Skip to main content

Equitable community software

Today Richard Stallman announced that he has been reinstated as a board member of the Free Software Foundation with the antagonistic comment:

“Some of you will be happy about this and some of you will be disappointed. … In any case, that's how it is and I'm not planning to resign a second time.”

RMS originally stepped down after defending the late Marvin Minsky, an AI pioneer accused of assaulting an underage girl trafficked by Jeffrey Epstein. Epstein’s victims, RMS alleged, were likely “entirely willing”, and sex with a minor should not be considered rape. In the most generous possible interpretation, he was utterly clueless in a way that caused harm to women both inside and outside the movement, and to the free software movement itself.

His reinstatement was not announced on official Free Software Foundation channels, but rather informally in a talk given by RMS himself. He has since reappeared on the board page of the FSF website.

It’s a colossal own goal for the figurehead organization of the free software movement - and more so, for the movement itself, which in many corners still holds the Foundation and its work in high regard. It’s a clear example of how considering software as if it sits in a vacuum - outside of the social context in which it is run and built - can perpetuate harms, regardless of the intentions of its participants. It’s also yet another reason for women to walk away from the movement.

Software development communities often talk about running through rough consensus and running code. But who is consensus achieved between, and who is the code run for the benefit of? Traditional free and open source software models don’t consider this. The result is infamously hard to use interfaces and opaque documentation, as well as hostile, monocultural communities.

We need to move from a source-code-centric model for free software, to a community-centric model that shares ownership of both the software and the process used to build the software with everyone involved, from (yes) the coders and designers, to the people the software is being built for, to research subjects and the communities touched by it.

A world where software is enshrined to, and co-owned by, its community is not one where socially-oblivious developers making apologies for child rape could be venerated. It’s also one where design, documentation, and community care become equally important to the software itself, overcoming one of free software’s most longstanding issues. The benefits are real, and only extend the principles of openness inherent to the movement. After all, a software community that does not welcome anyone other than affluent, white men cannot be genuinely considered “free”.

This commitment to distributed equity requires both an extension of contributor covenants and free software licenses. Together, these legal frameworks need to provide co-operative-inspired processes for sharing ownership - a sort of “partner with community” to sit alongside the emerging exit to community idea. It goes beyond the mechanics of how source code is contributed and shared to define how the product and community are defined, designed, and governed. It’s a promise to distribute equity as widely as possible throughout the whole community.

You can’t separate the software from the community that built it. Therefore, true openness must dictate how that community is formed and run. We are what we choose to tolerate; in the same way that free software communities do not tolerate proprietary lock-in, they should not tolerate exclusionary social practices that lock people out.

RMS shouldn’t have been reinstated. It’s a disappointment, to put it mildly. But it’s also a line in the sand for the entire movement, and an opportunity - finally - to build something really new.

· Posts · Share this post