Hacker News Reader

You are wrong though, reasoning produces way better results as evidenced by benchmarks. The amount of bullshitting reduces drastically - just like when humans think before giving an answer.

Why should we accept your anecdotal evidence in favour of statistical evidence on the contrary?

View on HN

> reasoning produces way better results

I doubt the whole concept of calling it "thinking" or "reasoning". If it's automated context engineering, call it that. The bullshit is in the terms used.

Why is thinking not a good word to use for it?

Because it unnecessarily anthropomorphizes it to create the illusion that there is intelligence behind it (in the traditional sense, where the intelligence is synonymous with being aware of its existence and embodied)

But I personally don't have a big problem with the term in this context. Our industry have been using misleading terms since the beginning to describe things that somewhat resemble whatever it's called after.

Like literally from the start, "bootstrapping"

Reply on news.ycombinator.com