Open source's new mission: To boldly go where no software has gone before
FOSS's license to exist depends on helping users. It has to learn to think that way
Opinion Bruce Perens is unhappy. He sees the spirit and potency of FOSS decaying into obsolescence as the big guns learn to game the system and users don't see the point.
What comes after open source? Bruce Perens is working on itREAD MORE
He should know. Perens co-authored the Open Source Definition document, laying out the principles and purpose by which open source licensing should work. This came out in early 1998, right at the moment the internet burst its banks and became a global flood. The astonishing histories of both have been inextricably intertwined ever since.
It's obviously true that the internet freed software from the physical media it had previously been distributed on. To reach a million users, you no longer had to copy and send out a million spinnable lumps of plastic and metal. You didn't have to be Microsoft to create and broadcast a new operating system, you merely needed to be a Finn with attitude.
Equally importantly, the internet was bidirectional. There was no longer an automatic barrier between the producer and the consumer of software, or between consumers who wanted to become producers themselves. The gatekeepers were, if not gone, no longer protected from competition through economic heft. With open source licensing, moreover, they no longer set the rules of the game.
In return, FOSS gave the internet the tools to grow at its own speed. The early adopters were never going to hang around for Microsoft or IBM to figure things out, they were going to build and share their own servers and infrastructure tools. By the time realization dawned in the boardroom, those rickety early packages had iterated into fully functional, universally deployed, reliable foundations of the new economy. Oops. Double oops: In so doing, they had validated and exemplified how to do FOSS of all kinds. FOSS and the internet reached critical mass, and kept on growing.
Fans of astrophysics will know what happens to massive objects that keep growing. Gravity overcomes everything and boom – black holes appear. In this case, the critical transition took place in the cloud, the quintessential evolution of FOSS and the internet. Imagine if Larry Page and Sergey Brin, the founders of Google, had gone to Microsoft in 1998 – it's that year again – to ask for a licensing package to build out Google on Windows. Nothing scales like free. And while a lot of big company FOSS usage feeds back to the community, a lot does not. If you don't distribute the code at all, you can lock up your changes too – and cloud services built on FOSS can be a secret as if every byte of code was written inside the event horizon.
Or you can be IBM and just wriggle out of things. As Perens bleakly notes, after all these years, the loopholes have been found and exploited, while the inherent problems with FOSS have not been addressed. Equitable distribution of funding, lack of user-focused design, and lack of user awareness of what FOSS can give them, even what it is, means it's time to find a new path.
This arc is inevitable. All revolutions are tamed. The youthful, angry rebels who drove rock and roll, punk, and hip-hop are now establishment-approved fossils in gilded display cabinets. Or dead. Or both. There'll never be another Apache, like there'll never be another Johnny Rotten. There doesn't need to be.
Traditional FOSS and the internet consumed the world because between the two of them they could hyper-evolve to colonize the new lands of fast comms and ever-cheaper digital devices. There are new horizons for the fleet of foot and itchy of genes, but you have to look at things differently.
- Navigating the truth maze in a world of clever machines and cleverer marketers
- How the tech toy century has troubled Santa's sack
- CLIs are simply wizard at character building. Let's not keep them to ourselves
- How to deorbit the Chromebook... and repurpose it for innovators
Perens says that users don't understand the freedoms FOSS gives, and developers don't understand users. He was right on both counts. The software that developers build is the software we talk about – it lives in repos, in source files, in distros. The software users see a multi-headed mutant shape-shifting dragon that swallows up and disgorges data – whatever that is – according to its own rules. Userland isn't built out of software at all, it's built out of data that flies from the screen to the eyes, from the mouth to the microphone. Users know that once inside the dragon, their data can be stolen and abused. They didn't know how to do anything about it, like they don't know why they have to use two calendars and three video conferencing systems. They don't know why the app they've been using for three years just changed beyond recognition.
This is the landscape that needs to be colonized by ideas that help the users. Open source by itself cannot be the cure: a Minotaur's labyrinth costs nothing to enter. What we need is string to find our way out again.
Stop thinking of FOSS software as clever code that lives in git boxes. See it instead as a paving stone or a wall or a destination in the flow of data that actually describes the user's digital world. Make it a condition of code use not just that modifications must be freely available, except when they're not, but that the flow of data must be discoverable to the originator of that data. Where's it coming from? Where's it going? Imagine a traceroute that goes up through the application layer into the data flow between components of what lives up there.
This would extend the inherent transparency of FOSS – something only visible to the highly trained – to the user. The details won't matter to them, but the ability for their device to warn at once if a service or an app or a platform is using a compromised component? With an open, real-time manifest captured available, that becomes possible. As does a "where's my data?" magic button that works independently of service suppliers. It might sound an incredible burden on providers and vendors, but this is 2024.
The megatonnage of useless data flowing through our infrastructure is incalculable. We know how to define efficient APIs and set rules that can be efficiently served at global scale. Open source has been intrinsic to all that. Now it can blossom into the user-focused, data-centric ecosystem we so badly need. ®