The open data movement produced a genuine shift. Satellite archives, weather records, and hydrological measurements that once required institutional access are now downloadable by anyone. The barrier of permission has largely fallen.
What remains is less visible and, in practice, just as limiting.
Bandwidth, hardware, and time
Working with large datasets requires reliable internet, capable computers, and unstructured hours. All three are unevenly distributed. A student with a shared laptop and metered data has technical access to the same archive as a research lab and cannot use it the same way.
Knowing the data exists
Discovery is a real barrier. Most people never learn which datasets are public, how they are documented, or which processing level to use. That knowledge circulates informally through universities and research groups, which means it follows existing networks.
Interpretation
Environmental datasets carry assumptions, known biases, and quirks documented in technical papers. Using them responsibly requires reading that documentation, and the skill to do so is taught rather than intuited.
What follows from this
Publishing data is a starting point. Tutorials, worked examples, mentorship, and tools that hide unnecessary complexity are what convert availability into use. This is a large part of why we build education programs alongside software.