In my days as director of a leading management association and editor of its journal, I was constantly amazed at the fad mentality that would sweep through otherwise staid, serious organizations. For a couple of years, everyone was abuzz over something called "quality circles" and "Theory Z" forms of Japanese-style management. Then there were various incarnations of "participative management," and everyone went "searching for excellence" before they went gaga over "process re-engineering."
Is SOA just the latest bandwagon nobody wants to miss -- or something deeper?
Is service oriented architecture that kind of fad -- which dominates the headlines and conference sessions for a few years, but doesn't seem to add much to the business when all is said and done? Is SOA a fad that is driven more by "bandwagon" thinking than actual business value? I'm sure many readers out there will agree that SOA certainly has been hyped beyond reason. But, still, there's substance to the methodology.
John Evdemon references a new Harvard Business Review article that discusses the "fad" mentality that sometimes gets the better of management, and ponders whether this is what's driving SOA to a large extent.
In the article, titled "When You Shouldn't Go Global," authors Marcus Alexander and Harry Korine question whether popular trends (or "bandwagons") may lead to sloppy thinking among executives. They cite the example of globalization, today's fad du jour, promoted by analysts, consultants, and media as the only way to succeed. Alexander and Korine say companies attempting to extend their reach across borders into new markets and ventures often find they can't integrate these far-flung parts of the enterprise, and things end up falling apart. Great vision, lousy execution.
The HBR piece included a sidebar (unfortunately, not accessible to non-subscribers) titled "The Susceptibility to Managerial Fads," and Evdemon plugs in SOA as an example of such fad thinking (and provides some great historical references as to how SOA thinking evolved):
"According to the sidebar, one example of 'sloppy thinking' is that labels used to describe a trend get stretched far beyond their original meaning. SOA was initially described by Gartner back in 1996.... In that original article, SOA was described as 'a style of multitier computing that helps organizations share logic and data among multiple applications and usage modes.' Seems fairly straightforward, right? If we jump back to 2008 SOA has been stretched, pulled and prodded to mean many things, of which multitier computing is only a small part. People have defined SOA in so many different ways that its become almost meaningless. Industry pundits have only added to this confusion by assigning version numbers (SOA 1.0, SOA 2.0). defining segments (WOA, POA, etc) and creating maturity models (far too many to list here)."
The original intent of SOA -- such as what Gartner said back in 1996 -- got lost in the noise and hype as everyone tries to get on the bandwagon, Evdemon points out. "This leads to people focusing on things that are often irrelevant or unrelated to the benefits the insight promised to deliver.... Are we implementing solutions or buzzwords? Far too often it seems like the latter."
Is it that, or have people taken the concept and really run with it, expanding the possibilities of what SOA -- or variations thereof -- can deliver? The original Gartner vision couldn't imagine the transformative effects reaching beyond IT that people are exploring today. I agree with John Evdemon that all the definitions an buzzwords are crazy. But SOA is also a work in progress, and is evolving as the world around it evolves. And yes, it also still makes sense to search for excellence.