Posted by: Jack Vaughan
Enterprise mashups, SOA infrastructure, SOA registry/repository
SOA has had a bit of a rough summer. The Best and Brightest of the SOA bloggers have publicly ruminated long and lamentably on SOA’s future. There has been a bit of SOA fatigue in evidence. Could it be because many SOA projects are ready to roll out and some people want to be elsewhere when one or two implode?
SOA fatigue may be traced to aspects of SOA that people have sometimes rightly described as bloated. For SOA repositories and SOA governance the jury is still out.
People have yet to conclude how much is too much and what is the right mix of best design practice and ‘good-enough to ship’ development push for organizations that are – let’s face it – in the business to make apps that save or make money.
From days of old we know about project bloat – isn’t it fair to say we have to watch out for SOA bloat? Simple SOA should be as simple as possible, but not simpler. Yet, it should not become Super SOA – super-sized and expensive. Being too infrastructure-oriented and too far removed from killer apps with big ROI – that is sure to cause disenchantment.
The impression has grown only more stark in recent weeks as bloggers, sites, and magazines have begun to press the idea of Small SOA, or more recently, WOA, or Web-Oriented Architecture, to the detriment of Super SOA. That could be a good thing.
Sometimes people are too quick to assign ‘bloat’ and too quick as well to pretend a lighter version of middleware architecture is going to solve all problems. Yet, it is true that things can grow beyond common sense proportion.
What all this chatter belies is that Services and Architecture are the important parts of SOA. The goal is to build the right app, make sure it works, and ensure it will fit with other apps. Architects and developers in organizations are using that services approach. It is not a product, and it is not magic. It is a way to do work.
We are just reaching the moment when many of SOA initiatives are going live full time. Did everyone ‘get it right?’ Probably not. Is that a reason to run from SOA? No.
SOA was no magic bullet. Neither is WOA. Is this an interesting time? Yep.