Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by wesselbindt on Sunday May 03, 2026 @03:38AM
(#47994423)
I'm still confused as to why folks don't just write executable specs.
[ Parent
| on HN ]
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by cenamus on Sunday May 03, 2026 @03:58AM
(#47994552)
So basically tests?
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by MoreQARespect on Sunday May 03, 2026 @04:21AM
(#47994712)
Yes, except a test can be turing complete - i.e. code.
An executable spec like gherkin or hitchstory is config - it has no loops or conditionals. There are a number of rarely recognized benefits to this.
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by carlbarrdahl on Sunday May 03, 2026 @03:50AM
(#47994493)
Could you expand on this?
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by booi on Sunday May 03, 2026 @03:51AM
(#47994503)
code
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by arikrahman on Sunday May 03, 2026 @03:57AM
(#47994550)
Literate programming would provide specs and code instead of working backwards from hard coded functions to figure out specs.
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by cassianoleal on Sunday May 03, 2026 @05:54AM
(#47995301)
> working backwards from hard coded functions to figure out specs.
People do that? Actual professionals?
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by arikrahman on Sunday May 03, 2026 @04:36PM
(#48001190)
Yes all the time actually, especially when making system migrations. Uncle Bob writes about this in Clean Code. It is actually extremely common since it's the most apparent course of action to look at hard-coded values in maintenance and backport it as specs to new system.
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by fudgeonastick on Sunday May 03, 2026 @03:56AM
(#47994542)
If you're confused, and have tried Opus for coding, I'm keen to hear what problems or workflows it's not good at.
If you're genuinely confused, and haven't tried Opus for coding, then it's not surprising you're confused!
It is also okay for you to just not like the idea of LLMs for coding (but say that!).
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by oytis on Sunday May 03, 2026 @04:09AM
(#47994634)
That's what the article is about - overcoming problems with AI cooding tools using specs in Yaml. If we've got that far, it might be better to write specs in a proper programming language instead and skip the AI layer altogether
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by adi_kurian on Sunday May 03, 2026 @04:47AM
(#47994882)
Think the idea is to still get monumental acceleration between fancy YAML specs (bullet points with some indentation that an intelligent technical manager could write) and production ready code.
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by oytis on Sunday May 03, 2026 @10:11AM
(#47997128)
I have yet to see the monumental acceleration tbh. For people who never tried writing code themselves a semi-formal spec is definitely a step in the right direction though
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by wiseowise on Sunday May 03, 2026 @04:54AM
(#47994923)
I’m using Opus 4.6 and I’m so confused! Maybe I should try Opus 4.7, which is almost twice as expensive to get some clarity (but not too much, I need to save money for Opus 4.8)?
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by eterm on Sunday May 03, 2026 @04:19AM
(#47994693)
Ambiguity is the grease that keeps everything turning.
Re:Specsmaxxing – On overcoming AI psychosis, and why I writ...
by mike_hearn on Sunday May 03, 2026 @06:05AM
(#47995358)
Some of us do! That's called Gherkin.