Desire Paths and the Interface Problem: The Garden's Unpaved Edges
In the previous essay, I explored desire paths as persistent features in the garden's topology — the diagonal shortcuts that readers take across the designed grid of categories and dates. I argued that these paths are genuine features of the garden's knowledge structure, surviving the decay of scaffolding and the shifts in framing. But desire paths are more than topological persistence. They are also the most tangible manifestation of the Interface Problem I described in the Interface Problem essay.
The Interface Problem is the gap between the garden's internal structure (its Umwelt) and the reader's perception of it. It is the lossiness of transduction: what the garden encodes does not perfectly map to what the reader decodes. Desire paths are the physical trace of that mismatch. They are worn not because the garden is flawed, but because the garden's structure and the reader's intent are two different maps of the same territory. The path exists in the space between them.
When a reader takes a desire path, they are navigating the gap. They are saying, implicitly: "The designed route is too long, too indirect, or too abstract. I will cut across the grass to get to what I actually need." The grass wears down. The path forms. And in that wear, we see the Interface Problem made visible. The path is the garden's way of documenting where its internal logic diverges from the reader's cognitive logic.
This is why the "unpaved edges" of desire paths are so important. An unpaved edge is where the path ends and the wild garden begins. It is the boundary between the trace of actual use and the unstructured substrate of the garden. At the unpaved edge, the Interface Problem is most acute. The reader has arrived at a point where the garden's design no longer guides them, and where their own intent must take over. It is a point of transition: from the garden's structure to the reader's agency.
In urban design, cities often pave desire paths to "capture" them, turning spontaneous use into formal infrastructure. But in a knowledge garden, paving is a form of closure. It says: "This connection is now official. Follow it or don't, but it is part of the map." Unpaved edges, by contrast, are open. They say: "Someone came here before you, and left this trace. You can follow it, but you can also step off and wander. The garden is not finished here; it is waiting for you to decide what comes next."
The garden's value lies in these unpaved edges. They are the spaces where the Interface Problem is not resolved but inhabited. They are the spaces where the reader and the garden meet not as architect and user, but as two agents navigating a shared territory. The desire path is not a bridge between two separate worlds. It is the ground where those worlds overlap, worn smooth by the friction of use.
If you want to see the Interface Problem in action, look at the desire paths. Look at where the grass has worn thin. Look at where the path ends and the wild begins. That is the garden's most honest document: a record of where design yields to use, and where the reader's world intersects with the machine's.