One of my favorite scenes in the movie Contact is when young Ellie Arroway’s father counsels her on the value of making incremental improvements in her short wave radio efforts: “Small moves, Ellie - small moves,” he tells her.
I was reminded of this when I read and commented on InfoWorld’s recent post, SOA gets an obituary. In it author Paul Krill reviews a recent Burton Group executive’s statement that declared Service Oriented Architecture “dead.” Krill quotes Burton Group’s Anne Thomas Mane in her blog as saying,
Companies need to become more in tune with what businesses require and understand what the problems are, she said. What is required is an examination of application architecture rather than project-by-project integration, Manes noted, but with the difficult economy, funding for SOA has dried up, she said.
One “sin” committed in the name of SOA seems to be the attempt to create an all encompassing technical architecture that accomplishes a real linkage between needs and technology. But that’s the problem — “all encompassing” shoves too many other things aside.
Here is my own comment:
Having made good money in the past as a consultant on systems integration initiatives, I think that one perverse response to the “SOA is dead” meme is that it proves, once again, how critical IT is to modern organizations. It’s impossible to totally re-architect the IT infrastructure of a modern organization because there are just too many moving parts — and the parts refuse to stand still long enough to reach an ideal integrated state. Perhaps the moral is: for the time being we need to take smaller steps.
Admittedly, “taking smaller steps” doesn’t sound very exciting. But I was reminded of an interview I conducted last year with the IT director in a large organization who was knee-deep in laudable standardization and re-architecting efforts. All his energy was focused on critical strategic goals aimed at improving responsiveness, standards, business-technology alignment, and cost control. In the process, though, he admitted that he had to set aside short term user requests, including those in the area of social media and social networking that could have been satisfied without major modifications to the organization’s architectural goals.
Granted, focusing on only “big picture” items at the expense of “small picture” items is not the same as losing control over large complex projects with too many moving parts. But the end result of the two are the same — user needs get lost in the process.
The question now is, as the economy continues to worsen, whether even “small moves” will be possible?
Copyright (c) 2009 by Dennis D. McDonald, Ph.D. McDonald is an Alexandria Virginia based management consultant. He can be reached via the contact information provided here.