Hacker News Reader

Comment by xienze | Hacker News Reader

I find it amusing that software developers have no issue with having an LLM churn out slop code but have such a visceral reaction to slop articles.

View on HN

You are falling into the trap of thinking there's a single monolithic being called Software Developers that has inconsistent opinions. In fact, you're observing different people with conflicting values.

Yeah yeah. But LLMs certainly have been embraced by a large number of developers. Many of whom I've observed react with disgust when they see "not X, but Y" or emdashes in an article. But when it comes to code, "wow this is so awesome!"

One is made for humans to consume, the other for a compiler or interpreter. Good code is supposed to look like other code and follow common patterns. The best writing is original and novel.

I have no issue with with code generated by e.g. Claude because it's not "slop".

On average, it's probably better than the code I would write.

I say "on average" because AI doesn't make stupid mistakes, doesn't invert logical conditions. I know I do. Which I eventually fix, but it's better to not make them in the first place, hence "on average".

And in cases that AI doesn't generate code up to my quality standards, I re-prompt it until it does. Or fix it myself.

I'm not a hapless victim of AI. I'm a supervisor. I operate a machine that generates good code most of the time but not all of the time. I'm there to spot and correct the "not all of the time" cases.

But that's my point. LLMs generate good prose "most of the time", certainly better than most people are capable of doing. Yet we frequently react with disgust when we see tell-tale signs of LLM-generated text in articles. Why? Because it indicates the person was probably too lazy to write it themselves and are simply chucking a half-formed thought over the wall? Why don't we hold generated code to the same standard?

Reply on news.ycombinator.com