Stop Slopware: Build Small, Clean, Human-Written Software

This piece defines “slopware,” criticizes misuse of AI that amplifies poor software practices, and reassures beginners while urging authentic writing and deliberate learning. It offers concrete steps to fix existing projects and guidelines for starting new ones focused on clarity, small scope, and maintainability. The site serves as a quick, honest link to deliver constructive feedback and encourage reflection.

Key Points

Slopware is low-effort, sloppy, noisy, and unmaintainable software, and misuse of AI makes it worse.
Beginners are encouraged to learn openly, but to avoid overreliance on AI, which weakens learning and authenticity.
Fix existing projects by slowing down, cutting clutter, rewriting what you understand, learning what you don’t, and making every detail reasoned and intentional.
Start new projects by solving one real problem cleanly, keeping scope small, writing your own README, and using AI sparingly.
The website provides a quick, honest link to share constructive feedback without repeating the same critiques.

Sentiment

The community leans moderately in favor of the article's core message — that careless, low-quality AI-generated software flooding open-source ecosystems is a real problem. However, there is significant pushback on the article's tone and framing. Many commenters feel it crosses from constructive criticism into gatekeeping territory, and a vocal minority dismisses it as virtue signaling. The strongest agreement comes around concrete harms: wasted maintainer time, deceptive project presentation, and degraded ecosystem signal-to-noise ratios. The strongest disagreement centers on whether AI itself is the problem or just poor usage of it, and whether the article unfairly discourages newcomers from using powerful tools.

In Agreement

AI-generated open-source submissions waste maintainers' time because the code is typically low-quality and the submitters cannot evaluate or fix what they don't understand
The problem mirrors AI slop flooding other creative platforms like fiction magazines, streaming music, and Etsy, where low-effort automated output degrades the ecosystem for everyone
The core issue is false advertising: AI-generated projects present themselves with the appearance of competent design but fail on basic functionality, wasting evaluators' time
Over-reliance on LLMs is genuinely harmful for junior developers' learning, as they skip the friction that builds foundational understanding
Publishing open-source software carries an implicit stewardship responsibility toward the shared commons, including clarity of purpose and honesty about limitations
If AI-produced code were actually good, nobody would object to it — the complaint is specifically about poor quality being obscured by polished presentation
The signal-to-noise ratio for discovering quality libraries and tools is already dropping and will continue to worsen

Opposed

The article reads as virtue signaling and gatekeeping, dismissing newcomers and non-traditional developers who use AI productively
Software that works and solves real problems has value regardless of how it was created or its internal code quality
AI is an excellent learning tool when used intentionally, and telling beginners to avoid it entirely is counterproductive advice
The craftsmanship movement in software has historically been used for heavy gatekeeping in code reviews, job interviews, and workplaces
For personal tools and private use cases, AI-generated code with low quality is perfectly acceptable and dramatically more efficient
Most working developers are just solving business problems — expecting craft and passion from everyone is unrealistic and elitist
Critics of AI-generated code are motivated by fear of job loss and skill devaluation rather than genuine quality concerns