IT Management

IT Management:  What Is a Mainframer?

When I first started working on the mainframe as a CICS systems programmer in 1987, the mainframe had already been around for 23 years and had a well-established culture surrounding it. In fact, the culture was actually older than the mainframe itself, having begun to self-assemble around the time of the emergence of business computers in the ‘50s. One of the first significant manifestations of its existence was the arrival of SHARE in 1955 in response to the IBM 701. Consequently, much of the pre-existing culture of the folks who first joined SHARE became foundational to the mainframe culture.

A strict sense of responsibility and a dogged pursuit of solutions to very challenging technical problems became a hallmark of this culture. At the same time, although very technically focused, mainframers have always been business people. No matter how deeply buried we are in the technology we support, it’s inescapable. Corporate culture, bottom line, departmental politics: It’s all part of the world of the mainframer. In that context, over the past 46 years, the various roles on the mainframe have emerged, built on this basic culture, and added its own facets.

So, in June 1987, after spending four years at the university learning about UNIX and PCs, I came face-to-face with a technology and culture that were completely different from anything I’d ever experienced before.

Among the first things I learned about were 3270-type terminals, 80-column-wide text members in partitioned data sets, 132-column-wide printouts (actually, 133 if you counted carriage control), 44-character, all-uppercase file names, and controllers that took everything you typed on your terminal and sent it all at once to the mainframe when you hit Enter. I also learned about the Reset key, or, as one of my colleagues once labeled it, the “I’m sorry” key, because if you typed when or where you weren’t supposed to, you weren’t allowed to type any further until you hit this special key to apologize for doing it wrong.

I also learned the culture, including just how careful you had to be before changing anything. It didn’t take long for change control to become a standard part of my existence.

That was half the mainframe’s lifetime ago, and many of the generation I learned the mainframe culture from are still keeping the mainframe running today. They’re also encountering a growing number of new mainframers who are ready to learn all about the technology and culture.

What will make this new generation “mainframers”? For the first time, I’d say it won’t be having to work with technology that seems obscure to anyone not steeped in it, thanks to the new, leading-edge graphical workspaces that are now becoming available. Instead, I’d say it’s that culture of scrupulous responsibility, of being a business person and not merely a technologist, that will define future mainframers. In fact, let me be the first to suggest that, ideally, “mainframer” may become a word something like “gentleman,” which used to designate a particular role, but came to represent the conscientious and considerate behavior associated with that role, no matter who played it.

This is particularly relevant with the merging of platforms implicit in IBM’s recently announced zEnterprise server. Suddenly, the “traditional” mainframer, who has likely already learned UNIX with USS, PCs with terminal emulators, and often Linux (both PC and mainframe versions), will have the opportunity to work with platforms not traditionally associated with the mainframe as part of that context. Of course, this means we’re going to need more mainframers. So, who will they be and what will they be doing in the future?

It’s likely, except perhaps in the largest mainframe shops, that each mainframer will have many distinct roles to cover—sort of like being a general practitioner. But, if a new mainframer starts by being a “jack of all trades,” he or she will certainly be likely to master many of them over time. However, time is no longer a luxury (if it ever was) that will be afforded to new arrivals on the mainframe space, so doing things the hard way will not be an option.

For that reason, we can expect fewer local customizations, as organizations look to reduce loose ends that could tangle up new mainframers. While in the early days of the mainframe, homegrown SVCs, exits, and various and sundry kludges were the order of the day, today’s mainframe management software offers a wealth of functionality that makes such exceptions unnecessary. This is especially true of leading-edge graphical workspaces.

However, none of this will change the need for key mainframers to be intelligent, knowledgeable, experienced, and possess strong work ethics. In other words, to be true professionals. And that professionalism, already tried-and-proven in today’s IT world, would appropriately be the dividing line between those who are given the opportunity to prove their worth as real mainframers and those who are encouraged to consider other career directions.

The future of the mainframe is brighter than ever, and I’m looking forward to knowing and working with the outstanding people who will be known as tomorrow’s mainframers.