Everyone who reads the IT trade press knows that SOA has moved from the hot trend stage to mainstream adoption. And how do they know? By reading the IT trade press.
Granted, there's a certain circular logic here. But the people who need this knowledge -- like CIOs or developers trying to build the right skill set -- simply don't have time to call "120 IT and business professionals" and interview them. They have to rely on what's available from the media. The question is, how reliable is that information?
An IDC survey conducted in mid-2006 with the above-mentioned 120 IT and business professionals pegged SOA adoption at 90 percent. This squares with the Yankee Group's figure of 84 percent, published in the same time frame.
But sometimes, the figures are much more ambiguous -- so much so that journalists can look at the same report and come up with radically different take-aways. There was a shocking example of this phenomenon last week concerning SOA adoption in the federal government.
The raw news was a survey conducted by Merlin International Inc., leader of a vendor group called the Merlin Federal SOA Coalition. According to Government Computing News (GCN) in a story titled "Feds Getting Comfortable with SOA," the IT professionals associated with federal agencies "are becoming well-acquainted with the concept of service-oriented architecture." The story goes on to report that SOA is now seen as a "safe decision" for agency program managers, and cites a variety of statistics to back up its positive spin.
A story in Datamation, "SOA and the Government: A Slow Process," tells an entirely different story about a federal government survey in which most participants said SOA projects were only partially successful or not successful at all, with 14 percent characterized as "a fiasco." The same survey found that half the respondents had never even heard of SOA, and of those that had, roughly one-third got the definition wrong. The shocker: This article was talking about the same Merlin survey!
Knowing these two publications, we feel quite confident that both stories were accurate, and that the conclusions involved simply reflect a cup-half-full vs. cup-half-empty situation. But there's a lesson here: If you really, really need to know about a particular report or survey, find a way to get your own copy.