent. The top-down portion of the world can be designed to nudge bottom-up uglies out of the way without killing them off
  • A combat fan looks beautiful, but hides its violent edge. The top-down portion can distract the audience until they reach a point where suddenly they realize that the bottom-down portions are living and breathing and have them in a treacherously dangerous position.
  • A potter slowly forms a pot on a wheel. The bottom-up parts can dance around the top down, slowly molding it into a desirable shape

Humans have been interacting with inorganic things and eachother for so long, it forms a good source of inspirations for new ways to weave top-down and bottom-up thinkers together.

This view has one drawback: it always is from an organic perspective, because we usually think of inanimate objects as not having any "will," so they just sort of sit there as the organic being wields it. However, if we draw from fantasy, this drawback goes away:

  • A devlish sword convinces its owner to draw it. The top-down portion of the world begs the bottom-up portion to sharpen it, draw it, and use it.
  • A magic mirror slowly deceives its owner. The top-down portion draws the bottom-up portion into a position where it can suddenly ensnare the bottom up portion and begin building the world outward from it

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .