There’s a lot of wishful thinking in 'The Value of Open Source in the Cloud Era', a recent IBM-commissioned O’Reilly Media survey.
For example, there’s the finding that 70 per cent of the more than 3,400 respondents “prefer cloud providers based on open source.” This sounds great until you ask, “What does it mean to be “based on open source?"
After all, every single software product in existence arguably fits that description. And then there’s the finding that 79 per cent turn to open source in the cloud because it somehow prevents vendor lock-in. (This is, as I wrote back in 2016, a bit ridiculous, for a variety of reasons.)
But buried in all that open source feel-goodism, there was one glaring truth: Cloud-specific technology offerings will help a developer ship their code faster, but open source technologies enable them to build a career that gives them independence from any particular cloud provider. In other words, open source is the ultimate career hedge.
Open source magical realism
But let’s get back to mythology. First, roughly 55 per cent of respondents said that “Learning cloud computing skills specific to a single cloud provider limits my career growth,” despite the fact that... pretty much every single developer does exactly this.
Why? Because most companies tend to focus on a single cloud provider. Yes, of course pretty much every company ends up using a smattering of different applications or infrastructure from a variety of cloud companies. But this is what I call “accidental multi-cloud,” not “intentional multi-cloud.”
Intentional multi-cloud does happen, but it’s rare. Why? Because, as former Citrix vice president Christian Reilly noted, “The problem is, and always will be, the lack of fungibility. Fungibility doesn’t drive revenue. The great promise of agnostic providers died as soon as the commodity concepts went out of the window. Smart folks use best of... breed. The idea of real multi-cloud is lunacy.”
When companies hire, they have a cloud in mind. Knowing how to use the native services for Microsoft Azure or Google Cloud or AWS or Alibaba pays.
The report authors then go a bit off-piste to try to explain why developers turn to open source to reduce lock-in, suggesting that with proprietary software:
- The vendor may impose steep price hikes.
- The vendor may remove a key feature that a client depends on because the vendor doesn’t want to support it anymore.
- The vendor may go out of business or may radically change its business model and abandon its former clients.
- The vendor may enter the niche in which the client is working, becoming a direct competitor and abusing its position to put the client at a disadvantage.
- Bugs or bizarre performance problems may crop up in the features the client depends on.
- The client may have trouble finding job candidates with expertise in the proprietary product.
Unfortunately for the report, all of these factors have been equally as true of open source companies as proprietary companies. Customers who have built on an open source product, for example, don’t want to be told, “Don’t worry that the project no longer gets actively developed. You’ve got the code! You can support yourself.” This is not comforting in the enterprise.
Indeed, in a separate (blind) survey that my team at AWS sponsored, what customers seem to want is to be able to get the best of open source without having to take on the burden of thinking too deeply about it. Customers, whether tapping proprietary or open source software, want it to “just work.” But for developers, it’s different, and this is where the survey shines.
Freedom to be me... anywhere
Developers are hugely influential in enterprise purchasing decisions, but generally don’t control them. What they can control, however, is career growth, and for that, developers laud open source.
As important as it may be for developers to know the intricacies of a particular cloud vendor, many open source technologies (Kubernetes, Linux, PostgreSQL, etc.) give developers skills that transfer between the clouds. Small wonder, then, that developers see open source as critical to improving their career prospects.
It’s not that developers don’t derive value from knowing, say, Google BigQuery. Rather, there is more option value in knowing TensorFlow or some other open source technology that can be used in a wider variety of corporate settings. When nearly 79 per cent of the O’Reilly survey respondents say that open source software offers more “technology flexibility” than proprietary software, this is what they’re indicating.
So, while some indulge in that wishful fantasy that open source doesn’t also breed lock-in (in the enterprise, every technology choice breeds lock-in), where open source really helps isn’t at the corporate level but at the personal level. That is, the more open source I know, the more valuable I will be wherever I choose to work.
Not surprisingly, developers understand this. In the O’Reilly survey, when developers were asked to quantify the perceived importance of Kubernetes to their career, 52 per cent said it’s “extremely important” or “very important.” Add in “somewhat important,” and we’re up to 80 per cent of respondents.
Clearly, developers continue to be driven to build their careers around open source technologies, and maintain independence through open source, even as they strive to increase their value by investing in knowledge of cloud-specific technologies.