We already know all the components, just need to configure to fit the problem
Input: needs, goals, functions
Output: Artifact that satisfies those needs, goals or functions
With design, both the problem and solution evolve together
Think AI agents that design AI agents
Configuration: A problem-solving activity that assigns values to variables to satisfy constraints.
Start at the top left w/ specifications (think footprint of the basement example)
Begin abstract and refine/expand on each iteration
Middle method of abstract and partial solutions can be represented as a design plan
Lower level plan or refinement
the arrangement may cause new specifications to come about, the process repeats
Output the arrangement model
loop(assign, refine)