Which is it, shortage of mainframe skills ... or jobs?


There is a lot of noise lately from 2 camps - one swears that the availability of people with mainframe skills is drying up rapidly - they either forecast dire shortages, or note problems hiring for certain positions internally. Most of the trade press articles are firmly in this camp.

Inevitably, when a trade-press article runs on the shortage of qualified people, individuals with mainframe skills lambaste the author by noting their difficulty in finding a job that requires their skills. I recently started monitoring a Twitter site - @MainframeCOBOL - which tweets mainframe job postings at the rate of several a day. Some folks note that it is a salary issue - employers don't want to pay conventional salary rates given the job market.

But the seeming disconnect between the 2 camps makes me wonder - Where does reality lie in your shops? If you use mainframe technology - are you having trouble hiring or keeping people with mainframe skills? Is there really a glut of qualified people? Is it any different than other skill-sets? Is the supply of folks with other skills appreciably different?


re: Which is it, shortage of mainframe skills ... or jobs?

I am an educator who teaches mainframe skills, and there seems to be a disconnect between the companies that use mainframes and those with mainframe skills. It's unarguable that mainframe professionals are retiring, but I don't think their leaving the job market in critical numbers yet. They will, and then mainframe users have only two choices: restaff or move off the platform. In most cases moving off the platform is not realistic

re: Which is it, shortage of mainframe skills ... or jobs?

Thanks Cameron, and welcome. I think that the call to move based on the fear of a lack of skilled resources often misses the root problem - lack of people familiar with the application.Moving existing COBOL apps to a new platform is a viable option below a certain MIPS threshhold - say several hundred to a few thousand depending on the apps environment and several other factors. But moving COBOL means you still need COBOL programmers, so that problem remains unaddressed. Most often I see small mainframe move to reduce the high cost of small mainframe environments - chiefly 3rd party software costs.Above a few thousand MIPS, (and we can argue where this shifting line belongs) I tend to agree with you - there is insufficient justification for gutting and replacing the entire technical environment (which goes WELL beyond just apps and IDEs) absent some compelling business reason. GGetting business people to sign-off on technically-driven projects is increasingly a huge issue because big change repersents big risk, business-people are more technically savvy today, and because the business case for technically-oriented environmental change makes no sense (or no sense to business people).Fix the disease, not the symptoms - the disease in this case is lack of apps knowledge. Applications for which we have little tribal knowledge on one platform will be just as much of an enigma on another platform, and training will address many of our skills issues.There MAY be factors that justify moves from truly obsolete platforms, DBMS and languages, but each is an isolated case, and must be evaluated that way. Broad proclamations of "move off technology X to technology Y" mostly benefit the vendors who sell technology Y, and vendors who move companies from X to Y.