So say a significant chunk of the 450 respondents comprised of CIOs, CFOs, and HR professionals surveyed by "modernisation solution" provider MicroFocus in a report released earlier this month. Only one third felt that their recruiting efforts were producing enough specialists in core IT competencies (identified as COBOL, PL1, and CICs).

My question is, is this really any different than it has ever been, and if so, why are people making such a fuss over it? Didn't we hear this same screed just ahead of Y2K, when our disturbing lack of COBOL experts was said to have our global information systems poised on the brink of doom? Now, it is Joe McKendrick, citing the survey in the wake of previous articles by he and Dave Linthicum bemoaning the slow death of SOA they are seeing as the recession sinks in.

I have some questions about the survey itself, which seems to draw a lot of conclusions from answers which are suspiciously unlinked in the actual answers provided. There are all the suspicious classics of ascribing meaning by combing questions which may have been unrelated in the respondents minds; two-thirds identify certain core systems as business critical (obviously), but instead of telling us that the respondents feel that recruiting problems have them worried about maintaining those systems, the survey instead says "Less than one in seven (13%) CFOs are very confident that the knowledge and skills exist within their organizations to maintain core IT assets into the future." Now, what happened to the HR guys and CIOs? And why say "very?" To me, that seems suspiciously like someone looking at a standard five-part multiple choice survey answer in which most of the respondents actually said "confident" or "moderately confident" and saying, "Well, hey, only 13% of them were very confident... let's go with that!" That's creating a Tempest in a teapot, and I don't believe it reflects any great crisis for IT.

But let's talk about recruiting and talent in the industry. In fact, there is always a shortage of talent in IT, but it's not defined by the number of people available who know COBOL or Ruby on Rails (to pick the opposite ends of the spectrum on technologies). This industry moves fast and few of the more talented people in it are dumb enough to stake their careers on one specific technology. The most valuable people are those who can learn whatever you put in front of them, who have experience in a dozen different technologies because they were smart enough to make the transitions when they had to. I've never met a hiring manager who was completely happy with their choice of recruits, either, even in times when the industry is on the skids and it seems as though thousands of technical staff have just been dumped on the street after layoffs. This is as much a psychological response as a technical appreciation of candidates. You always hope to be interviewing better people than you get.

CIOs are paid, in part, to be nervous, and being nervous about shoring up staff with these core competency skills may be more an expression of caution than a symptom of a skillset in decline. While the systems may be important, they clearly aren't unstable... if they were so failure prone, if candidates with such skills were really so necessary, you'd see a lot more of them on payrolls and in the market. There would simply have to be more of them around to keep the systems running.

McKendrick and Linthicum raise real concerns for SOA, but as I keep saying, SOA or any other new technological paradigm will succeed based on their value to business and the ability of executives to appreciate that value. I still believe, as far as SOA is concerned, that the roadblocks continue to be on the perception front more than the talent front. Both Linthicum and McKendrick identify consultants and vendors as among the forces holding back SOA, but those arguments are applicable to many new technologies which present a bandwagon for those of limited technical skills but great salesmanship to hop aboard. In fact, the stability that is being achieved in core business systems in modern IT means that most IT departments probably will neither want nor need to hire too many people specialized in those systems... there simply won't be enough work to make them worthwhile. Consultants and vendors are going to be the expert resources used to implement SOA. You certainly need internal architects to manage the effort, but it's silly to take on payroll people who should rightly be outmoded eventually by the success of their own projects.

There is room to dispute to what extent a successful SOA implementation will achieve this, but I think the argument is true of technology in general... new and old. We're simply finding better ways to apply these technologies, and easier ways to expose the information and power within them to average staff. There is going to be more outsourcing, and there accordingly will be less specialist hiring by organizations, regardless of the importance of their core systems.