Jul 15, 2026, 11:05 AM“A recurring question was how much we should still care about software design at the granularity of code. These are concerns such as naming, code duplication, dead code, mixing responsibilities, modularity, coupling, and cohesion. I can roughly group the participants’ opinions into two opposing arguments:”
https://verraes.net/2026/07/software-design-in-the-agentic-age/
Jul 15, 2026, 11:10 AM..
“* As with humans, code quality doesn’t matter much on short-lived projects, but has a huge impact on the long term maintainability and changeability of the code. Camp 1 is therefore not looking far enough ahead.
* LLMs’ context windows are limited. More accidental complexity in the code increases the context.
* Complexity increases the token cost.”
Jul 15, 2026, 11:13 AM@RuthMalan (he then proceeded to create what is, by all means, 4 camps described by two rules :p)
@RuthMalan What happens when an agent is no longer available? How does the software survive? It seems they've introduced a massive point of failure into their process and are completely blind to it.
I don't understand how this software passes even rudimentary V&V. Verification is mostly mechanical but validation seems completely ignored here.
Maybe it's because I work in an engineerong culture and not a software development culture that this all seems dangerously unsustainable and detached from reality. Handing implementation to a stochastic black box seems unwise to put it mildly.f