This article is more than 1 year old
Open source's totally non-secret weapon big tech dares not use: Staying relevant
Detachment from users' lived reality is how corporates shoot themselves in the foot
Opinion Last week, one fundamental problem for IT cropped up in three very different stories. One story was Google's parent Alphabet doing an internal audit of all its products on the back of falling profits. One was a highly critical look at Meta's efforts to put business into VR. And one was Linus Torvalds getting cranky that the i486 architecture was still in Linux's first-class lounge when it should be packed off to the Old Codes' Home.
Smart people want to do good things, things that are relevant to others
Meta and others are pitchforking hundreds of billions into a virtual bonfire of insanity, egged on by big-name analysts making risible predictions about trillion-dollar markets. That looks like a different universe to decisions about free software for a 30-year-old processor with zero modern relevance. Likewise, Google's habit of making things that nobody wants and breaking things that people do want, conversely, is a very Google thing.
All three are different ways of describing the same universal problems that all organizations face: resource allocation and opportunity cost. Meta is going hell for leather for one big idea that is delusional nonsense. Google's strategy of letting a thousand flowers bloom is a recipe for a backyard filled with weeds. Both these things are illnesses of the rich; you have the luxury of doing things badly for a long time if the cash keeps flowing.
Linux, on the other hand, is anything but rich. Its ecosystem of reward is enlightened self-interest, people and organizations that commit time and minds. They understand the importance of making things that matter and are very motivated to be a part of that. These resources are limited. Time spent developing, maintaining and testing for a processor that nobody uses is time not spent on current needs. Relevance has to be part of the reward; declaring something irrelevant is part of resource management. This balancing act has to work.
To a great extent, it does. FOSS has its flagship projects - the Linuxes and Apaches and LibreOffices - which means, perhaps miraculously, the desktop environment can be as productive and capable as proprietary alternatives.
There may seem to be no FOSS equivalent of Azure or AWS or Google Cloud Platform, but the intellectual infrastructure that open source provides through the ecosystem of libraries and frameworks is arguably as important to modern IT as the cloud itself. Relevance, it turns out, is a really good way to make resources self-manage, to be maximally productive in ways that match need.
It's amazing how making money the motivator can break that down. One cost of being rich is that you can ignore the outside world, or see it as something to be bought. That shields against relevance to users. Meta wants everyone to work in the metaverse, and is denying the awkward fact that strapping shoeboxes to our heads is completely irrelevant to our day jobs. But guess how long a career at Meta will be for anyone who says so. Relevance is poison to big wrong projects, and the corporate immune system recognizes it as such.
Another cost of plump revenues is that they become the only relevant things in your world. Google's incoherent product strategy is the opportunity cost of management mindset soaking away into the sands of advertising revenue and the resulting wars with regulators and governments. Meta has the same besetting sin, it just minimizes proper strategic planning by declaring it a "vision" instead.
- Google's Alphabet to review every project after $6bn decline in profits
- Most Metaverse business projects will be dead by 2025
- Linus Torvalds suggests the 80486 architecture belongs in a museum, not the Linux kernel
- Firefox points the way to eradicating one of the rudest words online: PDF
The irony of this is that both Google and Meta are successful parts of the FOSS ecosystem - and not just with headliners like Android, Chromium, Kubernetes, React, and PyTorch. Google in particular is proud of the tens of thousands of its engineers engaged with open source projects large and small. These are talented, engaged, valuable people, capable of doing great work. So how come the companies' branded goods don't reflect that?
Smart people want to do good things, things that are relevant to others. There's a word for behaving in ways that affect others in positive ways - ethics. Open source is profoundly ethical, in that it has rules that make deception extremely difficult. Proprietary models practically invite a divergence between public image and private motive. Guess which works better for motivating workers.
Torvalds can enforce policy decisions among an independent workforce by becoming mildly grumpy. Zuckerberg can bring billions of dollars and the full force of corporate discipline to bear on a Mighty Vision, and it ain't happening. What's happening is big tech becoming seen as the enemy of good governance, the perverter of politics through amoral algorithms monetising misery. Yikes.
The long-term innovation that big tech needs is cultural, not technical. The one big idea can be a great way to bring a revolution. An internal market of ideas can spark real fire. But to get there, means engineering a way to make the good of the end user more relevant than maximising revenue. That may seem utopian, but it was once the stuff of corporate will.
"We believe strongly that in the long term, we will be better served—as shareholders and in all other ways — by a company that does good things for the world even if we forgo some short term gains." That's quite a sentence. The company? Google, in its 2004 IPO prospectus. The three words that preceded it? "Don't be evil."
There's innovation for you. ®