You may have seen my previous posts about my initial attempts to code with LLMs. At the time, they seemed promising. But in the past few months, my mood has shifted over into a sort of depression. Below are a few of my complaints. [1]

• Using LLMs takes a significant amount of the fun out of coding. For me, a big part of the appeal of writing code is solving problems. Having a computer tell you the answer is much less fun.
• LLM coding tools largely appeal to the productivity-obsessed and quality-indifferent. When it comes to writing code, that is an unambiguously very toxic combination.
• The security implications of a fully-LLMed workflow are a bit terrifying. Simon Willison has written compellingly about this, and he claims that no one has solved these (rather severe) problems yet. This should really give us all pause.
• The old ways of coding are being sidelined, which seems short-sighted. There are those who are — perhaps Quixotically — sticking to the old approaches. I suspect that their work will prove prescient in the long run. But right now that seems like a lonely battle and I don’t think I have the fortitude for it.

The result of all this is that I have been losing interest in modern programming. And it is probably prudent not to program in a half-hearted way with dangerous tools. So maybe I should just focus on my work with Humanities Methods in Librarianship and worry about other things for a bit.

[1] I know that these points are not at all original, and quite likely a bit naïve, but I wanted to bring them together here because they speak to how I feel about this situation right now. That’s what blogging is for, right? :)

This entry is licensed under a Creative Commons Attribution 4.0 International license.