In brief:

  1. Open data means public data made available in an open, reusable format, including for commercial use.
  2. A dataset only counts as truly open if it meets four cumulative criteria: a usable format, a free reuse license, no cost and no prior formality.
  3. The publication obligation applies to local authorities and administrations above certain size thresholds, not to every public structure.
  4. Once published, this data feeds both citizen-facing uses and internal management platforms.

Open data covers a precise movement: making data produced or held by a public body available in a format that lets anyone retrieve it, combine it with other sources and reuse it without prior authorization. The term has circulated in local government for years, yet confusion remains common between simply putting documents online and genuinely opening up data in the sense the regulations intend.

What open data means

Open data refers to the policy of publishing public data in a structured, machine readable and freely reusable format. It differs from classic document transparency, which consists of publishing a report or a council decision as human readable text that is hard to process automatically.

An open dataset keeps circulating independently of the software that produced it. A spreadsheet, a geographic shapefile or a structured feed all keep their usefulness once extracted from the original tool, unlike a plain screenshot or a scanned printed document.

The criteria that define a truly open dataset

Four cumulative conditions separate a genuinely open dataset from a plain document upload.

A usable, documented format

The first criterion concerns the format. An open dataset must be available in a structured, documented format, tabular or geographic depending on its nature, so that software can read it without manual conversion first. A plain office document put online does not meet this criterion, even when its content genuinely comes from a public source.

A free reuse license

The second criterion concerns the license. A dataset is only truly open if its reuse, including for commercial purposes, is allowed without any prior request to the producer. The two remaining criteria complete this foundation: no access cost, and no registration formality that would slow reuse down.

Who must publish data as open data

The publication obligation does not apply uniformly to every public structure. It depends both on the size of the structure and on the nature of the data involved.

The local authorities concerned

The obligation applies first to local authorities whose population and staff size exceed certain thresholds set by regulation. Below those thresholds, publication remains a voluntary step, most often driven by a transparency goal or by a broader local digital strategy.

State administrations and public bodies

State administrations and a range of public bodies are subject to the same logic for the data they produce in the course of their duties. The exact scope depends on the nature of the structure and the type of data involved, which explains why implementation has rolled out gradually rather than all at once.

What the publication obligation actually involves

Publishing open data is not just about dropping a file on a website. It means choosing a compliant format, documenting the structure of the data and keeping it up to date as the source changes. Many local authorities rely for this on the same territorial data platform that already feeds their internal uses, which avoids duplicating the collection chain just for public release. The available options are presented in the comparison of data management solutions for local authorities.

The scope of the data covered

The scope covers data already produced as part of the structure’s duties, not an obligation to create new data for the sake of publication. A local authority that does not hold a given dataset has no obligation to build it solely to meet this requirement.

The exceptions provided for

Data whose publication would harm privacy, personal safety or secrets protected by law stays outside the scope, as do unfinished or preparatory documents. These exceptions bound the obligation without emptying it of substance.

What opening up data makes possible

Once published, open data serves uses that go well beyond the single administration that produced it. Nonprofits, researchers or companies combine it with other sources to build services the local authority would not have developed on its own. Internally, the same data also feeds into the flows behind urban hypervision, as long as the format stays usable without heavy reprocessing.

The general public, for its part, most often accesses this data not as a raw file but through an interface that shapes it for them. That is the role of tools built to visualise a territory’s data without requiring the user to handle a technical format directly.

Frequently asked questions

What is the principle behind open data?

Open data means making data produced by a public body available in a structured, freely reusable format, including for commercial purposes. The principle is distribution without prior formality, unlike plain document transparency, where content stays human readable but is hard for software to process automatically.

What criteria define a truly open dataset?

A dataset counts as open when it meets four cumulative conditions: a structured, documented format, a free reuse license that covers commercial use, no access cost, and no registration requirement. Missing any single one of these criteria turns the release into a plain document upload rather than open data.

Who is required to publish data as open data?

The obligation applies to local authorities whose population and staff size exceed certain thresholds set by regulation, as well as a range of state administrations and public bodies for the data they produce in the course of their duties. Below those thresholds, publication remains a voluntary step.

What are the benefits of opening up public data?

Opening up data lets third parties, nonprofits, researchers or companies, build services the public body would not have developed on its own, while reinforcing the transparency of public action. Internally, the same data often feeds the same flows used to manage services, with no extra rework when the format stays usable as is.