Misty Decker on Communicating the IBM Z Story
Reg Harbeck talks with Misty Decker about her unexpected mainframe career path, the IBM Z Academic Initiative and how to encourage others to join the IBM Z community.
By Reg Harbeck07/01/2019
Reg Harbeck: Hi, I'm Reg Harbeck and today I'm here with Misty Decker. Misty is in charge of the IBM Academic Initiative, which is really key to the future of the mainframe. Misty, welcome. Can you tell us, how did you end up on the mainframe?
Misty Decker: It was an accident. I went to college; my undergraduate I had to pay for myself and I wanted to be a university research mathematician so while trying to finish my math degree I ran out of money. I was literally starving. I had no food. I sold everything I owned to finish out my college degree and my roommate said you know if you get a job with one of these big companies, they'll pay for your master's while you're earning a paycheck. I said, score! That sounds awesome. So I interviewed with IBM even though I had zero interest in computers, pretending to show interest only so that they would pay for my master's degree. They offered me a job; they offered me an interview in New York and I'm from Virginia. I'd never been on a plane before.
Reg: Oh, wow.
Misty: So I took the interview only to fly. That was the only reason I took the interview because I had no intention of moving to evil New York, which is what we think of it in Virginia. So I was very impressed with the people and then I figured, you know I'm going to be getting my master's degree. I could live here for a couple of years until I'm done with that. IBM policy at the time required that you have to work for at least a year before you could apply for the tuition reimbursement program and during that year I fell in love with the mainframe and I fell in love with computers. I was very surprised. I am suddenly changing my career path and I pursued a master's in information systems and I've been in mainframes and computers ever since.
Reg: That is really cool. Now have you been based in Poughkeepsie the whole time?
Misty: I have.
Misty: So my family the first few years was like “Hey Misty, I got a truck. I could move you tomorrow,” and begging me to come back home but I fell in love with the Hudson Valley. I fell in love with the people that I work with. I then eventually fell in love with a man and got married.
Reg: Now did you meet him at IBM?
Misty: Actually I met him at Woodstock.
Reg: Oh my goodness.
Misty: So my husband is a retired IBMer. I didn't know it but we had a mutual friend and she asked me to go to the Woodstock 25th anniversary with her. I met this guy I thought was super cool and he was not at all interested in me because he was busy volunteering for Habitat for Humanity. I had a strict rule. I do not date anyone from IBM or anyone that I might possibly work with. I didn't know that he was an IBMer but on Monday, the day after Woodstock, I realized he sat four offices away from me.
Reg: Oh wow.
Misty: And I'd never noticed him before but now he's ruled out. I can't show any interest in him but now he is interested in me and being nice to me. I was like “No way man, I can't do this,” but over the course of the next year we got to know each other. I realized he was too perfect for me and he was worth breaking the rule so that's how I met him so it's kind of through IBM.
Reg: Yeah, interesting.
Misty: Not really.
Reg: So it was sort of a Woodstock experience of your own.
Reg: Now what was he working on at that time?
Misty: So at that time he was working in System Assurance Kernel (SAK). It's an OS that does nothing but test every single machine instruction. He then moved onto AVPGEN which generates little mini programs to test every single machine instruction and he led that team. He had a little side project that people may have heard of outside of IBM called zPDT because he is one of the few people that actually knows the entire hardware architecture end to end. They asked him and another colleague to put together a little prototype of an emulator to emulate the architecture on x86.
Reg: Cool. That's a pretty important emulator. Now meanwhile all of this time I gather you never were actually in the same group as him even though you had this parallel career to him.
Misty: Actually for a while we were in the same second-line organization. He did eventually move to work full time on zPDT and they put that in the same firmware organization where I was a development manager for LPAR. I had coupling milicode and Sysplex timing protocol, FICON and SCP so I was a first-line manager and he was a team leader in another department so kind of close.
Misty: But we never worked together per se.
Reg: So you've covered quite a bit of ground in your career while being on the mainframe. What sorts of things have you done and been in charge of?
Misty: I started in build. They were hiring a lot of mathematicians into build to manage the install logic. From there I was asked to join the brand new project management career path so I went through the PMP certification when it was first offered to IBMers. I don't know how many—I've been a PMP for a very long time and I took a job in release management so I was the release manager for z/OS 1.1 which is the OS that first exploited 64-bit architecture so I was able to release manage that.
Reg: So did that mean you got work with Bob Rogers?
Misty: Oh, yeah. I knew Bob Rogers since my first day on the job.
Reg: Oh, cool.
Misty: Bob had a habit of going around and greeting all of the new people at that time. There were not a lot of us getting hired in the early ’90s. Yeah, small numbers.
Reg: Because I know that 64-bit automatically triggered that in my mind because I know he was pretty heavily involved in that as well.
Misty: Oh, yeah. What hasn't Bob been involved in?
Reg: So now next after that what did you do?
Misty: I left on maternity leave and I came back to a job that I could do part time. I went into the customer sat office and I led the z/OS customer advocate program where we pair people in the lab to individual clients to act as their inside man so I managed that project for a number of years and then I moved into firmware. After firmware, I wanted to do something a little more fun and I saw this job called university alliances where you set up collaborative research between IBM and universities all around the world. I needed to find that sweet spot between something that makes a difference in the world and those technologies that they would be leveraging, that research would demonstrate the capabilities of the platform. What was interesting about that job was I only had one project that was on IBM Z. The mainframe is very hard to do research with for a number of reasons that we don't have time to go into. Most of the work that I did was on high-performance computing so that gave me a chance to really learn about that other world. It was kind of fun.
Reg: So that was your first involvement with academic stuff, but I gather definitely not your last. What happened next?
Misty: Oh, so now I was having a lot of fun in that job but I really was itching to get back into Z. Once a mainframer, always a mainframer, and I saw that Don Resnick who led the Academic Initiative was retiring and I applied for the job. They said, well it looks like you've been spending your entire career to prepare for this one job because I had been in hardware; I'd been in software. I'd run an advocate program which we have ambassadors volunteer their time all the time. I had done the research side so I somehow had managed by accident yet again to put all the pieces in place to make me perfect for this job.
Reg: That is so cool. Now I'm going to guess that was around the time that you met Dr. Cameron Seay.
Misty: Oh, yes, I met Cam a number of years earlier actually.
Misty: I was very active in volunteering as one of those ambassadors for the Academic Initiative and I went to the Enterprise Computing Consortium (ECC), at Marist College and Cam was on the board of directors for that.
Misty: So I'd met him through there.
Reg: Now you're in a really critically important position now because of course the future of the mainframe is really open for each one of us to be involved in making it happen properly. We can't just sit back and expect it's going to happen. What are some of the things you personally are planning to do to make the future of the mainframe bright?
Misty: Well empower everyone that listens to this podcast to go out and spread the good news. The mainframe is cool again.
Misty: So I am on the road a lot just talking, talking, talking. Good thing I like talking. So I talk to faculty; I talk to students; I talk to mainframers; I talk to non-mainframers; I talk to the poor person sitting next to me on the plane. I try to share the story because what is going on in the mainframe right now is just so exciting. It is the platform of the past but it is also the platform of the future because all of the major challenges going on—not all but most of the major challenges going on in the world of IT are things that the mainframe is naturally good at.
Security is a big question that a lot of people are struggling with. This is the best platform to be managing your secure data on. Large scale transactions, Internet of Things is driving vast numbers of transactions. Nobody else can handle that I/O like the mainframe.
Analytics. Analytics was great when you could move the data off the mainframe and then spend a week doing your analytics and then coming up with your customized mailer but now it's real-time analytics and that is best done on the mainframe because that's where the transaction is happening.
Misty: The other thing about analytics is a lot of people are moving to things like Spark with in-memory analytics and getting that large, large amount of memory necessary for any really good size data set is so difficult on other platforms; only the mainframe has that giant, giant pile of memory that you can use for something like that.
Reg: I have to just do a parenthetical remark because as I've been studying the history of the mainframe, I've discovered that a lot of these terms we take for granted and use as day-to-day terms on the mainframe that seem so different from everybody else actually have their origins in pre-computing and one of those is data set. When we say data set on the mainframe that actually has its origins in academia where they still use that term for something similar but not identical and you just used it that way.
Misty: So a data set for me is a math thing.
Reg: Yeah, exactly.
Misty: Right. We can start talking about fields if you'd like. Sorry, that's a math thing.
Reg: Which is your field.
Misty: Yeah. Oh my gosh Reg. I knew you had to get a pun in there somewhere.
Reg: So now for those listening to this podcast, I am going to give you the opportunity to actually ask them to do something to move the future of the mainframe forward.
Misty: Oh great.
Reg: So please give our listeners some motivation and ideas.
Misty: Oh great. So the one website to rule them all; this is the one website that you need to know so that you can find out all the things that we're doing, all the things that we offer to universities, all the things that we offer to clients, students. It is ibm.biz/zskills [hyperlink: https://ibm.biz/zskills].
On that website, we have links to so much, a lot of good information and what I encourage you to do is to go out and talk to students, talk to faculty, talk to your clients, to your company, to your peers. We have a fabulous story in the mainframe and I ask every mainframer to stop saying, please, please, stop saying the mainframe is not dead. The mainframe is not old because we are to the point now where the vast majority of people, everyone under the age of 30 don't think of the mainframe as dead. They don't think of the mainframe at all. They don't know anything about it.
So my analogy is if I were to go on a blind date with a guy who says I'm not one of those guys that's going to go out with you and then never call you again… I would start wondering, right? Why did he feel the need to say that? So if you talk to people and you say the mainframe isn't dead and they weren't already thinking the mainframe is dead, they're wondering and now you have planted that seed in their head.
Reg: Really important point.
Misty: So don't talk about what the mainframe isn't.
Misty: Talk about what it is just like I was doing a little earlier in this podcast. List all of the reasons why the mainframe is exciting, why it's vital, why it solves problems that people are struggling to solve on other platforms.
Reg: Excellent. Well with the last couple of seconds we have on this, anything else you wanted to make sure that everybody has in mind?
Misty: Just that we as mainframers love the platform and I really want to harness that energy, that love and that passion and bring more people into seeing what this is. I cannot tell you how many times I go to educator conferences and faculty come up, they hear what's going on in the mainframe and they say thank the Lord. I had to leave the mainframe to go teach distributed computing because that's what my university asked me to do but I always had a special place in my heart for the mainframe. Once you get into this platform, it gets in your blood. It forces you to stop planning to be a college math professor and do something else instead. It's that love and passion that I think all of us need to share.
Reg: Awesome. Well thank you very much Misty.
Misty: Thank you Reg.
Reg Harbeck is a mainframe enthusiast who has worked IT and mainframes for over three decades. He's the chief strategist at Mainframe Analytics ltd.