[gamedever]

Is this different than "I'm glad calculators didn't exist when I learned math"?

I'm mixed. I did try to get gemini to figure out how to do some fancy TypeScript stuff. It provided solutions but failed to meet the constraints and couldn't get it to really get what I meant. It would say "Oh, sorry, here's one dealing with that" and then spit out code that either still had the same problem or ignored a previous constraint.

Anyway, more relevant, I did wonder if I wasn't trying enough on my own to figure out what I was trying to do. I failed either way haha.

[genewitch]

Gemini is far and away my least favorite llm. So far I've had OK luck with windows copilot for a simple react site. The site was done in like 2 hours but I polished it up while the person I did it for was copyediting their whitepaper for the "site". I had 5 tabs in notepad++ and was saving it to a cifs share so I could test immediately. React is pretty snazzy but js with its decorations doesn't look quite right to me yet.

I used deepseek-R1-qwen-distill-15B to do a nodejs discord bot. That was done incredibly fast. I had asked for a few days for a cellphone app but llama, openai (didn't test 4o or newer yet), deepseek didn't let me know that what I was asking was technically impossible - service workers without a backend somewhere so I could keep track of timers and fire notifications. Like a PWA.

After asking copilot about it the other way around "can service workers URL be file:///?" Apparently not.

All of the serviceworker nonsense LLM gave me the two-step - make change, revert change, repeat.