Note: An LLM was not used in writing this microblog entry.
This entry is part of a microblog series called LLMs
I shared a thought on a big dev call at work, when talking about agentic “skill” based workflows, that one problem is that you lose Flow, which is often one of the most enjoyable ways to work as a software dev (most of the rest is frankly frustration and toil, but valuable). Today I saw this post by Gabriella Gonzalez about the same topic, with examples of interfaces that try to do a better job of that. I’m not as gung ho about LLMs as Gabriella is, but I’m enjoying these simple articles that discuss a particular technical point.
This way of thinking has been on my mind for some time, as generally I am always working to reduce feedback loops in my work, never increase them. (This doesn’t follow from using Emacs, but Emacs can help with that: look at magit.) It’s why slow CI ruins my day, and why most cloud ops like Terraform are painful (e.g. Amazon resources aren’t content addressable, so checking for changes to apply is always O(n)–brutal!).
Separately, but somewhat related; arguably, there’s a certain psychogeography to “walking around” a codebase when doing an update or refactor, taking the scenic route reminds you of the landscape and dark alley ways, it activates and gently tickles parts of your brain in a way that I imagine is valuable for the larger work, which is an experience removed by being teleported from A-Z.