Live data from Hacker News

From Today, Software Engineering Is Dead

building138.com

11–13 of 13 posts

Re: From Today, Software Engineering Is Dead

#11
post #6

> Before LLMs I would have to spend weeks or more implementing a design decision, and maybe weeks or months if I had to refactor one I wonder why people think that is a good argument. Spending weeks on a single design decision makes the case that you’re either not a good developer, your time was stolen by other concerns, or that design decision was a major process change. In the last case, this is one of the goals of…

Why do you think what you said was a good argument. You said, "hes dumb or distracted or deserves it" You just made lazy moralizing arguments.

Re: From Today, Software Engineering Is Dead

#12
"But the nature of painting changed."

No it didn't. You can go find wildly talented painters today who work in incredibly realist styles…in fact just a few years ago as visual social media really took off, there were many challenges to "fool" people into thinking they were looking at a photograph when it was actually a painting.

The whole "painters didn't have to paint a realist style anymore because of cameras" is an ad-hoc rationalization. That was entirely due to taste and other cultural factors, nothing really technological about it whatsoever. The only claim you could convincingly make is that painters _who made a living painting portraits for random people who hired them to paint their portrait_ had to pivot because now people could hang a cheaper photograph of themselves and their family on the wall instead of the more expensive painted portrait. That's about it really.

Handcoding great software for people who want it is alive and well. It's just not at all the industry "software engineers" who can't see past what Google or Meta like to shove out of their corporate offices have been working in from the jump.

Re: From Today, Software Engineering Is Dead

#13
post #6

> Before LLMs I would have to spend weeks or more implementing a design decision, and maybe weeks or months if I had to refactor one I wonder why people think that is a good argument. Spending weeks on a single design decision makes the case that you’re either not a good developer, your time was stolen by other concerns, or that design decision was a major process change. In the last case, this is one of the goals of…

Yes, and I'd argue having to spend weeks on a single design decision is an extremely good thing. The worst thing that can happen to software development is for it to become trivial to pump out reams of new "software". That only means that 99% of software will be freakishly terrible on an unimaginable level.
Post reply on HN