Science is undergoing a shift toward openness—but the tooling for working with data still lags behind. Researchers struggle with reproducibility, fragmented datasets, and collaboration workflows that don’t reflect how modern science actually happens.

DatWeb is an open-source project built on peer-to-peer principles that brings version-controlled data and custom tooling directly to the web. Inspired by the original Dat protocol, it enables scientists, educators, and open data communities to build and share their own data-driven applications with just a few lines of code.

Instead of relying on centralized platforms, DatWeb allows teams to:

• share evolving datasets with built-in versioning
• create reproducible workflows
• build lightweight, domain-specific tools tailored to their research

In this talk, we’ll explore how decentralized data infrastructure can support open science practices such as reproducibility, collaboration across institutions, and long-term data access. Through live examples, we’ll demonstrate how DatWeb is being used to prototype custom tools for scientific and educational use cases—without requiring large engineering teams or complex infrastructure.

Whether you're a researcher, developer, or newcomer to open source, this session will show how peer-to-peer technologies can unlock a new generation of open, composable tools for science and education.