Open data

The Open Definition, first published in 2005 and updated in 2015, sets the international standard for open data. It requires that the data be provided under an open license (such as Creative Commons Attribution or ODbL), be available in a machine-readable format, and allow for free use, reuse, and redistribution without discrimination against fields of endeavor.¹ The Open Data Institute adds that data must be accessible, available, and reusable, emphasizing the importance of clear licensing and technical accessibility.²

Open data is distinct from open source software, though they share philosophical roots. The term was popularized in the 2000s, building on earlier movements for open access to scientific research and public sector information. The Open Government movement, particularly under the Obama administration, accelerated adoption of open data policies in the United States.³

Governments are the largest producers of open data, publishing everything from weather data to census statistics. The U.S. government's data.gov portal, launched in 2009, hosts over 200,000 datasets from federal agencies.³ The G8 Open Data Charter of 2013 committed member states to release high-value data by default, including data on budgets, elections, and geospatial information.⁴

Open data policies have been linked to economic growth and improved public services. A 2014 study by the McKinsey Global Institute estimated that open data could generate $3 trillion in value annually across sectors such as health, transportation, and consumer products. However, implementation varies widely; many governments struggle with data quality, privacy concerns, and the cost of maintaining portals.

In the scientific community, open data is a cornerstone of the open science movement. Funding agencies like the National Science Foundation and the European Commission now require data management plans and often mandate open access to research data.⁵ The FAIR principles (Findable, Accessible, Interoperable, Reusable) provide a framework for making data open and useful.

Open data in science has led to significant breakthroughs, such as the Human Genome Project, which made its data freely available, accelerating genetic research worldwide.⁶ However, challenges remain, including the cost of data curation, the risk of misuse, and the tension between openness and the protection of human subjects.

Beyond government and science, open data has niche applications. For example, the OpenStreetMap project is a crowdsourced open map that has become a critical resource for humanitarian aid and disaster response.⁷ In the arts, the Metropolitan Museum of Art released over 400,000 images of public-domain works under a Creative Commons Zero license, enabling new forms of creative reuse.

Open data also has a dark side: it can be used for surveillance or to reinforce biases. The Open Data Watch organization monitors the quality of open data in developing countries, highlighting gaps in coverage and timeliness.⁸ The concept of "open washing" — where organizations label data as open without meeting the full criteria — is a growing concern.

Open Definition

A set of principles defining open data, requiring free access, reuse, and redistribution.

FAIR principles

Guidelines for making data Findable, Accessible, Interoperable, and Reusable.

Open washing

The practice of labeling data as open when it does not fully meet open data standards.