We don't need more apps. We need better open source
First reported by Jordankrueger ·
Making open-source software easier to install and run reduces the technical barrier for organizations to control their own data, moving away from expensive cloud services.
Jordan Krueger's blog post argues that the current trend of building numerous small, AI-assisted applications, often referred to as 'vibecoding,' overlooks a more impactful area: improving existing open-source software. Krueger, who assists progressive nonprofits with technology and self-hosts many services, highlights that while open-source tools are often functionally excellent, they frequently suffer from poor performance, difficult installation processes, and high resource consumption. These issues prevent organizations from migrating away from costly Big Tech services due to the need for specialized system administration skills. Krueger proposes focusing AI-assisted coding efforts on four key areas: making software faster, simplifying installation, improving usability for day-to-day operation, and reducing resource requirements. The goal is to make open-source software more accessible and sustainable for individuals and smaller organizations.
The proliferation of AI-assisted coding is enabling rapid creation of new applications, but a significant opportunity lies in applying this capability to enhance the reliability, performance, and ease of use of foundational open-source projects. This shift from building novel, single-purpose apps to refining existing infrastructure could democratize technology ownership by reducing the need for specialized IT expertise within smaller organizations. By focusing on areas like faster execution, one-click installations, and reduced memory footprints, open-source software can become a more viable and attractive alternative to proprietary cloud solutions, aligning with the growing demand for data privacy and control.
This approach to software development, which prioritizes the improvement of mature open-source tools over the creation of countless new applications, has political and economic implications. It empowers organizations, particularly nonprofits, to adopt self-hosted solutions, thereby reducing reliance on a few dominant tech companies and retaining greater control over their data and operations. The author also cautions developers to engage respectfully with open-source maintainers, emphasizing collaboration, clear measurement, and small, well-tested contributions, especially when leveraging AI tools to avoid overwhelming maintainers with low-quality or unsolicited changes.
AI-written summary. May contain errors.