It seems like nothing has changed in open source for 10 years. Oh, sure, open source has emerged as a potent force for innovation, defining the cutting edge in everything from infrastructure (Kubernetes) to machine learning (TensorFlow). Our understanding of open source, however, seems decidedly retro.
Maybe, just maybe, this has less to do with our delayed learning and more to do with just how fast open source is moving.
You don’t need a weatherman…
Each day Google delivers news articles related to open source. And each day I’m dumbfounded by how little they change. For example, take this one from earlier this month: “The Pentagon is set to make a big push toward open source.” That’s amazing, since a 2003 Mitre report documented widespread adoption of open source within the US Department of Defense. By 2009 the DoD even had a website dedicated to open source code collaboration.
Lest you think this is just one headline writer’s mistake, pretty much every headline keeps churning over “news” that was outdated over a decade ago.
SEE: IT pro’s guide to working smarter with Linux (Tech Pro Research)
We’re still wading through the City of Munich’s ill-fated decision to try to dump Windows for Linux; still trying to make open source licensing less opaque (even as developers use it regardless of what their lawyers may think); still dithering over whether open source helps or hurts security; and still struggling to explain why financial services and other companies are embracing open source.
It’s 2017, people. Open source arrived long, long ago.
…to know which way the wind blows
Or maybe it hasn’t. If we’re still forced to swim through the same existential questions that plagued open source in the early days of its corporate coming-of-age, it’s possible that the market still doesn’t really “get” it. As much as the open source hipster crowd might consider itself “woke,” most people still scratch their heads over this bizarro world where you give away code that otherwise would be property. Property that could be sold. Property that, when not secured, is someone else’s fault.
SEE: 20 quick tips to make Linux networking easier (free PDF) (TechRepublic)
Or maybe our collective understanding of open source has improved, but simply hasn’t kept pace with the scale of open source. As GitHub’s Nadia Eghbal describes in her excellent “Rebuilding the Cathedral” talk, the allure of bazaar-style development has faded with the crush of keeping up with open source’s scale. GitHub now boasts approximately 70 million open source projects, growing double digits each year. To understand open source at some meta level may still leave us completely incapable of appreciating what’s happening at the micro level, project by project.
And so we’ll continue to see Open Source 101-type articles, as newbies (and everyone is a newbie to some corner of the open source universe, given its scale) come online and try to make sense of just how big open source has become. I’ve had some involvement in open source for 20 years now, and it seems like it should be mainstream by now. But it’s not, and won’t be anytime soon.
There’s simply too much happening, too fast.