The European Parliament’s Committee on Legal Affairs (JURI) rapporteur Brando Benifei has published his draft opinion on the Digital Omnibus that contains some welcome attempts to preserve the core principles of Europe’s open government data framework. Whereas the Commission proposal risks dismantling the Open Data Directive’s “open by default” approach, the JURI rapporteur rightly recognises that open data remains essential. Unfortunately, this recognition is not consistently reflected in the rapporteur’s proposed amendments. While some amendments seek to reinforce open data principles, others would undermine those same principles by allowing licensing practices that are fundamentally incompatible with them.
What the draft opinion gets right: defending open data principles
The draft opinion correctly identifies the importance of maintaining the current open data ecosystem that enables public sector information to flow freely across use cases. Amendment 58 strengthens Article 32r(2) by clarifying that conditions attached to the re-use of public sector information shall not undermine “the principle of open data as freely reusable for any purpose.” Amendment 59 improves Article 32r(3) by requiring that standard licences used by public sector bodies should, by default, be “open licences enabling free and unrestricted re-use.”
By reaffirming the principle of non-discriminatory re-use of public sector data and documents, the rapporteur re-aligns the Digital Omnibus with one of the foundational principles of the Open Data Directive. And by expressly requiring the use of open licences, the draft opinion goes a step further than the Directive, which merely encouraged the use of standard licences.
These are positive improvements that reflect an important understanding: standard open licences are a key driver of research and innovation, because they provide legal certainty, eliminate transaction costs, and enable the interoperability of information originating from different public sector sources. Unfortunately, that understanding disappears in the very next amendment.
What the draft opinion gets wrong: open licences and special re-use conditions cannot coexist
While the rapporteur’s amendments to Article 32r(2) and (3) defend open data and open licensing principles, its amendments to paragraph 4 retain the Commission’s proposal allowing public sector bodies to establish “special conditions for the re-use of data and documents by very large enterprises”.
This creates a fundamental contradiction. Open data is based on the principle that everyone receives the same permissions to re-use information. Standard open licences such as CC0 and CC BY grant universal rights that are not dependent on who the user is. The moment a public sector body introduces actor-specific conditions for certain categories of users, the licence is no longer open in the traditional sense.
To its credit, the rapporteur attempts to shield certain categories of users from the negative consequences of differentiated licensing. Amendment 60 specifies that any special conditions imposed on very large enterprises should not “restrict or negatively affect re-use by research organisations, non-commercial users, or open knowledge projects.” However, this safeguard does not solve the underlying problem.
If public sector bodies are permitted to establish special conditions for “very large enterprises”, they will inevitably move away from standard open licences and towards customised, conditional licensing arrangements. They cannot simultaneously make public sector information available under open licences and reserve the right to impose special re-use conditions on selected users.
The way forward: differentiated charging, not differentiated licensing
The good news is that policymakers do not need to choose between preserving open licences and addressing concerns about the market power of very large enterprises. Public sector bodies can legitimately charge very large enterprises more for access services such as APIs, bulk downloads, or other forms of data supply.
Charging and licensing serve different purposes and should remain separate policy tools. By focusing on the conditions for supplying data rather than the conditions for re-using data, legislators could address concerns about economic asymmetries without undermining the foundations of the open data framework. Such an approach would allow governments and public institutions to capture more value from large commercial actors while preserving the legal certainty and interoperability provided by standard open licences.
A missed opportunity: fixing exclusive agreements
The draft opinion also misses an important opportunity in another area: exclusive digitisation agreements involving Europe’s cultural heritage collections.
The Digital Omnibus incorporates the existing Open Data Directive rules governing exclusive arrangements for the digitisation of cultural resources. Yet the current framework has proven ineffective in practice. While the Directive states that exclusivity periods should “in general” not exceed ten years, there is no evidence that existing agreements have been reviewed as required, and some arrangements continue to grant long-term or even perpetual exclusive rights. As a result, an estimated two million Public Domain books from European library collections remain subject to exclusivity restrictions that prevent broader access and re-use, including for AI training purposes.
The Digital Omnibus provides a timely opportunity to address this weakness and expand access to valuable Public Domain materials held by Europe’s cultural heritage institutions. A straightforward amendment to Article 32k(3) could replace the current principle that exclusivity periods should “in general” not exceed ten years with a binding 10-year ceiling. This would prevent indefinite extensions of exclusive arrangements and ensure that digitised Public Domain works become genuinely available for re-use after a reasonable period.
Edit notice: An earlier version of this blog post incorrectly stated that the JURI Committee had adopted the draft opinion discussed above. The document is in fact the rapporteur’s draft opinion and has not yet been adopted by the Committee. We have updated the text accordingly.