The future of open source will be messy

Open Source Initiative (OSI) chief Stefano Maffulli says Meta is “bullying” the industry on the concept of open source. The Economist more colorfully characterizes the dispute as “bare-it-all purists … confronting Meta, the social-media giant controlled by a mankini-clad Mark Zuckerberg.” It’s all much ado about something as AI proves exceptionally difficult to stuff into a decades-old definition of the term open source, which had already been failing to keep up with cloud.

How important is Meta’s steadfast refusal to show obeisance to the OSI’s hitherto iron grip on open source naming rights? It’s both more important and less important than you might think.

Obsessively compulsive about open source

First, it’s important to point out that open source software is both pervasive and foundational. Where would we be without Linux and the vast treasure trove of other open source projects on which the internet is built? However, the vast majority of software, written for use or sale, is not open source. This has always been true. Developers do care about open source, and for good reason, but it is not their top concern. As Redis CEO Rowan Trollope told me in a recent interview, “If you’re the average developer, what you really care about is capability: Does this [software] offer something unique and differentiated that’s awesome that I need in my application.”

Unsurprisingly then, a coterie of open-source-is-what-the-OSI-says-it-is advocates (note: I’ve historically been in this camp) are castigating Meta for calling its Llama large language model (LLM) open source, despite restrictions that fall short of the OSI’s Open Source Definition. The industry’s response has been a collective shrug. See, for example, “Why Meta LLaMA Models Are Open Source” — a title that must drive OSI folks crazy. Part of this stems from, as one HackerNews commentator says, the idea that “Meta, through the Llama models, has done more for open source LLMs than just about anyone else.” Rewinding even further, open sourcerors can look to Apache Cassandra, React, GraphQL, PyTorch, and other Meta projects that met the OSI’s bar for open source.

It’s hard to get too grumpy with a company that has created some of the industry’s most important open source projects.

And yet some people are very grumpy, despite the fact that there was (and is) no settled definition for open source in AI. Yes, the OSI finally released a definition of open source for AI, the Open Source AI Definition 1.0, but, as with cloud, the OSI is playing catch-up, and its definition has disappointed some of its most ardent supporters by not dictating that training data also be open.

Meanwhile, Meta and the rest of the industry keep releasing new code, calling it open source or open weights (Sam Johnston offers a great analysis), without much concern for what the OSI or anyone else thinks. Johnston may be exaggerating when he says, “The more [the word] open appears in an artificial intelligence product’s branding, the less open it actually tends to be,” but it’s clear that the term open gets used a lot, starting with category leader OpenAI, which is not open in any discernible sense, without much concern for any traditional definitions.

Does any of this matter?

Open is as open does

I’ve been pretty blunt on the topic: Developers want software that is easily accessible and that works. Sometimes that means they want open source, but in the age of cloud computing, it has just as often meant that they wanted a solid API and a free or low-cost cloud service. AWS took the dominant early lead in cloud not because of its open source bona fides but because it offered the most compelling cloud services, as I’ve argued. Developers will continue to use Llama so long as it offers a fast track to generative AI applications, no matter its license.

Still, the fact that Meta says Llama is open source and that OpenAI wanted open in its name (despite the fact that its software runs obscured from view on servers in the cloud), suggests that openness does matter. Call it marketing, if you wish, but these companies recognize that they need to at least nod to open source.

I’d argue they’re giving more than lip service to open source—at least Meta is. Yes, their license blocks use if you have more than 700 million monthly active users, but for 99.99% of the population of Llama developers, corporate or hobbyist, that is a meaningless footnote. Is it open source? No, not by the OSI’s definition, but that’s the point: Meta isn’t willing to cede policing of the term open source to the OSI in an area where Meta is the expert.

And developers don’t seem to care.

So, yes, openness matters, but not so much that the industry is willing to let one nonprofit try to retrofit a concept originally designed for packaged software. The OSI failed to keep pace with cloud, which allowed the big cloud vendors to disproportionately take from open source without contributing back. If Meta isn’t willing to let that happen in AI, an area where it leads, it’s hard to blame them.