Kent Beck says YAGNI was never about saving effort. The thread says the essay is AI slop

YAGNI, re-derived from net present value

In "The cost YAGNI was never about", Kent Beck throws out the familiar defense of You Aren't Gonna Need It — that it saves programming effort — and rebuilds it on two financial costs. The first is optionality: structure built on a prediction usually misaligns with the feature that actually arrives, so you pay twice, once working around it and once tearing it out. "The value was never in the structure. The value was in the option to build the right structure once you knew." The second is net present value: "Money has time value. So do features." Building now for a feature three months out pulls cost forward and pushes revenue back. His hardest claim is that being right doesn't rescue you — "Perfect foresight doesn't save you, because the discounting doesn't care whether you were correct." In the AI era, he argues, cheap generation makes the violation more tempting, not the principle obsolete.

The thread audits the prose, not just the logic

The essay carries a disclaimer that its second half is "a genie-generated description of YAGNI" — AI-written. Readers in the thread noticed, then went looking.

"I fed it into Pangram and it came back as '70% AI generated'… I know that this author can write, so if this really was partially done with AI, it's disappointing." — a commenter who ran the detector

The complaint wasn't only provenance but texture: one reader found Beck's own introduction clear and the machine section opaque, quoting "the discounting doesn't care whether you were correct" as the kind of line that scans but resists meaning. Another regular was blunter.

"I don't know what happened to Kent Beck, a few of his recent articles seem to be devoid of substance." — a developer who recognized the pattern across recent posts

dang stepped in to keep the pile-on honest, quoting the guidelines: "Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something."

Where the actual disagreement is

The substantive pushback splits on whether AI changes the math at all. A founder argued the opposite of Beck — that AI has lowered the cost of restructuring, so the ability to safely change large parts of a codebase is now the thing to optimize for. A high-karma reply called that contradictory: if restructuring is cheaper, the value of pre-building for it goes down, not up. And a recurring counter-case is that YAGNI hides real debt — "table stakes" features developers pretend they won't need.

"YAGNI and related tenets were all excuses for 'we are consultants in a field we don't understand'." — a commenter with \~3.9k karma

The sting for Beck is that the strongest objection isn't to YAGNI but to the form: an essay arguing that cheap-to-produce work tempts you into building the wrong thing was, by its author's own label, half-produced by the cheapest writer available — and a 70%-AI verdict is the readership testing the thesis on the text itself.