The lingua franca for developers is code, and predominantly open source code. Even more than code, however, is convenience, which is why Amazon Web Services (AWS) rules the cloud computing roost. And yet, there are signs that Google's mixture of complex machine learning and AI code with open source software is a winning strategy—one that could put AWS under serious threat.
Getting real about open source
Google has made no secret that it intends to differentiate its cloud by baking in superior machine learning and AI capabilities. Having deep data science, however, is not the real differentiator. Rather, as David Mytton has highlighted, it's all about making that machine learning and AI approachable to a broad swath of users: Google plans to "standardize machine learning on a single framework and API," then augment it "with a service that can [manage] it all for you more efficiently and with less operational overhead."
The key to that standardization? Open source.
Redmonk analyst James Governor nailed the reason a year ago:
[B]y 2015 Google realised that open sourcing the code itself, rather than just publishing papers about its approaches, made sense. Why watch somebody else create another Hadoop or Mesos when Google could build a community around stuff it actually built—and so Kubernetes was born....The decision to open source some of Google's core machine learning technology—TensorFlow—followed naturally on the obvious and growing success of a better, more collaborative model for applied science.
Fast forward to today, and TensorFlow is "effectively the de facto standard open source library toolkit for machine learning," Governor wrote. Kubernetes is also a community darling, with open source heavyweights like Red Hat lining up to sing its praises. Google may still dominate development, but that's rapidly changing, and Google has learned to open up and get out of the way.
Training a machine learning army
Machine learning and AI are extraordinarily difficult. Though there is plenty of hype about enterprises embracing the two, the reality is that most of this so-called AI adoption is just marketing fluff. To turn enterprises into real users of machine learning and AI, someone needs to make it both easier to consume and make the users savvier.
By generously giving away the keys to its machine learning and AI kingdom, Google Cloud effectively ensures that the next wave of data-informed applications will be pushed to the platform best equipped to run them. In so doing, Google also ensures that it, and not AWS, trains the next generation of data scientists to grow up on its code. In both ways, open source helps Google to place itself at the center of machine learning and AI for years to come, with its cloud business reaping the revenues as a result.
AWS, meanwhile, "has definitely not taken the open source pill," as Governor quipped. As AWS increasingly competes with the open source projects it hosts, Google's cloud and Microsoft Azure, which is also very open source-friendly, become obvious homes.
To be clear, AWS has a strong lead and significant AI and machine learning work of its own. But AWS still errs on the side of hiding code as it offers services. By opening up its machine learning and AI efforts, Google may be offering a one-two punch that AWS seems ill-disposed to counter.
- The cloud war moves to machine learning: Does Google have an edge? (TechRepublic)
- Google's problem with the enterprise cloud is that it's too innovative and not practical enough (TechRepublic)
- Cloud computing policy template (Tech Pro Research)
- Google finally gives developers access to its cloudy secret sauce (TechRepublic)
- Why Kubernetes may be a bigger threat to Amazon than Google's cloud (TechRepublic)
Matt is currently head of the developer ecosystem at Adobe. The views expressed are his own, not those of his employer.
Matt Asay is a veteran technology columnist who has written for CNET, ReadWrite, and other tech media. Asay has also held a variety of executive roles with leading mobile and big data software companies.