I’m getting some pushback on my recent prediction that OpenStack will implode as a result of its inability to innovate sufficiently to keep up with the marketplace. I admit I’m sticking my neck out a bit on this one, but the basic idea is that OpenStack has been driving toward maturing the various individual projects for a while now, based upon architectural decisions made early on. What happens if some of the smarter people in the consortium realize that maybe there’s a better way to set up the whole thing? At that point OpenStack might collapse under its own weight, as the best developers bail on an obsolete monster to build the next great thing.
To illustrate this point, take a look at the recent “Grizzly” release’s logical architecture in the following diagram.
You’ll notice many moving parts and interrelationships connecting seven main rectangles which represent the core OpenStack Projects. What if sometime down the road (but perhaps in 2014), some members of the OpenStack team look at this diagram and say to themselves, we can do a lot better than this. Let’s start with a blank sheet of paper and the hard lessons we learned by trying to build this OpenStack mess, and come up with a newer, better approach.
In other words, innovation, in all its messy glory. Too many people have been tweaking the diagram above for too long for anybody to be sure they’re still on the right track. Simply tweaking it as they go is likely to miss fundamental innovations that will lead to quantum leaps in improvements.
There’s no such thing as failure when innovating – there are only successes and “learning experiences.” My guess is that OpenStack will end up being the latter.