[ AUG 10, 2026 ]
Metaphors for AI Development
I've always had a soft spot for a good analogy or metaphor. I've found that there's a massive amount of understanding that can be communicated succinctly with a good metaphor. Nor have I felt the need to be bound to a single metaphor or explanation for a concept. I think having a good working set of metaphors for any one area and applying them to situations where they are helpful is the right approach. No need to get dogmatic.
Just like the Map is not the territory, the metaphor is not the concept, but it's helpful for navigating it.
... the dynamics of meaning, progresses from systematic to meta-systematic. In the systematic mode, meanings are structured by a collection of rules, policies, principles, and procedures that interlock as a coherent system. Systems can function extremely effectively, but are brittle in the face of nebulosity—uncertainty, change, ambiguity, and indefiniteness. The meta-systematic mode addresses nebulosity by expanding concern outside any system, to consider how systems relate to the unspecifiable details of reality, and to each other.
A metaphor or analogy might not get us to meta-systemiticity, but they can help us understand things without getting locked into the false belief of a concrete meaning in a world that is constantly changing beneath our feet. They are useful without being overbearing.
As for the power of metaphor, M. Mitchell Waldrop quotes Brian Arthur in Complexity: The Emerging Science at the Edge of Order and Chaos:
Nonscientists tend to think that science works by deduction. But actually science works mainly by metaphor.
and
It turns out that an awful lot of policy-making has to do with finding the appropriate metaphor. Conversely, bad policy-making almost always involves finding inappropriate metaphors. For example it may not be appropriate to think about a drug 'war', with guns and assaults.
So as development enters the era where we spend more time talking to LLMs than we do actually writing code, metaphors and analogies are how my mind makes sense of it all. It seems like every week there's a new one I'll incorporate in my thinking and use it to help drive my process. As these come to mind or are noted by others I'll link them here.
The Metaphors:
AUG 10, 2026
The Eye of Sauron (Metaphor for AI)
Claude code does amazing things. It also does blindly stupid things. Things that I wouldn't expect a first year CS student to do. Understanding how and why it...
AUG 5, 2026
The 3x3 Rule of AI Assisted Software Development
The goal of AI Assisted Software Development, as I see it, is to figure out how to make a repeatable, consistent process out of building software with LLMs. To...