Old sysprog shoes fill easy, too - Jay Maynard

> Recent entries
> Calendar view
> Friends page
> User info
> Jay's web page

Wednesday, 2 March 2005

Previous Entry Share Next Entry
1023 - Old sysprog shoes fill easy, too

wilwheaton pointed to an entry in his friend actor/producer Shane Nickerson's weblog about how he slipped easily back into an old, familiar role. I've noticed that, too.

I was a mainframe systems programmer ("sysprog") for 15 years. I was, with all due modesty, very, very good at it. Good enough, in retrospect, that I got away with a lot of crap that I probably shouldn't have been allowed to. I could, and did, handle nearly all facets of installing, maintaining, and managing the software infrastructure of IBM mainframe computing systems. One guy I worked for told me that he loved to see me walk into the computer room and sit down at the console when the system was having problems, because my touch was almost magical.

When you do that kind of work for as long as I did, some of the things you do ingrain themselves into your brain indelibly. I shifted my career focus in 1995 because I just couldn't stare another packaged system installation in the face. I moved from the mainframe arena to Unix systems because I thought that was the next big wave, and my experience across multiple platforms would make me even more valuable. Time has proven me only partially right.

Part of my consulting work is in doing mainframe stuff, although I very seldom get my hands on the actual systems themselves. I do, however, do a fair amount of systems work as part of my work with Hercules - after all, a mainframe (or a mainframe emulator) isn't very useful without anything to run on it.

I routinely scare myself with just how much I remember when I do sit down in front of a mainframe console. My fingers seem to know their tasks without conscious intervention. I have little trouble remembering which of the hundreds of system configuration files contains a particular parameter, or the meaning of a specific ABEND (abnormal end, what the operating system does to a misbehaving program) code. If I do need to look in the documentation (which, thankfully, is available online), I know exactly which book to go to out of hundreds and how to interpret the IBMese found there - perhaps the system programmer's greatest skill.

Realistically, I don't think I could get a job as a full time systems programmer. The field has advanced a lot in the past decade, and there's too much I don't know. I would also be astonished to discover there was a mainframe within 75 miles of Fairmont. It is nice to know that I haven't completely lost it, though.

current mood: [mood icon] contemplative

(Leave a comment)

> go to top