Individual judgement about the presence of software freedom in a license is not the same as community consensus expressed through OSI approval.
Does it really matter if a copyright license is OSI Approved or not? Surely if it looks like it meets the benchmark that’s all that matters? I think that’s the wrong answer, and that OSI license approval is the crucial innovation that’s driven the open source revolution.
“Open Source” describes a subset of free software that is made available under a copyright license approved by the Open Source Initiative as conforming with the Open Source Definition. Having a standards body for licenses — one which ratifies the consensus of an open community of license reviewers — saves individuals from needing to each seek out a legal advisor to tell them whether a given license does in fact give them the rights they need to build or deploy the software they want. By providing easy certainty, open source gives people permission in advance to meet their own needs and innovate with technology.
The only arbiter of OSD compliance is the license review process conducted collaboratively by the open source community and summarised and ratified by the OSI Board of Directors. Others have no role outside this process and are not entitled to assert that a non-approved license satisfies the OSD. As such, licenses that have not received OSI approval don’t satisfy the process and can’t be considered open source.
The strength of OSI’s approach is that it is objective; a license is either on the approved list or it is not. Licenses on the list are known to give permission in advance and unlock software freedom; those not on the list cannot be guaranteed to do either. The FSF uses a subjective approach that encourages speculation about whether a license is “free”. Meanwhile there are many with vested interests in diluting free and open source software who want a subjective approach where every individual may act as their own arbiter. Despite these pressures, it’s the OSI’s approach that has made open source succeed.
That’s not because a legalistic tick-in-the-box is really interesting. Rather, it’s because developers can gain certainty as to whether they can use a project simply by checking its approval status. No-one has to be asked for permission or clarification. Significantly, there’s no need to retain a lawyer just to check the license is in fact safe to use.
It’s easy to get overwhelmed by all the details of the many open source licenses, losing sight of the reason they are important. They’re important because every open source license guarantees the freedom to innovate without seeking permission first. OSI approval means you have the unconditional right to use the software in question for any purpose (sometimes calls “freedom zero”). You also have an unconditional right to make new software based on that software for your own use, and a conditional right to share the software — modified or not — with other people. The final case comes with some complexities beyond the scope of this article, especially for copyleft licenses.
That freedom to innovate, unlocked by the permissions the OSI-approved license guarantee in advance, is the powerhouse of open source. Developers know they can incorporate open source components without seeking legal advice. Users know they can deploy the software with the confidence that they have a license and won’t be persecuted by rent-seeking proprietary software companies. Together, this liberty has realized the potential of free software and propelled open source to dominance over the course of the last decade.
(A variant of this article was published in the Linux Voice section of issue 199 of Linux Magazine, June 2017)