As you know, open standards work and open source code development have some similarities. Both systems are designed to permit strangers to collaborate in joint design. Both have rules for discovering and filtering private patent or copyright claims, with the goal that end-users can access and use the outputs safely. While they’re two distinct systems, usually they are complementary, not in opposition.
Standards bodies are a microcosm of the marketplace: alternative ideas compete for inclusion in a draft. Some organizations permit royalty-bearing work to be included in a standard; some don’t; and some (like ours) permit each standards project to select a set of terms. In OASIS’ experience, the overwhelming majority of member-initiated projects choose a royalty-free approach, and so only accept contributions that do not come with royalty claims.
Demands for royalties are becoming rarer, in our own domain. Structured data standards that bear royalties have grown less acceptable to the marketplace, during this last decade of widely-distributed, Internet-driven information transactions. Proprietary and royalty-bearing methods have a wholly completely legitimate place in computing and ICT. But they are less likely to be embraced for use in globally-networked open systems that depend on user elaboration. In a multi-modal, constantly-growing open network, unlike a closed telecomm network (for example), participants add themselves spontaneously, and often modify shared data constructs. In such a system, to test each new participant and data exchange against whether proper rights are in place — perhaps many times a second, in the case of some networks — would pose obvious scalability challenges.
That may be why, in our field, standards panels often seek workarounds for proposed methods that have restrictive licenses. Whether to do so is, itself, the decision of the participants. The benefit of a transparent, stable process is that all contributors may participate, and conduct an open debate about what kind of terms ought to be required for a given project. We believe that this leads to better disclosure and better results.
In free and open source software work, like open standards, there are some established venues that operate under clear and known terms. The Open Source Initiative, the Free Software Foundation and the Apache Software Foundation are obvious examples; each with a transparent governance structure and a track record of stability. Many open standards consortia benefit greatly from those “FOSS” software development communities. As a practical matter, we see these two spheres interact constantly. Many FRAND standards have open source implementation projects which pose no license compatibility problems. See, for example Softwarefreedom.
A number of more informal models also have emerged. Not all of them have binding assurances for openness or re-usability in place. Ad hoc projects are created, to pool cooperative effort for a project, and may even be called “open-source” or “open-standards” work; but often these collaborations have no participation rules, or license or permission practices, to give assurance to the end-user. Increasingly, we see governments carefully distinguishing between such ad hoc efforts, and established stable processes, in their purchasing requirements and policy development.
Patent and copyright regimes do not always deal in certainties, even at their best. Some of the legal devices used in the above environments are relatively untested. One only needs to look at the Rambus cases in the US, to see swings in the interpretations that regulators and courts apply to the duties of participants in collaborative projects (e.g. 2006 case and 2008 case). A clear definition of the enforceable meaning of “fair” or “reasonable”, when the terms of proposed licenses are challenged, sometimes seems beyond a court’s reach. Also, issues around whether one may, or ought to, apply patent protections to software, generally, under the laws of various nations, is so well known an issue this year as to require no citations.
One recent response to these uncertainties, in our community, has been an increased use of waiver forms that may make licensure diligence, and perhaps enforcement details, obsolete. Standards participants may issue a statement that supports a standard — regardless of assertions about the state of their rights — by simply disclaiming any right to demand a license for a conforming implementation. See, for example, the model “Non-Assertion Covenant” . It’s likely that developer and user marketplaces will continue to seek agile and creative solutions, or work-arounds, to the traditional high-cost methodologies involved in parsing and threading through often poorly-mapped patent minefields.
Join the Forum discussion here: http://www.talkstandards.com/questions-for-standards-and-oss-open-forum/