Defining open source by its spirit


Johannesburg, 07 Aug 2020
Muggie van Staden, CEO, Obsidian
Muggie van Staden, CEO, Obsidian

How do you know you’re getting real value from your open source software? At first, it may seem like an obvious question: open source is free, right? And free is terrific value. But, while this narrative resonated with an earlier time, when open source was still fighting for traction against established proprietary brands, that dynamic is very different today. For one, open source is now nearly ubiquitous. It's rare to find a major IT system or service that doesn't incorporate elements of open source.

Yet as the prominence of open source has changed, its definition became more ambiguous, says Muggie van Staden, CEO of Obsidian: "The open source landscape has changed a lot. Fifteen years ago, people laughed at open source vendors, thinking that they were crazy. Now, everybody wants to do it and almost all vendors claim to be open source vendors in some way. Open source is literally everywhere. And the definitions get blurred because it is such a pervasive way of building software."

Many adopters look at cost as the main value behind open source, and parts of the market have been happy to sell it as such. This, though, threatens to undermine the actual value of open source. It's not about cost, but about collaborative development.

Back to the open source well

What is open source? It's the ability for a third party, without a licence or prior arrangement, to see an application's code and make changes to it. The concept terrified companies charging for access to their proprietary systems. Certainly at a base economic level, open source didn't seem to make much sense. Why give away what you can charge for?

Yet that assumption overlooks a critical factor: open source is intuitively collaborative – every piece of software benefits from continual testing, quality assurance and bug squashing. Proprietary software can only pursue that as far as the resources and processes of its owners can allow, while open source uses large independent communities who take it on themselves to establish quality.

The most obvious example is Linux. This open source operating system not only exists in many complementary flavours to suit different needs, but is also famously secure because of its community's prolific debugging and development spirit.

Open source's value is in how many contribute to the greater good. Built on top of these communities are enterprise-level open source solutions – services that leverage free open source, but include paid-for extras such as support and customisation.

The problem today is that many using open source think the value comes from it being free, and they consequently fail to feedback into the cycle that keeps improving open source options: "Open source means a lot of different things to different people. But the power of open source sits in the collaborative development model. It's not that something is cheaper or free. This is important, because we see companies take on bad open source products. They think they are scoring value because it's labelled as open source. There are a lot of products out there that claim to be open source, but they do not support the collaborative method. They only take and give nothing back. That is a dead end."

The right stuff

Open source is fuelling the modern era of connected technologies. Pieces of open source code are being used inside most applications and architectures, because they are often the most eloquent and tested solutions to a problem or requirement. This ubiquity has led to the blurring of open source's definition, and it can be hard for customers to know if they are engaged with a provider that does justice to open source's collaborative power.

Yet, suffice to say, if you use open source components from a company that doesn't pay it forward, the solutions you get from them will soon lose its open source value. So how can one distinguish between 'good' and 'bad' open source? You could ask if they contribute some of their work back into the open source community, though this could invite complex and befuddling answers.

Instead, Van Staden suggests, look for something that also often defines good proprietary software: "You can get good open source and bad open source, just like you can get good proprietary and bad proprietary. For me, one of the key things is to focus on open standards. You can have a piece of open source technology that doesn't adhere to any open standard, and that in my view is a bad decision. In contrast, you sometimes can get a piece of proprietary software that complies with all open standards and therefore reduces the risk of deploying something like that in an enterprise. Good open source software should offer the same."

He adds that supporting enterprise versions of open source is also a powerful way to both feed the collaborative spirit and ensure you receive the best software. Non-enterprise options are still potent, but enterprise-grade open source offers support and also pays back into the community.

"Why must I pay for a piece of open source software from an enterprise vendor? They plough enormous amounts of resources back into open source development. If you look at vendors like Suse and Red Hat, a lot of their money is paying developers who collaboratively build Linux."

This concept goes by different names, such as frenemies or pre-competitive collaboration. But instead of a complex joint partnership or other vehicles, it draws from the DNA of open source: take but also give. The presence of open standards, which also reduces the risk of lock-in, is a good sign that these principles are in play: "It should never be about just the cost. It's a value discussion. This makes more sense in the world today where companies are less and less concerned about owning the technology. They're more interested in owning the data, owning the algorithm, owning the components that go into the technology. That's part and parcel with open standards and good open source."

Share