Each layer of abstraction is commented with a fabric like flag attached with two strings.
As you type and describe what you want, this is saved. inputs are indicated and outputs expected are automatically displayed. imports are specified in code, these are separate from the rest of the code, but once imported it acts as any other node.
describing the layers in this way allows for us to search for previously written abstractions and transformations.
we can zoom in and out of our abstractions, depth visible can be specified.
most important thing is clear primitives and transformations. clear pipelines. clear ports. import export.