Yesterday, I got a short drive in a “Kick-Ass” Fisker Karma. As someone who converted a car to electric, it was a great treat to see the amazing quality, polish and sophistication of the Karma. Especially since, six years ago, I had to build my own EV. Today there is a diversity of choices ranging from the Nissan, GM, Tesla, Aptera, Fisker and others. Yet even with all these choices, EVs are far from the main stream.
How does that relate to Cloud *aaS? It’s all about innovation cycles and adoption.
Cloud platforms (really, IaaS software) have transformed from DIY to vibrant projects in the last few years; however, we still don’t know what the finished products will look like - we are only at the beginning of the innovation cycle.
With yesterday’s Citrix’s “CloudStack joins Apache” announcement painted as a shot against OpenStack, it is tempting to get pulled into a polarized view of the right or wrong way to implement cloud software (NetworkWorld, JavaWorld, CloudPundit). I think that feature by feature comparisons miss the real dynamics of the cloud market.
The question is not about features today, it is about forward velocity tomorrow. There are important areas needing technology development (network, storage, etc) in the cloud infrastructure space.
So the real story, expressed eloquently by Thierry Carrez, is about open collaboration and the resulting pace of innovation. That means that I consider all the cloud platforms in the market to be immature because we are still learning the scope of the “cloud” opportunity.
The critical question is how the various cloud projects will maintain growth and adopt innovation. Like the current generation of EVs, we must both prove value in production and demonstrate our ability to learn and improve.
From the perspective of building innovation and attracting developers, the tension between OpenStack and CloudStack is very real.