Post

Never let an LLM speak for you!

To each their own, I guess. In general I find the discourse around this lacking some nuance. If we agree to move forward with some PR then we can improve — in this case — Knip.

While I totally get most folks don't like it, I can just read through it and see if there's value for the project or not. Knip got better for it lately and I welcome anyone to prove me wrong, show me slop that got merged. Note that I've also rejected plenty.

The human must stay in the loop, such a shame in too many cases the overall experience is ruined these days. I'm positive it'll be better. We should first and foremost take care of ourselves before we can help others indeed 🫶

The only idea I have right now is to throw more AI at it from my side, which doesn't feel like the right approach. But getting 5 slop PRs for open issues is not great when it's not even proven that the issue is valid :/

I always wonder: where does it lead? Ah want to stay sane while working in software. You need to use AI as a filter for everything... it feels a little dark lol Invent a spam problem, then force me to pay for AI to prevent AI spam lol.

Some people don't bother talking anymore or trying to understand the issue/project/roadmap. I get walls of text pasted from an LLM. In Slack/Discord, I wonder if it's even a real person. A few times now, it's just been a bot using the person's account. Cripples my motivation a little...

Different people, different priorities. Different project, different rules. Walls of text are annoying for sure, but I feel like that's the low-hanging fruit solved easily. In a chatbox you should def know if you're talking to a human or not, that feels odd.