Eucalyptus refocuses on open source values

Upcoming release to merge two branches to single GPL product

Even as its competitors generate the headlines and talk more than a little smack, Eucalyptus Systems has been quietly and carefully re-aligning its business practices, while steadily growing. Now the cloud computing company is ready to take on the sector with less quiet and more open source attitude: including a radical shift in how it will deploy its main product.

The conversation with Eucalyptus this week started with a question I raised Monday based on a statement made to the New York Times by Eucalyptus CEO Mårten Mickos, in which Mickos made the assertion that some in the open source community still felt that making profits was immoral.

[ FREE DOWNLOAD: Selecting the right cloud: A step-by-step guide | 10 things to check before moving to the cloud ]

So when I saw a note in my inbox from Mickos about an hour after that post went up, I knew I was in for it. Mickos can be pretty diligent about correcting what he believes are errors in reporting, something I have learned to expect.

Actually, it wasn't all that bad. There were some clarifications Mickos wanted to make on his comments, to be sure, because Mickos has a lot more to say on the subject of open source and business than the original NYT piece let on. But mostly the conversation centered on Eucalyptus and what the Santa Barbara firm has been doing since Mickos joined the company in 2010.

The short version of the story? A lot. The Infrastructure as a Service company has made a great deal of technical improvements to its enterprise and open source editions, and has--while others have been tooling up in this space--been nicely growing its customer base.

But it hasn't been without some bumps along the way. Bumps that have shown up in the press and have been exaggerated by competitors and some in the media (including, regrettably, me). Mickos recognized those problems when he came on board with Eucalyptus, even as he was fully engaged with the vision of the six co-founders of the company.

That core team, Mickos knew, love open source and are passionately dedicated to its principles. And when they developed the Eucalyptus product, they applied those principles with the same laser-like focus they applied their engineering and programming skills.

It would be that highly intense focus, however, that would create the rough spots Eucalyptus would endure.

"They didn't have the outward focus on the community and the customers that they needed to have," Mickos explained. "They were focused on each other and on building the great product."

This inward focus would lead to the misunderstanding about Eucalyptus and NASA. The apocryphal version of the story is that the Eucalyptus team refused to incorporate changes from the NASA team because those contributions would be too close to Eucalyptus' proprietary add ons to its open core enterprise product.

In reality, Mickos explained, the four or five members of the Eucalyptus team who were dedicated to the NASA project did not communicate as effectively as they could have (due to the issue of inward focus on Eucalyptus development). There were also competing motivations from within NASA. Then-NASA CTO Chris Kemp, who argued the most vehemently about Eucalyptus at the time, would divert the NASA Nebula team to create a new fabric controller, Nova. Nova would later be integrated into the new OpenStack project, and Kemp would later go on to launch his own company (also called Nebula) in 2011.

Reports that NASA completely dropped Eucalyptus were not entirely accurate. NASA still continues to be a Eucalyptus customer, Mickos clarified, adding that it was the Nebula team alone that dropped Eucalyptus software.

The misunderstanding about this and other incidents would lead to a lot of friction between Eucalyptus and its community, as well as a lot of pundits decrying Eucalyptus' use of the open core business model and pointing to the NASA kerfuffle as a prime example.

This kerfuffle appears to have been overblown, and for the part I played in the pile-on, I regret the error.

That doesn't mean Eucalyptus was without its problems, however, as Mickos was very forthcoming to admit. With all the conversation focused on Eucalyptus an its business model and licensing, the company became distracted from what it really wanted to do: make good IaaS software.

Now the company is about to address those problems head on.

"When [the founders] chose the business model, it became overly complicated," Mickos explained. Beginning with the upcoming release of Eucalyptus 3.1, therefore, the company will be dropping the two-branch development tree that stated with Eucalyptus 2.0 (one open source community version and one commercial enterprise version) and going to a single open source branch, licensed under the GPL.

Greg DeKoenigsberg, Eucalyptus VP, Community, recently delivered a thoughtful explanation on the decision in his blog.

"Looking back, it's probably not accurate to say that the decision was a 'mistake,' per se, but it's definitely true that the open source version suffered. In retrospect, it was inevitable; whenever a hard choice needed to be made about the allocation of scarce resources, the choice was always, always, always to solve the customer's problem. So the subscription version was patched and tested, while the open source version atrophied, with the lack of commit activity leading many observers to conclude that Eucalyptus Was Dying. Which was completely wrong, of course; the customer base grew the whole time, and the company grew with it. The new version of the product marched along, with a shared understanding among Eucalyptoids that when the time came, the codebase would be rebuilt the right way. The proprietary hooks would be pulled out into modules. Open source would never again be treated as a second-class citizen."

Mickos added that this was a decision that was one-and-a-half years in the making, and not a knee-jerk reaction to recent bad press about the open core model. Given the amount of effort DeKoenigsberg described it took to merge the two code branches, that seems just about right.

The company is also changing the way its dealing with the community, hiring new staff and training veteran team members to focus outward.

"We now have a lot of people who are there to take care of the world around us," Mickos said.

Does this change repudiate Eucalyptus' open core model? Not hardly. While the initial release of Eucalyptus 3.1 will be completely open, Mickos did allude to the possibility of add-ons down the road.

That's because Mickos is a firm believer in using open core to enable scalable commercial growth. Relying on just service and support as the "special sauce" limits how an open source company can grow, he maintains. Adding strong software features that customers are willing to pay for is key to sustained growth, like the kind Red Hat enjoys.

Mickos makes a strong case for the way his company is doing business, and if industry observers are correct, then it seems to be working out quite well. With this new focus on its core open source values, Eucalyptus is about to be a much stronger and louder presence in the cloud computing sector.

Read more of Brian Proffitt's Zettatag and Open for Discussion blogs and follow the latest IT news at ITworld. Drop Brian a line or follow Brian on Twitter at @TheTechScribe. For the latest IT news, analysis and how-tos, follow ITworld on Twitter and Facebook.

ITWorld DealPost: The best in tech deals and discounts.
Shop Tech Products at Amazon