Intel claim open source driven by 'enthusiasts' is 'complete rubbish' says Weaveworks founder

"Enterprise-capable product" can only be produced once open source is "professionalised" by distributors, claims Intel

Weaveworks founder and CEO Alexis Richardson delivered a verbal drubbing to an Intel senior architect yesterday after he suggested open source software is still driven by "enthusiasts" who alone don't produce "enterprise-capable product" without distributors 'professionalising' parts of it themselves.

Richardson, speaking at an open source panel debate hosted by Rackspace, described Markus Leberecht's claim as "complete rubbish", leaving the solutions architect floundering.

When discussing the increasing relevance of open source software to the enterprise, senior data centre solutions architect Leberecht volunteered the notion that "open source has become a natural thing for enterprise to consume when distributors have professionalised certain parts of [it]".

"So just to re-emphasise the role that some of the companies on the panel here [companies included MongoDB, Red Hat, and Rackspace, as well as Weaveworks] are taking in this particular way of getting open source to market: by itself open source is attention-driven, enthusiasts driving a certain topic, but that doesn't give us enterprise-capable product."

"I'm sorry, but that's just complete rubbish," Weaveworks' Richardson shot back.

"Most of the successful open source products are driven by companies, not by 'enthusiasts' or whatever you call them."

Rackspace principal engineer Igor Ljubuncic then waded in to try and make common ground, suggesting that "if you really want to teach others how to do something, you have to master the domain enough to be confident enough to do it.

"And open source makes you really vulnerable, because you expose yourself to the audience and the developer community, who can actually take your product and surpass you, so you have to balance between your desire to sell products, and the knowledge that you can potentially [lose ground].

"But if you excel, you become a market leader."

Leberecht suggested the panel had "got a bit hung up on when I used the word 'enthusiasts'."

"As well as "attention-driven" - it's just not," replied Richardson.

"It is in so far that in the development process, those things that [depend] on core functionality, on fail first, are usually the ones that are done first," offered Leberecht.

"I don't know what you're talking about, I'm sorry," concluded Richardson.

"There are relatively few purist open source companies out there," said Martin Percival, senior solutions architect at Red Hat as he picked up the increasingly icy ball.

"There are a lot of companies that will do an open source core or an additional pathway to make an actual revenue. Red Hat is relatively rare in that it survives from taking software and literally putting it out into the world come-what-may, and we'll continue to do that."

Percival observed that even when Red Hat has bought companies outright, it "turned that into open source as part of our model".

"I think when we talk about the history of open source, you can build a market with something like that, and there comes a tipping point when you take it, rather than not take it. There reaches a point where everybody is talking about it or doing it, whereas before there was always a question as to whether you should be doing it."

Reflecting on the history of the open source movement, Richardson opined that application server JBoss - founded in 2004 but now owned by Red Hat - was "the first time open source was taken seriously" by the business community, coinciding "exactly with" what Percival claimed.

"[It was] the really early days of what they call professional open source, and the idea was that you could get consultants for professional services around open source projects, and the reason for this was you wanted experts working full time on the project to get paid for 40 hours at a premium, and they would do that buy selling those hours to business customers.

"Red Hat was successful enough at the time for a handful of companies with about 50 staff each - and that was the dream."

It was a 'professionalisation', said Richardson, that began many, many years ago.

An interesting debate for sure, unfortunately marking Intel out on the day as a company that may need to do a little more of its homework. Attempting to brand open source with the stereotypical, beardy "enthusiast" sticker in 2016 is neither smart nor true, especially when sitting on a panel filled with industry peers who have built their entire business on exactly the opposite spirit.