Could the 'death' of SOA bring it back to life?

The jig is up.

Wow. Anne Thomas Manes, one of the Burton Group's SOA analysts, really stirred up the SOA pundits with her recent post about the death of SOA. You'll find my coverage here.

As you recall, I had this in my SOA predictions for '09 (I posted in October). In essence, I said that we would move away from the SOA TLA, largely due to the complexity of SOA, and businesses' inability to solve such a large problem. Moreover, the patterns of SOA would live on under other names, such as enterprise architecture and/or cloud computing. Anne just declared the time of death, something I think many of us were waiting for somebody else to do.

What is surprising to me is the number of pundits and thought leaders who are basically nodding their heads. Most interesting was a post from Mike Kavis, who did a better job as summarizing the issues than I did on Monday:

In my opinion, we suck at architecture! Why do we suck at architecture? Let me count the ways.

Of course, Miko Matsumura had to get his own mark on this by changing the name of the "SOA Center" to "Whatever-you-want-to-call-it Center ... The Artist Formerly Known as SOA Center." That's so Miko, and he makes a good point.

Moreover, I suspect that those in SOA marketing are freaking out a bit this week, including IBM, which has invested a great deal in the notion of SOA and ESBs. Indeed, I'm noting a clear pattern that the vendors are pushing back on the conclusions from the Burton Group, whereas the thought leaders and pundits are saying, "Well, somebody had to say it."

So, what's the end game here? The reality is that the patterns of SOA predate "SOA" TLA and won't go away when the TLA does. However, do think that SOA, or whatever you want to call it will be practiced differently, and that may be a good thing. In short, I think that architecture will be the small mammal that survives the meteor strike, and live and thrive for years to come in creating good enterprise architectures, enabling infrastructure for cloud computing, and many other uses. What won't remain is the hype and confusion.

Also, it's time for those who made a bunch of money off the notion of SOA, without providing a lot of value, to move on to the next hype wave. The jig is up.