170
AI coding assistant refuses to write code, tells user to learn programming instead
(arstechnica.com)
this post was submitted on 13 Mar 2025
170 points (99.4% liked)
Not The Onion
14901 readers
1069 users here now
Welcome
We're not The Onion! Not affiliated with them in any way! Not operated by them in any way! All the news here is real!
The Rules
Posts must be:
- Links to news stories from...
- ...credible sources, with...
- ...their original headlines, that...
- ...would make people who see the headline think, “That has got to be a story from The Onion, America’s Finest News Source.”
Please also avoid duplicates.
Comments and post content must abide by the server rules for Lemmy.world and generally abstain from trollish, bigoted, or otherwise disruptive behavior that makes this community less fun for everyone.
And that’s basically it!
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Honestly, that's a smart thing for AI companies to do. AI is surprisingly decent at extrapolating from existing codebases, but it's useless at starting from scratch. If one model says "I can't do that, Dave" and another spits out garbage, you're getting the same amount of useful code out of both and a much better signal-to-noise ratio from the first.
imo it's the opposite, AI is good at starting projects by giving you boilerplate code, but bad at considering the full context of an existing project. Better to be doing the larger structure stuff yourself and only giving the LLM self contained tasks.
No, it's not smart, I pay for Cursor to generate code, not to patronize me. I would stop paying for it and instead switch to something that at least tries different ideas to get my feature to work.
(Also it's definitely not useless at starting from scratch, you just need a strong design and good understanding of what's possible)
What does it help you with? I can definitely see having snippets or "modular" code on hand as being useful, but you don't really need a LLM for that. What sets it apart? Is there a big time commitment necessary to get to the desired outputs or does it just do what you want right away?
It helps with avoiding learning templates in your IDE or literally any other feature that's been around for decades.
On mobile but two real quick examples:
The prompting in both of these cases started with an idea ("give me a cute penguin SVG logo") and refined ("make the beak smaller and more round.. add a purple background.." etc)
For more in-depth rather than one off features, this whole app was basically coded with AI (and I use it everyday, the quality is fantastic): https://play.google.com/store/apps/details?id=com.widget.uvindex
Hmm, I guess if you're happy with the output here that's all that matters. For me, the visual elements look really uninspired and mediocre. But if you're still in the startup/iteration stage then I suppose the unfinished look makes sense.
I guess it's a good place to start? Maybe? Depends if the code is easy to maintain. I was more interested in how you feel the AI adds to your existing coding workflow, but maybe you aren't a professional programmer? I am getting the impression that the AI is doing most of the work here but maybe I have the wrong idea.