Maybe this can help

https://simonwillison.net/2024/Oct/25/pelicans-on-a-bicycle/

The issue is that benchmarks that look insightful will end up being gamed by labs quickly (Goodharts law)

The best LLM benchmarks test around the margins of those behaviors, tasks that are difficult and correlate with usefulness while being removed enough to stay unpolluted

Yeah, i'm also surprised people just read post title and jump to conclusions ...

"Community Activity: How to guarantee you'll get free, fast support when you hit a bug."

Why is this a pillar of using open source in a business ? Take every thing for you enrichment and give back nothing ? What a f* mindset.

Thanks for useful unpolite feedback, and yes, opensource community can provide fast and free support at some extent

Corry's enshitification is in charge

So we are cycling back to kind of Waterfall Development ? No more agility ? The trend is dead by AI ?

"Waterfall" got a bad rep because it meant "we stay months in the requirements gathering, then months design phase, then months in development, then months in validation". If you compress "months" to days/hours, what you obtain is something that nobody from the 90s would recognize as "waterfall"; it is not the end of agility, far from it.

Using librairies, building TUI was easy before AI too.

Ben, that was one of my favorite "internet" stories... What a beautiful generational link!

Total resonance with this part :

"They’re writing TypeScript that compiles to JavaScript that runs in a V8 engine written in C++ that’s making system calls to an OS kernel that’s scheduling threads across cores they’ve never thought about, hitting RAM through a memory controller with caching layers they couldn’t diagram, all while npm pulls in 400 packages they’ve never read a line of."

Why not if you are free to fork it and do whatever you want with it ?

Maybe they can use an LLM to help them implement this feature ? Oh wait a second...