Dr. Joe Gulla Still Hears the Music of the Mainframe
The "z/OS and Friends" author goes on "IT Social Hour" to discuss how he went from music theory major to mainframe devotee, why he still gets emotional about the mainframe, and why, at age 72, "z/Architecture Principals of Operation" still fills him with wonder
Click here to listen to the audio-only version
The following transcript has been edited for clarity:
Andrew Wig: Hello everyone, and welcome to another episode of IT Social Hour. Today I’m excited to welcome Dr. Joe Gulla, who has been working with the mainframe in some capacity for about 50 years. He spent many years with IBM. He’s worked as a technical consultant. He’s been a professor. He wrote for IBM Systems Magazine before that turned into Tech Channel, and most recently he’s the author of the TechChannel series zOS and Friends where he explains the mainframe layer by layer. He blends in lessons he learned along the way from his illustrious career as well. So if you give z/OS and Friends a read, you’ll know it’s written by someone with a deep passion for the mainframe, and he’s going to share some of that passion with us today. So without further ado, let’s welcome Dr. Joe Gulla. Joe, thanks for joining IT Social Hour.
Joe Gulla: Hey Andrew, thanks for inviting me. I’m so looking forward to this.
Wig: Yeah, I’ve been looking forward to this too. We talked a little bit last week as a little preview, and that just got me extra jazzed for this session. I know you’re going to have some interesting perspective to share with us today, and I figured a good way to start out, let’s talk a little bit about z/OS and Friends. Now, last year you reached out to us with kind of a full plan, 24 articles, kind of picking apart the mainframe bit by bit here, and we’re like, yeah, let’s do this. So didn’t even really have to give a second thought to whether we would want to publish this project. And so really, I just want to hear from you what prompted you to reach out to us like that and take on this z/OS and Friends project.
Gulla: Well, thank you, Andrew. Thank you for giving me the opportunity to do that. Been writing, I hadn’t written for TechChannel at all. Prior to that, I had written somewhere between 400 and 500 articles for IBM System magazine. At the time, there was a print magazine, there was a newsletter that went out periodically, and then there was a channel that had to do just with mainframe operation and things like that. So I had different editors and I had written for all three of those channels. So when IBM Systems magazine stopped, I just took a break. I was really on a mission to get my work done every week. Every month I was really banging them out, and then I took a break, so I decided it was time to come back. I didn’t know who the editor was, but I wrote to one of your leaders and she connected the two of us, and I thought the least I could do, Andrew, since I didn’t know you, the least I could do was write a plan for the year. That was careful. That was thoughtful. And so that’s why I had a vision of doing 24 articles over the year. I figured that was enough to cover the domain of z/OS and the Z system. And so that’s the mission I’m on right now. I think I’m two-thirds of the way through. Six articles, six more articles ‘till Christmas.
Wig: Six to go. Wow. I’m kind of sad. We’re on the final stretch. To be honest, you pump these things out like clockwork. The methodical approach you brought just with the proposal of the 24 articles is kind of evident, I think in every single article that you sent to us. Can you give me, just, I guess as a writer myself, I’m always interested in someone’s writing process. Can you just tell me how these come together every two weeks?
Gulla: Sure. Well, I’m trying to do two at a time because every month the ones that I plan for each month are on the same subject. Usually the first one is a little more introductory and a little more survey, and the second one is what’s happened lately or a deeper dive. So that’s the way I did it for CICS and for IMS and for database with Db2 and other databases. So that’s the thing. So what I try to do is I try to outline them both at the same time. So right now I’m working on messaging and integration, and so I’m working as deep an outline as possible for the two of them in the same document. And then I put some things down. I usually, I’m very happy that you’re letting me do this. You didn’t, not complain, but you didn’t respond negatively about the fact that I do so much personal history in there.
I have a deep relationship with CICS, for example, or a NetView. These are things that I worked on and worked with, and I have a lot of personal experience and I wanted to share it. So I tested you early in the year by sending you some things that the first couple paragraphs are autobiographical in a way, and you didn’t say, “Well, Joe, I think you should tone that down a little bit.” So I guess you’ve noticed I’ve kept that going. I’ve kept that going in the article. So that’s part of it is the history.
And then I do have a lot of passion for these things. And now it sounds a little crazy, but I kind of love, I mean, for a guy who likes to program and also likes to implement products, it was a perfect thing because whenever you did an implementation, a customer would always say, well, can it do this?Can it do that? And you never had to say no. You could say, sure, you want to do something every same time, every day, no problem. And I would show ’em how to do it, and it was easy to do because when we worked on NetView, it was pretty mature from the early days. It was pretty mature. So anyway, so I try to do history, personal anecdotes, and then I try to, in all the articles, I try to make sure I cover the topic. I don’t want to assume that the reader knows everything about the product. So I kind of er on the side of explaining things, putting a lot of focus on explaining things.
Wig: Yeah, comprehensiveness, I think you’re kind of a completionist, aren’t you? In terms of you want to cover the domain in its entirety, right?
Gulla: I do. It’s really a challenge to do that. When I worked on my dissertation at ‘Nova in 2000 and I was writing my dissertation, my advisor told me, “Joe, you’re digging a hole for yourself.” And I said, “What do you mean?” He said, “Well, you’re covering the whole topic. You’re very comprehensive in the way you’re doing this, and you can get away on your project from doing a subset of that, doing less of it.” And I fought that and I ended up with a 466-page dissertation report, my final report, and I was my own enemy in a sense. And so Z and Friends, a little bit like that. I should be writing maybe a thousand words on an article. That’s probably what I should be doing. But sometimes I’m up around 2,500 words because in my mind, that’s what it takes to cover. That’s what it takes to cover the topic in a reasonable way.
So it’s my nature. I think I told you last week that I don’t buy one Beethoven Symphony. I buy all nine and I listen to ’em and I want to know how they do—the first one, which is lighthearted and airy, and how they do the Monumental Ninth. And I wonder what did they do with Seven? Seven is the one I studied in college. I was a music major and we studied. I spent a whole semester on Beethoven Symphony Number Seven, chord by chord. That was the harmony class analysis of that piece. So naturally when I hear it, it brings back memories from 45 years ago, and I want to hear in a new performance, I want to hear how does the Seven sound by this guy? So it’s kind of my nature to do that, try to be complete in a way more comprehensive than otherwise, even if it hurts a little.
Wig: Just a compulsion almost. Can’t help yourself,
Gulla: Perhaps it’s a disease.
Wig: So you mentioned the music thing, and so maybe let’s go there. You were a music theory, you actually got a music theory degree before you got into computers. How did that happen?
Gulla: Well, I got turned onto music in high school, probably late in high school, and I actually went to Sam Goody in Philadelphia and bought a recording. I went to the cutout aisle where they had the cheap price LPs, and I picked up a recording of Fritz Riner, Chicago Symphony, Bayla Bartok Concerto for Orchestra, read the program notes and said, wow, this is really important. This must be something really important. Let me read about it. I read about it, and then I listened to it, and at that moment in time, I decided I was going to be a music major. This was the model. This piece of music by this composer was the model that a young guy like me could use to create music.
That’s crazy, right? I mean, nobody in my family went to college. So nobody said to me, “Joe, I don’t think that’s a good idea. You shouldn’t major in something you don’t know anything about. You should maybe get an accounting degree or something.” So I paid my own tuition, so nobody applied any pressure on me, and I went after it. I went after it really hard in college, and I don’t regret it. I had a great experience, and I was basically a theory guy because I was so interested in the system of music—how is it put together? We studied Bach and very easy Bach things like two part inventions. So things in the left hand, something is in the right hand, and I realized that every single note in that piece, well, for one thing, the music is beautiful to listen to, but the other thing I found about it was that every note has a relationship to another note. How is that possible? What kind of system is that? So I just kept that up in college, and so that was the theory piece of it. And then I also did history, and since I didn’t really know the literature, I never heard Tchaikovsky Four or Five and Six symphonies. They’re famous pieces, and I know them and love them now, but I never heard them. So I had so much catching up to do. I didn’t grow up with music as a kid. I had to learn it all in a short period of time.
But the theory was the main thing. History is something that takes a whole lifetime to consume. I mean, I’m 72 and I’m still, I was at a concert on Saturday night, and I always knew that Brahms wrote three sonatas for violin and piano. I knew them. I listened to ’em all the time. I found out at that concert that he actually wrote five, and the violinist played one of the ones that was a transcription. So there’s so much to learn about music and the experiencing music just from listening and attending concerts you mentioned was the theory thing really translates well to computing because programs are part of systems. And what I found out that was, I wasn’t such a great composer, but I was a pretty good programmer. I had more success with the system of programming than I had with the system of music, and that really fueled me to keep going with programming, and it fueled my desire to create systems, to create programs that were part of application systems.
Wig: And basically you started working out for a railroad that was your first foray into the world of these computer systems.
Gulla: And I didn’t know anything about computers. I worked for Conrail for a couple of years, Consolidated Rail Corporation. They were set up by the U.S. government and they were made up of a bunch of bankrupt railroads that had been, they were successful railroads, but they were so governed by the rules, rules of the road that they couldn’t make any money. So yeah, I was like a general purpose helper in the medical department for a couple years, and I was identified as somebody who was underemployed. I had a degree, but I wasn’t really doing anything with it. So somebody approached me from personnel and said, “Hey, do you want to be a programmer?” And I had no idea what that meant, but I said, “Yeah, you bet. I’m going to do. Yeah, sure. When do I start?” So they put me in a three-month training program and taught me COBOL, basic JCL. And so when I graduated from that program the next week, I was a programmer.
Wig: Did the language learning, did that come pretty easy for you then basically with your music theory background, and I assume just a general predisposition towards math and those sort of things?
Gulla: I think it did. I think I had three months. It wasn’t like you do some 10 class and then you graduate. We had to complete assignments and programming assignments. I think there were nine assignments, and they kept track of how many times you ran your job to make it work, to get the successful completion. It wasn’t just getting a compile, it was getting a run, a successful run. So we produced reports. It was batch programming. So yeah, I did it. I did it. And then the next week I started in transportation and billing. And the TAB system just turns out that the reason they were so desperate for programmers was that they were trying to redo basically what was done in the rail yards. The rail yards had it, they had intelligent typewriters, input output devices, async typewriters, and they were going to do a major upgrade to the system, including CRTs, right?
The green screen, 3270 devices. So that was the big upgrade, and they were doing a lot of changes to the system itself. So I finished on a Friday, graduated on Monday, I was assigned to an analyst, and I was the luckiest guy in the world because the analyst they assigned me to was Daniel J. Murphy, DJM. I’ll never forget him. And I worked for him for a year, and he was a wonderful human being, and he loved a mentor and he helped me through the difficult transition of being a batch programmer to being an online CICS programmer. So it was the very early days of CICS was as crazy as it sounds now, you can no longer do this, but at the time you actually put assembler language macros in your COBOL program.
So instead of saying link to a program, you said D-F-H-P-C type equals link name equals. And so you had this weird assemble language macros sandwiched in your COBOL program. That was the state of CICS at the time. It was a hodgepodge, and so naturally you had problems, you had challenges. And so Daniel was there to help me desk, checked all my code and helped me with the testing and everything. So he was a great guy. He was so important to the company, to Conrail that there was a C-I-C-S transaction called DJM. If you typed in a production system, if you typed in DJM followed by the car initial, which was four letters, and the car number, which was six letters, it would dump the record onto the screen so you could do debugging in production. So he had his own special system-wide CICS transaction. That’s how important he was. So he was a great he guy.
Wig: So that was basically your introduction to the mainframe. Did you view it at the time as this world spanning foundational platform that was still going to be around 50 years later that you would still be working on 50 years later, or was it just it was a platform where you could kind of apply your skills or did you have an inkling that of what would go on to progress and become
Gulla: Yeah, I’m sure I did have, I had more than an inkling. What I had to do at the time was separate the rail’s implementation of the mainframe with what the mainframe could do. And I mentioned Conrail was made up of a bunch of bankrupt railroads, so they didn’t have enough resources to do the things they wanted to do. So computer time was submitting for compiles and things like that was challenging to find enough computer time to do that. I remember that the test system would come up at nine o’clock in the morning and somebody would walk on the floor and say, test system’s up, and sometimes by nine 10 it was down. And it wasn’t because the mainframe was terrible, it was because some programmer didn’t know what they were doing and brought the test system down because they didn’t have a Daniel Z. Murphy helping them deal with the complexities of that environment.
So I knew it was a big deal though. I knew that when I signed onto TSO and I looked around me and I saw all these other people on TSO, and then I saw the test, CICS and the production CICSs and production database regions, I knew that something big was coming on here. There was a tremendous simultaneity. I mean, what system is managing all this concurrent activity and sometimes not competing with competing with one another. I mean, there was the high priority work, production work can be done, and then there were these programmers getting things done. I mean, I knew there was something magical going on, but I never saw the machine. I never got into the computer room. It was an abstraction to me at that point.
Wig: Okay, interesting. It was just this, yeah, I feel like in people’s imagination it has that kind of mythical place.
Gulla: That’s the right word. It’s mythical really. I mean, now that I’ve studied architecture and I’m not really an engineer, but my whole life I’ve gone back to the architecture, try to figure out what’s going on, and just in the writing of the articles this year, I got into microcode more than I ever got into, which is an engineering topic, and I got into the topic of instructions being hardwired quote, which means they’re relatively simple. And so one instruction cycle to do it. And then I got into microcode where, well, I didn’t get into it. I read about it and the microcode says, the more complex instructions, there’s a little program that actually is behind the scenes, and it consists of several instructions and sequence that actually make the instruction take place. So it’s a mini program that’s part of the hardware for the more complex instructions. Those kinds of things I didn’t know about. I couldn’t even understand if I read them because I didn’t have the background. I didn’t have enough experience to really understand all that. So I’m still learning. I’m still learning those things and looking forward to reading about them.
Wig: I guess there’s always something to learn. How is something as complex as the mainframe z/OS—I wonder if there’s anyone who has a complete library of knowledge in their head. I don’t know if that’s possible.
Gulla: Yeah, I met some people. I think when z15 happened, I was writing for IBM Systems magazine and I interviewed part of the team that was working on z15 and some of the engineers that I met. It’s totally unbelievable what they know. It’s really just humbling, humbling their insights. They understand the history because of that instruction set, and then they understand how to extend it, which is two different things. To have a brain that can handle enhancements while keeping what’s there not to break, it’s pretty remarkable, pretty remarkable thing.
Wig: And with z/OS and Friends, it gets pretty technical, but also you blend in emotion with that and your own personal history and it gets almost sentimental, and you kind of maybe don’t expect someone to take both of those approaches. Maybe it’s not, kind of an unfair air presupposition, but you kind of think the technical guys are going to just be technical and they’re going to throw all the emotional stuff away, and then the emotional guys will take that approach. But you are on both sides there. I’m interested in that side of things. Where does this passion for the mainframe come from? What gets you excited about it?
Gulla: Well, you’re right. It seems to be with people one or the other, the intellectual dry about it, and the emotional people don’t really know what’s going on, but have a feeling for it. It seems to be one or the other. I don’t think I’ve ever met anybody that speaks in the emotional way that I do, writes in the emotional way that I do about mainframes. I haven’t encountered that. I know there’s plenty of people that make a living doing it, but I think it comes from my arts background. I think the fact that something like z/Architecture Principles of Operation, that 2,000-page book, the fact that it can fill me with wonder, and I mean that it fills me with wonder. I routinely go back to that and read sections of it. I just read the preface over again the other day, a formal section called The Preface, and it’s so wonderfully understated.
It says, this book is for several language programmers. It doesn’t tell you how to write a program or anything like that, but it tells you if you’re going to use the thousand or so instructions in the architecture, this is how those instructions work. That’s really what it’s about. It says, and if you’re not, and if you’re not in a summer language program, you only find this interesting too. It’s so humble and understated, and then it proceeds to present itself, and it’s just like, oh my God, this is a tremendous human achievement. I’m not saying that as a former IBM employee, I’m viewing that as a person who is observing a technology that is so deep and rich, and so it fills me with wonder. So what else?
I would say it’s like a great piece of music, but I don’t really know a piece of music that has this history of z/Architecture. I mean, I could say it’s like the Beethoven Ninth. Well, Beethoven Ninth is one person’s work, and it takes about an hour to perform. So the architecture is the result of many humans working, being preoccupied with making the best thing they could make over decades. So I can’t compare it to the Ninth. Can I compare it to all the nine symphonies of Beethoven? Maybe? Do they fill me with wonder? Yes. So I don’t know. I think it’s because I have a deep interest in love of music. I think it helps me to appreciate other things that have the potential of filling you with wonder. I think that’s where my emotional connection comes from.
Wig: So you’re geared to kind of appreciate the music the mainframe makes, I guess, in a way.
Gulla: Yeah, it’s a mix. It’s a mix. And the theory and theory too. I mean, once you dig into something and you understand how it’s put together, you can pretty readily determine if it’s remarkable or not.
Wig: We were talking earlier this interesting concept of, I think there’s an idea of knowing why or how something works can take away from the magic of it. If you know how music works, if you know how the notes fit together and the math of it, then maybe that takes away from your pure enjoyment of it. But that’s not the case for you. You were telling me, no, it’s not, not the case.
Gulla: I mean, I mentioned the Bach two-part inventions earlier in this talk. I listen to them. I often listen to them when I write because they’re easy to listen to. There’s not too much noise. There’s two voices working together, and I listen to it over and over again, and there’s no explaining it how beautiful they are. You can’t. It’s timeless. I listen to the late Beethoven piano sonatas. There’s a group of sonatas starting around number 28. He wrote 32. I listen to 28 to 32 all the time, all my walks, and I’ve listened to him, I don’t know, a hundred times. Some of them I’ve listened to the E major number 29. I love major. Something about the way that makes my cells feel. I’ve listened to it over and over again. I don’t think it’s an exaggeration to say when I listen to it, it seems like the first time there’s such a level of abstraction, and I know what’s coming next, but I don’t necessarily, doesn’t feel like it’s a repeat. I don’t know. So I think that abstraction, I get an emotional response to abstraction that you can be in music or it could be in computing. I think there’s something similar about that experience.
Wig: So maybe there’s someone out there listening to this hopefully, and they’re like, wow, maybe they don’t work on the mainframe yet, and they have a developer background. Maybe they’re just getting out of school and maybe they’re inspired. I really want to check this mainframe thing out. Where should they start? What are the things they need to learn first to get going down the path where they can appreciate the music of the mainframe like you do?
Gulla: Well, before we got on this call together today, I went on to Copilot, which is Microsoft’s AI, and I said to Coilot, “I’m new to z/OS, what do I need to know?” And I got this long answer back. So lately, I don’t know about you, I don’t know if you’re using any of the AI tools, I’ve been kind of using them in place of a Google search, and I’ve been using Copilot in place of a Google search, and I ask it all kinds of questions, and I don’t usually ask it a simple question. I treat it like I’m talking to Andrew and I make, sometimes it’s a paragraph with periods and question marks, the whole deal then. So anyway, I think that a good way to get started with z/OS would, of course, the IBM website has tremendous materials for beginners, beginners, and all the product details you need, but maybe something as simple as an AI search and using that as a starting point would be an okay place to go.
Wig: Okay. So you endorse the use of your chat GPT or your copilot or your—
Gulla: I haven’t really tried anybody else. I like Copilot because it puts a little end note at the end of a, when it makes a statement of fact, it often puts an end note, and then there’s a list of sources underneath. So I can go to the sources myself, so I don’t have to trust Copilot. I can look at it Copilot sources. So I think it’s useful for anything in the area of z/OS to get started.
Wig: Yeah, I mean, I’ll say some of these GenAI tools have helped me learn the things I need to know just for this job as any article, quick information, especially if you want something on the fly. And of course you’re always double checking things, but you don’t want to sort through pages of search results and just get what’s probably the right answer fast. Yeah, super helpful.
Gulla: Probably the right answer. Yeah. If I ask questions about mainframes, I usually have a pretty good idea ahead of time what the answer is. So one thing that I’ve found with copilot is that it has a more complete view sometimes than I do of the topic. There’s a few things that I didn’t know about, and I explore those few things and I say, yeah, they fit in. Yeah, that’s right, they’re right. That’s appropriate to include in the broader discussion. So it’s useful. So I hope you don’t think, Andrew, that’s a cop out, but in the days of ai, it kind of makes sense to use it for simple things and see what you get.
Wig: Yeah, you’ve been a mainframe instructor and I think there’s a lot of talk out there about the future of education with AI and what the role’s going to be, and sounds like maybe, yeah, someone really wants to dive into the main, I mean, you can go down the rabbit hole there and probably learn a lot. I can imagine.
Gulla: Sometimes people ask me, if you were mentoring somebody, if you were mentoring, and I have mentored people not for, it hasn’t been a long time since, it’s been a long time since I mentored anybody in z/OS, but what do you do? How do you deal with that mentoring situation? And first thing I want to know, is this person going to be an application developer or are they going to be a systems programmer? And if the answer is systems programmer, then I have five things that I want to explore with them. I want to know what their general knowledge of z/OS s is. Because you don’t want somebody to be a trained monkey. They just do one thing. You want ’em, you want ’em to do one thing or several things well, but you want to make sure they have a broader knowledge. So general knowledge, and if you’re a systems programmer, what programming skills do you have?
I think it’s still pertinent that a systems programmer know basic Assembler language, have taken, studied BAL. I mean, I don’t think that skill, maybe you won’t use it as much as you have in the past, but that’s a really important skill to have to understand better the nature of the programming environment you’re in. The other thing is what about experience in installing and maintaining products? There’s the S and PE program that’s used with IBM and other suppliers we receive, apply, accept. Do they know that process? Do they know how to protect the system if they’re going to do an install, do they know how to back up in case they have to make a mistake and they have to restore? So those kinds of things. Do they have experience planning? If you work for somebody, they’re going to want to know what’s the plan to do something.
So you don’t have to have a comprehensive, maybe you don’t have to have a comprehensive project plan, but you do have to have a plan. So what’s the plan? And then, can you put that plan in writing? Because everybody knows that supervisors and managers want to know what you’re going to do, and they want to read it and they want to read it in English. So that’s the kind of way I answer. If somebody’s new to ZOS, how do they get going? It depends on what their job is, but in the systems programming side, there is a path to take it. That’s the path I took when I was a beginning systems programmer. My manager sent me to Assembler language workshop in New York, 4.5 days. I went from not writing code to writing code in 4.5 days. In Assembler.
A couple months later, he sent me to the SAMDAM interface, sequential access method, basic direct access method for Assembler language programmers. So I’m really happy that I got that training because it’s useful. I actually did write some Assembler language in my career when I worked for IBM we had a customer that I was having problems with terminal cleanup. When the session was over with CICS, if the client didn’t log off, there was some history of that transaction that stayed behind and it was a security exposure. So I got involved in diagnosing what happened there, and I wrote the Assembler language program called the Terminal Error Program, ATEP. And what the ATEP did was cleanup when someone rudely ended a session, and the only choice was to do it in Assembler language. So you may think you’ll never use it, it’s nice to know, but you’ll never use, and then next thing you know it, you’re using it. So I still think it’s an important skill for assistance programmer.
Wig: And there’s stuff that you’ve worked on or written that is still being used today, right? You’re talking about stuff you did with NetView, and how cool is that to see things still in operation that you worked on however many years ago?
Gulla: Yeah. Well, I, I don’t know. It’s a matter of pride, I guess. I don’t know. I don’t exactly know why I feel the way I do about it emotionally. I think it’s, but I was in the field. I was, when I worked for IBM early in my career, I was a guy in a branch office. And the difference in our branch was our branch was a services branch. This is before IBM had a big services business, so we were like $140 an hour kind of guys. So somebody in the client site would say, my client has a problem and they want IBM to help us fix it. Can you help me fix it? That’s the kind of guy, it was more than just fixing things. So anyway, IBM came out with this NetView solution pack. NetView was available. It was a collection of other previous products put together, integrated better.
And we were doing these solution packs, and I really was excited about doing them. It was the same project over and over again for different clients, but since every client was different, the experience was different. So it wasn’t boring. It was the same thing over and over again, but it wasn’t boring because the client mix was different. You always learn something. And I guess I came to the attention of the development team in Raleigh and they wanted to hire me. I was doing a lot of these. I was doing a lot of these. And then I was doing things like enhancing education and things like that and saying things back to them. The next thing, they wanted to hire me, took the job, and I transferred my family down to North Carolina. And I worked on the Automated Network Operations Solution Pack, ANO Solution Pack, and I worked on the ACO solution pack.
ANO became a feature, it’s currently a feature of NetView. So you can go in there and it has a help desk and other tools. So I didn’t do much code work on there, but I did a lot of communication with ANO. So I was so happy when I wrote the NetView article last month to see that that feature was alive and well. And the other thing I worked on was the ACO solution pack, Automated Console Operation, which became System Automation for z/OS. So that’s the product that you add onto to NFU in order to automate the startup shutdown and recovery of a z/OS system. So it became a strategic project product. I worked on the CICS, IMS and and workload features of that, and they’re still alive as well. So I was really happy to see them. And actually they’re better than they ever were. The lab they went to really improved them.
They put ’em through a lot of rigor. I think they ended up in Germany and a lot of rigor and better than ever. And it was emotional to see that. And I admit that we were a small team when we were working on that stuff. I mean, it was a handful of people. So take it from a solutions group we had where we were just doing, we were doing add-on things to a program product, and you put it into a lab environment where that’s all they do is churn out program products. They’re going to improve it. They’re going to make it better. So it was exciting to see that. No question about it. I was really happy with it. It’s still a lab.
Wig: And so this stuff is still in use on machines such as z17 that just came out. And I’m curious about your take on that latest hardware release and really, I mean, what it says for the direction the mainframe is going. What do you think about the z17 and where things are headed?
Gulla: 17 is great. I got started on taking a close look at the Z around 15, I think 15 was when I wrote some articles for the IBM System magazine. So I met some of the people that were doing the development work, work project manager, not so much nuts and bolts engineers, but project managers and the people that were leading the direction of it. And I learned something about them that gives me a lot of confidence that they are maniacally focused on wants and needs, customer wants and needs. If you ask me what is the current mechanism now, or they have customer counsel and all that stuff, I’d have to refresh myself on what their current practices are. But they are very much focused on two things. One of them is doing, putting into the product things that customers need. So that’s a big, big focus.
So I remember when I was reading about z15 and making a list of the items, there were some changes made in there to allow programs to run in the batch window faster. So there’s kind of a window in time, a window in time in some customer environment where they have to run a bunch of jobs and they have to be done quickly. And those jobs do math. They do ad subtract, multiply, and divide. So IBM re-engineered the ad, subtract, multiply, and divide instructions so that they use registers and not memory. And so they executed percent faster, thereby reducing the window.
What I just said to you is probably not a hundred percent correct, but that’s what I came away from change the instruction set, actually change the instruction set to do more efficient arithmetic so that arithmetic would be completed faster and the overall job would be completed faster. So I mean, if that isn’t a powerful response to a customer need, I don’t know what is. And that was just one of many things. So that’s one thing, paying attention to wants and needs, and those are special because these are large enterprise, these are enterprise customers and they have very challenging needs at times. And the other thing they’re doing though is they’re not closing their eyes to what is needed that perhaps isn’t a want or need right now, but it’s going to be something that’s going to lead. And so I see a mix of those things. I don’t know if it’s 50-50 or whatever.
And I used to use the word, I think I mentioned this to you last week. I used to use the word that the mainframes were evolving, and then I had a sweet dream one night and the voice said to me, “It’s not evolving, it’s being made by humans.” And I said, “Thank you for that vision.” So it’s not evolving. I met these people. These are dynamic people who care about the future of the mainframe. They care about their customers, they care about technology, and they’re doing things to those machines that are going to make them continue to be useful in the future. So I mean, I’m not an insider, but I have a high degree of confidence that this work is going to continue.
Wig: You’re not the only passionate one about the mainframe out there. So it makes sense that there would, yeah, there’s others out there that also.
Gulla: It makes money, too. There’s a revenue implication of new releases. I know the mainframe, I was reading about the business cycle of the mainframe, and somebody reached out to me, I don’t remember who it was, but somebody, one of the top salespeople for Z architecture machines, and I guess he read something on TechChannel. He read one of my articles in TechChannel and said something nice to me. And then he said, “We got a real nice bump from Z 17.” So I looked into the sales cycle for a bit, and the way the mainstream sales cycle works is there’s always a bump when a new Z comes out and then there’s a tail. It’s almost like a book sales. You get a bump and then there’s a tail, there’s long tail, and then you get another bump. So there’s an incentive from a sales point of view and a revenue point of view to keep the releases coming at a pace that could be consumed. It’s not too rapid, and that it accumulates a bunch of line items, development line items that are once in needs of customers, and then looking out on the horizon and seeing what technologies are important.
And there’s always changes in security. Security is one of the sections in z17 that it’s improved now and it’s been constantly improved. Another area of course is AI, because AI is used for things like when you make a credit card transaction. I paid somebody to do some web design for me and one of my other businesses, and the Mastercard didn’t take, the Mastercard didn’t take, because AI said, you’re doing something over the web with a vendor we don’t know about. You’ve never done work before and you’re spending a lot of money spending $3,000. So I’m going to check on that. I’m going to deny that. I know that that seems pretty straightforward, but I know that there’s a decision engine making that decision. So that’s important. That’s important to credit card companies. And the algorithm has to be more important than Joe Gulla spending $3,000 on a vendor he never worked with before. There’s a lot more to it than that. And I’m sure AI is playing a role in that. And it’ll continue to play a role. We would want to play an even more sophisticated role. If I was American Express or Mastercard, I’d want it to do more for me smartly.
Wig: For sure. Yep, yep. It’s good to know that something’s out there making sure everything’s on the up and up. You mentioned books in the book industry, and that’s one thing we have to wind this down here pretty soon here, but before we go out, I wanted to hear a little bit about that. That’s one thing we haven’t talked about that yet. That’s kind of a left turn for you. Seems almost, you’re in the publishing industry now publishing children’s books and memoirs. How did that come about?
Gulla: Well, my wife is an educator and she had very good relationships with authors and illustrators through the job that she had in school systems. She would bring them in as somebody to come in and do storytelling for kids. And so she got to meet some important people in the book, important authors and illustrators I should say. So she had the idea that we would start a press and she wrote a fairytale, and we used the fairytale and we published the fairytale, and we used it as a calling card so we could show it to other people and say, “Hey, we’re a new press. We know what we’re doing. Look at this book and do you have anything for us to publish?” So that’s how we got started. We used her book as a calling card, called “On The Wings of the Swan.” One of the first books we got to publish is this book here.
This is “Walk Together Children,” which is one of two books of black spirituals, written and illustrated by Ashley Frederick Bryan. Lemme see if I can show you what his artwork looks like. Those are linocut prints, and the music to go along with it. So he became a friend of ours and through my wife’s work with him, and he said, “Well, I have a couple books that are out of print. Maybe I have rights back to them. Maybe you want to do those for us.” So we did two volumes, that one and another one via one and volume two. And that put us on the map. If a little press like Alazar Press could publish two books of, important award-winning books by Ashley Frederick Bryan, oh my God. Look at the back of this. This is Rock and Jerusalem. It’s so interesting. These see that it’s a color image, right?
Wig: Yeah.
Gulla: The original was black and white. And I said, Ashley, well how come you have a colored version of that? He said, “Well, when I rubbed the print, it didn’t come out so well. So rather than throw it away, I colored it.”
So that unbelievably beautiful background is actually a rejected print. So then one thing led to another, this is our most successful book. This is called “The Women Who Caught the Babies” by Eloise Greenfield, and it’s about the midwifery. And when Eloise died about four years ago, she had an obit in the New York Times, and this was one of the two books that appeared in her obit, this and “Honey I Love,” which, I love, “Honey I Love.” So not only did we get, not only did we get into publishing, but because of my wife’s contacts, mainly with these very, very important writers and illustrators, we ended up getting a few things handed to us that frankly people don’t believe.
I met a woman in a parking lot after a meeting and whatever, and she was telling me, this was a parent teacher meeting, I was there for my grandson, and she told me that she’s into midwives and she’s a researcher at Duke and that’s what she does. And I said, “Oh, yeah, we have a book called ‘The Women Who Caught the Babies,’ by Eloise Greenfield.” And she looked at me like I was a freak. “What do you mean you published ‘The Women called the Babies?’” I said, yeah, we did. My wife and I, we know always, and it just didn’t make sense that me and my wife did this book. That was the best thing she ever saw. I mean, it didn’t make any sense to her. So on some days, it doesn’t make any sense to me either, because we’re a two-person operation.
Everything we do, everything else we do, we do through relationships with people, pay them. So we’re the new definition of the small press. So we’re not so much, now what we’re doing is memoirs. We decided we were going to branch into memoirs. So we had one last year, and we had one coming out next year from a professor at Duke who’s a psychobiologist, and this is his memoir. And he wanted to get, I approached him about getting it back in print. He came out of print and I asked him to write some new things for it. And he did. And just last week, I was at his house, I was recording him. He made a recording of two sections of the book, and when I put it on the QRC on the back, and you use these quick response codes you can create. And what I did was I put his voice files on the website, on our website, and the QRC, you get the chance to hear him read. I’ve discovered that people really like that. They liked it with children’s books. I don’t know if you can see there’s a QRC back here.
Wig: Yeah
Gulla: Yeah. That’s Eloise Greenfield reading her poems. When I read this in Good Reads, I read about, I’m telling you this about, because it’s the intersection of technology and publishing, right? So I don’t know where we got the idea from. I got the idea that I want to record these people. They’re humans. I want to know. So I went to Eloise’s house in DC. I recorded her reading the poems. She was practically blind, practically deaf. She was very old and made the recordings, put them out there. And then people on Good Reads, they buy the book and then they find out there’s a QRC in the back and they hear Eloise read and they break down in tears.
It’s such an emotional thing to hear. “I heard Eloise Greenfield read her own poems. Oh my God.” So I try to do the same thing with every book we published. I try to get them. So I went off to Durham last week and recorded John Staddon doing two sections of his book. I mean, I was emotionally wrecked by the end of the second reading because of the events of his life that happened since he published the first book. His wife died and he remarried his first wife. And I mean, it was like it was so moving, I was so happy that I recorded him doing this prologue.
Wig: Yeahs a whole nother layer to it when you can hear the author reading what they wrote, and there’s emotion to it. And yeah, that’s really—
Gulla: I think because of my background with computers, I’m more open to doing these kinds of things. They’re not that hard to do, but you need to be a bit tech savvy. So I’ve taken my ideas, my skills in the mainframe world and moved them to other places. So, that’s the publishing house. That’s Alazar Press.
Wig: All right. Well, wow, I love that twist in your career, and it’s great to hear the success you’ve had. And then we’re running out of time, but we got to talk about another book project you’re working on. I don’t know if it’s with Alazar Press.
Gulla: Yeah, we’re going to publish it through the press.
Wig: Ok, great. Tell me about that. I know that’s coming out next year,
Gulla: So I wanted to take the 24 articles that I’m writing for Tech Channel and include them in a book called “A Mainframe Reader.” And I felt like it, well, I had a couple of things in mind. One is for somebody who doesn’t know about mainframes, then it might be a reader, something you would do and supplement to something else. And if you were away from mainframes for a while, name came back to them, this would be a good book for you too, because the structure of the articles is, tell me about it and then tell me what’s new. That’s kind of what the A and the B articles do. And none of these IBM products are standing still. If you look at what CICS has done and IMS has done with embracing new technologies, those technologies open the door to the data more fully. And they also are appealing to a younger generation of programmers. So it kind of does two things at one time. I mean, young people don’t want to write COBOL. They just don’t want to write COBOL. So they’re looking for a language that does more for them and a language that has greater abstraction. So all these major subsystems from IBM are embracing, have embraced, not are embracing, they have embraced all these new technologies and this should appeal to a younger workforce.
Wig: Alright. Yeah.
Gulla: So you asked me a question, about the book. Right. Okay. So it’s a reader and it’s going to come out in June of next year. And it’s basically the 24 articles packaged and yeah, I’m looking forward to finishing up and getting it out there. And Andrew, I’m hoping to do something with you next year.
Wig: Yeah. I can’t wait for your next methodical proposal for a yearlong set of articles. We’ll see. That’s up to you. But I hope we can keep doing something.
Gulla: Yeah, I think I told you last week that I was thinking about modernization. That’s the topic I’ve come back to a number of times, been on a number of modernization projects. But anyway, I want to take a new look at it and see what I might do next year and propose it to you. And of course you don’t have to take it, but—
Wig: Well, no, I mean, yeah, we will cross that bridge when we come to it, but I mean, at first flush, it sounds awesome. Obviously I shouldn’t commit to everything on live, well not live, but on recording. But yeah, well, I’m excited to hear more about that for sure. Yeah, well that really does, we’re coming up in an hour here, so we really probably got to get out of here. But Joe, thanks so much. It’s been fun just chatting with you here and having you on Tech Channel, and it’s really a privilege just to be an outlet for your deep well of expertise. So thanks for joining us.
Gulla: You, Andrew. And I’ve already told you many times that I’m enjoying working with you, my editor. I have complete trust in what you do, and it’s just really great, it’s really great to work with you. I, I’ve had editors in the past that have kind of wanted to rewrite things, and it’s just not fun when that happens. And you always make what I do better, but you don’t get in the way of my message. I didn’t mean for that to sound snotty, but you know what I mean. You make it better, but you don’t try to rewrite what I’ve done. You have a lot of confidence in this, what I’m writing. So I appreciate, I’m—
Wig: Definitely, I’m not going to try to tweak your technical stuff for sure. So I definitely try to tread, tread lightly and let the voice of Dr. Joe Gulla shine. So yeah.
Gulla: Andrew, what you do, I think what you do, which is so much better than me, is you rewrite the title of the article. You routinely, you really take my dry article and heading and turn it into something that every single time you do that I say to myself, yeah, yeah, that’s right, that’s what I’m reading about.
Wig: There’s usually some little nugget in there that that’s like, okay, let’s turn that into
Gulla: You, the NetView one. I forget what you did with the NetView one, but it was sort of like a love affair. Yeah. I mean, you talked about, yep, I didn’t have the nerve to put that in the heading, but you were able to find a way to do that. So anyway, so I encourage you to keep doing that.
Wig: Okay. If I go too far afield with one of those headlines, let me know and we can dial it. Thanks. But yeah, as our audience probably knows, you can find Joe Gulla’s work techchannel.com, look for “z/OS and Friends.” He’s got something coming out twice a month. So that’s going to be a great place to just find a very comprehensive source of fundamental mainframe knowledge. And you can also, if you don’t want to miss any of it, you can just subscribe to techchannel.com/subscribe, and you’ll be all set up and won’t have to worry about the thing. And I think that’s it. So Joe, thanks again. Thanks everyone for tuning into IT Social Hour. We’ll see you next time. Bye.
And we’re back. Hey, just wanted to add this little addendum on here. We were talking with Joe after we were done recording this podcast session. The recording was still running, and he got out a letter that he had saved that he wanted to read during the podcast, didn’t get a chance to, but it’s from 1984 when he first started at IBM. It’s a personal letter from John Oppel, who is the chairman of IBM. And it was just kind of touching to hear Joe read it and see how much it meant to him. And it kind of ties into some things we were talking about. We’re talking about AI can help you learn a lot these days on the mainframe and all sorts of things, but there’s also still no substitute for good old fashioned human inspiration. And so just wanted to provide this little glimpse of Joe reading this note that has meant so much to him over the course of his career. So here you go.
Gulla:
Al. He says, “Dear Mr. Gala, welcome to IBM. I hope your career with us will be a rewarding one. We have been successful largely because of the dedication of our employees and their pride in being associated with the IBM company. Our success in the future as well as your own depends on your sharing in that tradition. I’m very pleased that you have joined us. Sincerely, John, R Oppel, chairman of the board. I can’t tell you how much that letter meant to me. It ignited me to be everything I am and to give my energies to IBM Corporation, trusting that IBM would continue to be successful, and that just being me and chasing the things that I thought were important and the right things for IBM would be good for me, would be the right thing for me and nothing had happened over my career diminished that. Amazing.
Wig: Yeah. That’s really something to hear from someone right at the top when you’re just starting your career.
Gulla: I don’t think they sent letters to everybody. He sent letters to maybe professional hires. I wasn’t hired right after college. I had six years of experience and I’d just gotten my MBA. So he may have known that. They got the letter. The thing about that letter that really means something to me is he said the IBM Company, he didn’t say the IBM Corporation and that sound like a small thing, but the old guard IBMers viewed the company as a bunch of people representing IBM. They were the company. And so that generation of people never called it the IBM Corporation. The code word was IBM Company and you are in the company. Nobody ever told me that. That’s just what I gleaned and that meant a lot to me. It really meant a lot to me to be in the IBM Company.
Wig: It’s good. Yeah, that sense of belonging and everybody’s got to have inspiration too, I guess. Right. So you’re talking about the, we can use the AI to learn a lot, probably. It’s not something to sneeze at, but also it’s like, I don’t know, I feel like you still, you got to have a human in there somewhere for that dose of inspiration.