As part of the research project on “The Interaction between Open Source Software and FRAND licensing in Standardisation”, a workshop was organised by the European Commission, Joint Research Centre (JRC) in collaboration with Directorate General Communications Networks, Content and Technology (CONNECT) to present and discuss the intermediate results to date. The workshop took place in Brussels on September 18, 2018. I presented a set of observations from the research on the case studies performed as part of the project that are outlined below. Other speakers where Catharina Maracke on the issue of legal compliance between Open Source and FRAND licenses, Bruce Perens on “Community Dynamics in Open Source”, and Andy Updegrove on “Dynamics in Standardisation”.
You may ask what the relevance of this debate is for the wider Free and Open Source Software community. The obvious answer is that to distribute software “without restriction”, the user needs all the usage rights associated with the program. While most FOSS contributors assume that this is naturally the central motivation for anybody to contribute in the first place, there is a long history of attempts to maintain some sort of exclusive control over a piece of FOSS code, possibly using other rights than copyright.
Contributor license agreements (CLA) are a well known instrument for that. The original author agrees to an exclusive or non-exclusive license to a central entity (usually a community organisation or a dominating company) to use their work. This may just give a community organisation a right to adopt newer licenses. Or it may give a single Open Source vendor the right to use other contributor’s work commercially. Whether CLA embody FOSS culture or violate them by creating an asymmetric distribution of benefits between contributors and a company therefore depends on the specific terms of the CLA. The contributor agreements configurator helps projects to set up a CLA and understand what the motivation behind the terms are (Endocode contributed the implementation). CLA are controversial, especially if they include copyright assignments to a single company. Projects and contributors are well-advised to make sure to fully understand what they are agreeing to. The main take-away for the argument here is that the norm in FOSS communities is a symmetric distribution of rights between all contributors, and CLA are one instrument to modify this to the benefit of a single entity.
But that is not all. It just helps to explain how such asymmetric rights can be used to give a single entity more power in a community of contributors. There is a long history of such attempts, and they usually involve some other legal rights beyond the terms of the FOSS license applied. There is a good reason for that: FOSS licenses give contributors and users a time-tested and well-balanced framework for collaboration. “Use, study, modify, redistribute” is a simple, easy to understand mantra that every FOSS contributor knows and understands. CLA augment the FOSS license terms with an additional contractual agreement, but for the most part stay in the realm of copyright just like FOSS licenses. Other instruments not so much:
- Trademarks are used to enforce a vendor’s expectations when others adopt and ship their software. This is used by communities to protect their names, like in the Mozilla Trademark Policy, or to make sure specific services are included in devices, like in the Android Compatibility Program.
- Trade secrets like delayed code releases under FOSS licenses or proprietary “extensions” that are effectively required for the software to be useful are used to maintain an edge of an “enterprise” version over the FOSS release.
- The software patents debate may have slowed down, but is still relevant. Patents exist, and applications for patents that include claims that are implemented in existing FOSS solutions never stopped. It has become harder to be granted a patent for a software-only solution. Institutions like the Open Invention Network create a safe space where FOSS innovators cross-license freely and avoid litigation. However vendors shipping FOSS in products still need to clear potential patent claims implemented in those. Conformance with patents does not usually follow from FOSS licenses.
Which leads to round five of the match with software freedom in one corner and exclusivity in the other – standard-essential patents (SEP). Everybody loves standards, and FOSS especially: We use standardised programming languages, unicode tables and system call interfaces. Standards are the foundation for interoperability, and some standards like the Open Document Format are the result of a long struggle to establish formats that re-open markets for competition. All of these standards are Open Standards that can be implemented in FOSS without restrictions. But what if implementing a standard would be required to interoperate with others, or even mandatory under existing regulations, and this standard would include functionality that is covered by patent claims, possibly with royalty-bearing patent licenses? That is the definition of SEP, which are commonly required to be licensed either royalty-free or under “fair, reasonable and non-discriminatory” (FRAND) licenses. It is also a relevant and current concern because the European Commission considers SEP under FRAND licenses a useful instrument for industrial policy in the European Union. Until now, the wider FOSS community has contributed some positions to this debate that is for the most part dominated by incumbent industries, recognized standards-development organisations (SDO) and policy makers.
The study currently conducted under the supervision by the European Commission Joint Research Centre (EC-JRC) on “The Interaction between Open Source Software and FRAND licensing in Standardisation” aims at providing a better understanding of the interplay between FOSS licenses, governance and the IPR policies of SDO. At the workshop that was just held in Brussels as part of this study where an initial set of case studies on the subject have been presented. Theses case studies (that I am the author of) will become part of the overall study report prepared by Knut Blind and me. As the study progresses, a number of early observations from these cases will be shared in this blog and other channels. It is important that the wider FOSS community recognizes the relevance and importance of this subject and actively makes itself heard. CERN, the ECMAscript community, the Eclipse Foundation, the Linux Foundation and others have already contributed insights to the field work for the project. Others, especially FSFE, have been notably absent from the debate. The Open Source Initiative is represented on the project’s steering committee.
The standards that will be shaped by this debate, like 5G mobile communication or IoT protocols, will affect us all, as will welfare gains or losses from policies that influence the innovativeness of FOSS. There are opportunities to contribute: communities can communicate what their needs are with regards to standards and FOSS. Contributors can provide input to the case studies or the stakeholder survey that will be performed in the next phase of the study. Feel free to reach out if you would like to participate.