Comment by frevib
4 hours ago
My Pavlov reaction when I see the Claude icon next to the committer is: not touching this. Is trust in OSS now completely gone?
4 hours ago
My Pavlov reaction when I see the Claude icon next to the committer is: not touching this. Is trust in OSS now completely gone?
The code generated by Claude is fully under my control, and I make sure it works exactly as I intend. I could have hidden this by committing it myself, but I choose to acknowledge that I had Claude's help
It’s not your fault. The problem is now that my mother in law can vibe code a complete operating system, I cannot distinguish quality from shit any more. And the last few OSS projects that I tried that had the Claude icon where shit and breaking in many ways.
I think your prior assumption that all OSS projects had a degree of quality is where your problem is. But I get it.
We’ve had low quality code for as long as we’ve had programmable machines. I’ve seen a lot of manually written bad code well before LLMs took over. Nowadays, I primarily look at how long a repo has been around and how active it is. A repo that has consistent activity is probably going to be a higher quality one. Even better if I can follow PRs to see how the code base is evolving.
I will admit it’s become more difficult to tell low quality git repos from higher quality ones. Missing files or incomplete README files are a thing of the past. Claude is very good at filling in the gaps to make all repos at least appear respectable.
It's kinda like eternal September but for software engineering. There was a time that just being on the internet, having an email address, etc. was a signal that indicated a level of technical proficiency. Now everyone has that connectivity, it doesn't mean anything anymore.
Same concept, but github repos now.
The Claude-generated README doesn't inspire confidence in that. I believe you checked all the code, but a lot of people are going to see that README and not give the project a second chance.
> I make sure it works exactly as I intend
I appreciate the effort. May I ask how do you tackle this?
This is also why I allow it to add these things to the commit. I think it's worth disclosing despite me checking everything it did myself.
As a former AI skeptic I get it and still get it.
But now other people's code is just a starting point for me forking an app to do exactly what I want it to - it's inspiration because code is cheap.
Of course for larger projects we will still need solid components underlying them, but what you're really talking about is just maturity - it's very easy to produce a lot of output now but you can't replace actual history or having more then exactly 1 maintainer.