On the Mainframe and a Changing, Challenging IT World
Joe Winchester of IBM Hursley and Reg Harbeck engage in a discussion and debate on ISPF, molecular biology and the cello
Reg Harbeck: Hi, I’m Reg Harbeck and today I’m here with Joe Winchester, who is an IBM mainframer with quite the personality, as you will hear over the coming minutes. He’s based near Hursley, I guess actually. Somewhere between London and Hursley. Well Joe, let’s just dig into this. First of all, how the heck did you end up on the mainframe?
Joe Winchester: That’s a really good question, Reg. I’m actually based in Hursley right now.
Reg: You are based in Hursley?
Joe: Right now, I am in the Hursley lab as it is sometimes called, the Hursley laboratory, which is also where I live. I have a marvelous commute to work. I haven’t got it pinned up on anywhere but I have the kind of Google maps, you know other online map tools are available that shows a dotted line between my front door and my office, and about 70% of that is on IBM property. I’m literally IBM’s neighbor.
Joe: You go outside a fence, past some trees, past a very pretty little church in the village of Hursley, and you’ll get to my house. Yeah, my home and work location.
Reg: Yeah, for those who are not deep mainframe nerds like myself, the reason I’m so excited about Hursley CICS,0r “C I C S” was my first product on the mainframe back in 1987, the first thing I worked with and CICS is maintained in Hursley. Now it was actually created in the United States, but then development was moved over to Hursley, England, pretty early on and I have so many things I love to talk about with CICS... [crosstalk]
Joe: It was about the 1970s or something like that. You implied it was last week.
Reg: 1968 I think, ’68 or ’69 CICS went—I guess it was ’69 it was officially GA’ed and yeah, I guess it must have been in the 1970s that it moved to England.
Joe: Yeah, yeah. When you said ’68, I thought you were going to like, outdate me. I thought I was going to come up with something about rounding up—yeah, so just for folks that are watching who think it just got moved over last week or something and you know MQ, WebSphere MQ as it is now called—got to get the brand right—the messaging, queuing platform and lots of other awesome software. Yeah, so it’s my home. Yeah. I’m glad you’re nerding out on Hursley and a lot of really cool products. z/OS Connect was created here. For folks who know about things that are over here—you know CICS transactions and Db2 stored procedures and all that cool stuff. Yeah, a lot very, very smart people in Hursley.
Reg: Oh, I just realized. I mean see as I look at my bucket list, one of the things on my bucket list is to get to Antarctica because it’s the only continent I can’t go to as part of my mainframe job because I haven’t found any mainframes in Antarctica. So I have to go there on my own dime. But one of the other places I really want to go to first—
Joe: You can find a Rockhopper penguin.
Reg: Yeah, yeah.
Joe: In joke there. Next one Rockhopper. Anyway.
Reg: The question is are they near Argentina because that’s the only part of Antarctica I’m likely to get to you know or are they in a different part of Antarctica, but anyway—
Joe: Argentina is South America. That’s not part of Antarctica, but it’s close.
Reg: No, but Argentina has a slice of Antarctica.
Joe: It does?
Reg: When you can take a cruise to Antarctica. You basically will go to the Argentina part of it, in all likelihood.
Joe: Which way? Does a ferry go there or something? I stand corrected.
Joe: I shouldn’t quarrel with you about geography. You are always the smartest man in the room if there’s a discussion of geography.
Reg: Well, I doubt that, but anyway—
Joe: Anyway, I yield.
Reg: I guess I’ll have to establish just how smart you are, because there’s a reason I’m interviewing you, and you and I have great conversations at SHARE. So let’s back up a little bit. How the heck did you end up on the mainframe?
Joe: That’s a good question. Sort of by accident, really.
Joe: I sometimes tell people I’m not a mainframer. I know it’s not politically correct, but I’m not sure if I’m allowed to say this. You know I might be an old bald white guy but so I look like a mainframer, but I’m not a mainframer, and you can always edit that out because I don’t want to insult some old bald white men. But so interestingly, my background depending upon how far we go back in time is actually genetics. I have a master’s degree in what’s called natural sciences from a very nice university. I went to University of Cambridge in England, and I ended up going there wanting to study physics, fell in love with molecular biology and genetics, which I still adore as two topics, but what I found really interesting was basically computers. When I was 14 years old, there was a type of computer called a Sinclair ZX81. It was like a home computer in the UK, and my grandfather bought one—
Reg: That had a flat keyboard, didn’t it? Like, there was a touch-sensitive keyboard.
Joe: Flat keyboard, 1 KB of memory. Just for folks you didn’t realize: 1 KB of memory. We’re not taking about a megabyte or 100K. We’re talking one kilobyte of memory. You plugged it into—
Reg: It went to screensaver and if you touched the wrong key, it would erase all your programs that were in memory.
Joe: Yeah, yeah, so that thing was difficult to program, and I learned to program it. My two brothers just wanted to play games on that computer—bless them—and you know they were gamers, but I was more interested in writing software. A lot of games when you got them were basically just listings. You just had to know a bunch of BASIC, or they would ship you the source code, right?
Joe: It also had a Zilog Z80 chip. So the really good games were built using Assembler, and I actually taught myself Z80 Assembler and BASIC, and I ended up writing game software. Then I ended up trying to do things where in the game software you could kind of punk it a little bit. You could get like infinite lives or go to the next level—basically cheat by reverse engineering it, and learning about what memory addresses you need to manipulate and things like that. I had fun doing that. I ended up buying my first motorbike selling game hacks to magazines and just weird stuff like that. This is before I went to university; I was probably like 15 or 16 at the time. When I was at university, I remember doing a project all about the structure of DNA, deoxyribonucleic acid. Most people are familiar with it now. Back at the time it was quite revolutionary, the sequence of DNA. This kind of zipper of, like, base pairs that, not only does it encode the template for creating proteins, which is the building block life. of everything basically made of proteins, but it’s also a self-replicating molecule that can just unzip and rezip as part of meiosis, cell division, and basically all organic life is based upon that. People were very interested in what did it look like and I was working on my thesis for my MA about what does it look like. I was writing programs in Fortran at the time and I remember seeing the very first Apple IIe computer—I think it was the very first Apple GUI—and I remember basically being able to code that and do sort of 3-D modeling where not only do you show what it looks like. Most people are familiar with the sequence of what it looks like unzipped, which is the crystalline form in solution. It folds itself inside in what’s called in vivo: inside the cell. It’s always just a big blob and that shape of it is very important. When you get mutations, a tiny mutation in the base pair can have a huge influence on the shape. When things are exposed or collapsed or concealed or revealed, that changes where things go back. Anyway, so I really geeked out on writing graphical software and being able to print it out and instantly have it printed. Beforehand I was dealing with like, dot matrix printers—all I could do was print dots and that took forever. I just remember literally thinking—back in 1980s, 1988 probably—that this is cool. This is going to be able to be used for a lot, graphical user interfaces, and more affordable computers are going to be something cool. So my parents have never forgiven me because I went to university studying to be a geneticist but I ended up switching into IT. I just thought this would be fun and I actually went to a job fair and joined a very small company. I said I know nothing about computer science, but I’m keen and that was it. I never really worked on mainframe computers. I always worked on GUIs, graphical user interfaces. I ended up writing VBXs for Visual Basic, I ended up writing DLLs in C, Windows programming. And then at some point in time I was proficient in a language called Smalltalk. Back in the 1990s and Smalltalk—there’s a bit of a lineage between Smalltalk’s OO concepts and C++’s kind of syntax into what became Java and the evolution of that. Then for various personal reasons, I needed a big employer with a healthcare plan, and a friend of mine who was working for IBM at the time, a director at IBM, he said when are you coming to see what we are doing? I never thought I’d work for IBM. So I joined IBM because I had skills in Smalltalk, and Smalltalk was used by IBM to build a software tool called VisualAge, VisualAge for Java, VisualAge for Basic, very successful tools like IDEs, integrated development environments. And I remember from there working on the tool that then became Eclipse, which is basically an open-source Java IDE that was extensible and was also very successful at building Java. It was a very successful project. It way outlived its sell by date because Java moved around between being a thing that basically made stuff animate on web pages towards being a desktop computing platform, neither of which really succeeded. They’ve both been replaced by better technology, but now it’s a very good server architecture with what’s now called MicroProfile—J2EE it was called at one point in time. So then I moved from there and then I was working in America at the time, and I wanted to come back to England. My wife and I wanted to come back to England, bring up our family in England, and I came back to Hursley. I was sitting around for about two years, looking outside my office thinking I should probably do something to do with the mainframe stuff that happened in Hursley. My expertise was all around building GUIs and then I said, can I actually come and do something interesting? They were like well, you don’t know anything about CICS or MQ or the mainframe, but you know quite a bit about stuff that looks flashy: laptops, GUIs. So I ended up building some GUIs, and they’re called the CICS Explorer and z/OS Explorer. Now I’m working on a project called Zowe, which fits very much into that swim lane which is where I’m most comfortable, which is basically building flashy stuff that makes things simple to use by having—by not through text interfaces is the best I can describe it. Anyway, here we go. Big answer to a simple question. Sorry.
Reg: So I’m going to ask you a question I never thought I would ask a mainframer. What’s your experience of ISPF?
Joe: The first time I saw ISPF, I thought you’ve got to be kidding me. I remember going to a training course. IBM wanted to train me, I went to this training course. I remember sitting in the room and the first thing we had to do was download an ISPF, a 3270 emulator, and it was just a mess. Nothing worked—absolutely nothing—and the first time I tried, the instructions were hit PF key 12. I’m looking at my keyboard. I’m like, where is the PF key 12? I think maybe if I did have 1-12 mapped at some point I have to hit a PF key like 19 or something and I couldn’t find it, and then I had to hit the CLEAR key and it was just—you’ve got to be kidding me. I remember, funnily enough, I was sitting on this course. I remember literally put my hand up and I said, is there not a GUI, a graphical user interface that I could use to just drive this with? I don’t remember when it was. 2000 and something? Anyway, whenever it was, perhaps 2004-2005, and the person said, “Why do we need that?” I ended up having a quarrel with my instructor—I was probably in my early 30s at the time—and he said GUIs are for wimps. I was like, what do you mean? He then went into this big thing. So wimp—it’s a kind of a slur, right? You know when people take an acronym and they weaponize it to be an adjective to insult. So he took the word Windows, Icon, Mouse, Pointer—which for me are fantastic superpowers. They’re power ups, right? I’m like, yeah. I’m all over that dude. He was like no, they’re for WIMPS—i.e., implying weakness if you have to use them. Then I said, but surely you want a graphical user interface? The instructor—for the entire rest of the time, he walked around, he was basically pretending that his feet were stuck to the carpet. He was saying oh Joe wants a GUI, I can’t move, it’s so gooey. I remember sitting there—thinking you may have to bleep this word out, Reg. I was thinking, you bastard, right? You’re making me look stupid in front of the other students, but the last laugh on me, because hey ho, I went back and built the CICS Explorer that is just celebrating its 15th anniversary or something right now. That is now the interface to CICS, right? I went back and I did what everybody should ever do if they receive a snide insult from somebody. Go internalize that, use your intellect, use your brain rather than your fists, or use your fingers on the keyboard to deal with that and basically create something and have the last laugh. So that’s what got me and that is basically—and I use ISPF if I have to, but I am so happy when I can join a call with a customer who says oh, I can’t do something. I can’t configure a key ring or I’ve got a hard problem with TLS or something like that. I’d be like, dude, this is my world. And they’re on their WebEx, other web conference tools like Zoom and Microsoft Teams and stuff like that, and they’ll show me something on SDSF, and do all these incredible commands, and I will say let me grab the screen. And I will show them something using Visual Studio code plugin or Zowe Explorer where we can just scroll around. We can do find and replace. We can grep things to, you know, use AWK, we can use some more advanced tools. And by the end of that call, I have got that system programmer downloading Visual Studio code. Yup, that’s it.
Reg: I’m going to get into the nerd weeds here, because on the one hand—
Joe: The nerd weeds?
Reg: Yeah, to get in the weeds on the nerdy side. On the one hand I see that you equate 3270 and ISPF, which is really an interesting way of thinking because that’s the kind of thinking that somebody who very much did not grow up on the mainframe would have. Because you talk to a mainframer and they don’t think of 3270 and ISPF in the same thought because-
Joe: Yeah, of course not. Yeah.
Reg: ISPF is one thing, but for you that’s been your experience. On the other hand, you said AWK, that great UNIX shell command that is a whole universe unto itself. You know it’s like—let me put it this way. I’ve written articles about regular expressions and wild cards and how ACF2, Top Secret, and RACF use them differently because they were all developed in the 1970s and early 80s before regular expressions were a thing. I mean they’d been developed but they weren’t widely rolled out. AWK on the other hand is sort of the ultimate manifestation of using regular expressions and anything else you can think of to do just amazingly powerful things.
Joe: You can pipe to GREP and AWK and pipe and pipe. By the time you’re on your sixth pipe to GREP, you are on fire. You are this multidimensional pangalactic super being, and I’m like, yes!
Reg: You might even be like Zaphod Beeblebrox and have two heads. Here’s a Gargle Blaster.
Joe: Oh yeah. Good shout out for the Hitchhiker’s Guide to the Galaxy. Douglas Adams, like it. Yeah.
Reg: So that said—although do you pronounce his name Zay-phod or Zaph-od?
Joe: Zay-phod Beeblebrox, who invented the Pan Galactic Gargle Blaster which—it’s a brick with lemon rind wrapped around it if I’m correct?
Reg: The Pan Galactic—the Gargle Blaster?
Joe: The Gargle Blaster. It’s similar to being hit around the head by brick wrapped in lemon rind.
Reg: Yes, but you actually do drink it, so unless you’re dissolving the brick, I think that’s supposed to be the net effect.
Joe: You do drink it but that’s the effect that you have, right?
Joe: It feels like that.
Reg: It sounds to me like using ISPF has a similar impact on you.
Joe: I just—okay, so I get it. I get ISPF. I get it when people had terminals and whatever and when the Jurassic period ended 65 and not 66 million years ago, because we have passed the millennium. So I’m constantly correcting people on that, using my genetics, nerding out. But so, ISPF edit: I have this argument the other day. I was presenting at the IDUG International Db2 user group in Prague last week, and I did this amazing thing. I was like, guys, what do you want to hear about? I want to hear about Visual Studio code. I was flying and I was dancing through the syslog and I was refactoring code, I was submitting jobs and running REXX. I was on fire, dude. I was on fire. Six engines: boom. We are just, okay, hold onto your seats, children, we’re going through the stratosphere—and up goes a hand. What about my ISPF macros? I’m like, okay. I’m not going to knock you over. I’m not going to kind of grandpa you in the room, but I like syntax highlighting YAML, and JSON and XML, right? Because for the mainframe to remain rev—for the mainframe to—tongue twister that. We got to make a limerick out of that one. I like Java languages like Java and Node and languages that are cross platform, and bringing in tool kits written by the Linux community, and they’re going to run the UNIX System Services, right? They’re not running in PDSes with fixed block 80 and stuff that’s basically just the grandchild of a bunch of punch cards. Okay, it has been prettified and stuff, but it’s basically just punch card technology. For that I need YAML and I need JSON, and ISPF doesn’t do that. I have to use function keys out the wazoo. And if I look at log files from Apache Tomcat that are like really wide and I talk to customers who… Then this person was telling me ISPF can highlight stuff and he said you can highlight but it doesn’t let you highlight JSON or YAML or XML or anything that’s come along this side of, I don’t know, like 1900-whenever. I think the closest I could find was I could say, imagine it’s C, because C comments starts with the pound sign. I just don’t like it. I’m sorry. I just don’t like it. I just don’t like ISPF.
Reg: Well, that’s interesting. I mean, I have a different perspective. You know for me, I love writing REXX macros for ISPF Edit, and I love the way ISPF Edit treats data in a columnar way that lets you hide rows and move stuff back and forth. The data can be edited really intelligently, but it’s a very particular type of data and I think that one of the essential differences that you’ve highlighted is the punch card attitude, that until you sort of drink the punch card Kool-Aid, the 80-column punch card Kool-Aid that is really the jet fuel of the mainframe, you’re so into the real world that it’s hard to get your brain into that perspective of thinking 80 columns wide.
Joe: But you’re just like a saxophone nerd in a jazz band, right? So I grew up playing the piano, and I’m a fairly good pianist. I grew up playing classical piano and I have a really nice keyboard at home I bought for my 50th birthday, and when I push the keys on it, it’s like, touch sensitive. It doesn’t have strings behind it. It’s not a proper upright piano, but it’s very good and it’s digitally sampled pianos it plays. I can say I want to sound like I’m in the Royal Albert Hall, click a button. It’s pretty good, actually. I can put headphones on, which is good for my neighbors when I’m practicing a difficult piece of music because it’s not noisy on them, but if I want to, I can hit a button and say make it sound like a saxophone, and it’s actually really, really pretty good. It’s like digitally sampled saxophone for all of the different pitches, and obviously I can play a jazz piece. It’s okay. A jazz musician would nerd out. It’s like no, I’m sorry, you’ve got to play the musical instrument. You’ve got to take a piece of—a cellist, right? I can make it sound like a cello and I can play cello and I can play trios where I lay down the piano tracks and then the cello and the violin and I can actually play trios, but they sound a bit weird because it’s obviously one of me playing-anyway but there will be somebody in the room. There will be the ISPF. Now the cellist, what do they do? They take a piece of wood with horsehair and string it and bow it and make it sound beautiful. Respect. That’s going to take you 10-20 years to master, and you can play cello music. I’m like, play the piano, hit a button that says play cello. So what my mission on the mainframe is to say, you can hit a button in Visual Studio code—I don’t care what it is, IntelliJ, you know Jenkins, whatever—and you can say, “See that thing that can talk to every other computer on the planet, that everybody coming out of school knows how to operate it, hit a button and say, play mainframe.” There will also be the ISPF macro people saying oh you just don’t know how to use PF keys, and here’s how to remap the CLEAR key and you’ve just got the wrong emulator saying—you know, the Luddites, throwing their shoes at Visual Studio code—and I’ll be like, you know what? Come on. Step this way, dude. 100,000 downloads on Visual—sorry. This is starting into a bit of sort of aggressive posture I’m taking here.
Reg: Oh cool. This is fun. This is first time I’ve heard the word Luddite refer to somebody who favors a technology. It’s just the technology happens to be the established legacy technology.
Joe: The first time? Oh, my goodness. Dude, you ought to come to some of my—actually no, I have to be careful. I haven’t yet at a SHARE conference gone full-on attack dog. I do realize there are some boundaries, but there’s an element of me that wants to. Steven Perva and I have talked about should we do the talk that basically says look, we need to just leave this baggage behind. Lock it in a museum case, look at it, respect for what it did, but let’s skate towards the future. I’m on fire. Woo!
Reg: This is really important, because on the one hand—I mean if you want to find somebody who’s—well, we have a word in Canada, keener. It’s not a word anybody else who speaks in English-see, if you go to the United States or the UK and say keener, it means two things. It either means it’s somebody who is more keen or something is more keen, such as keening of a knife, or somebody who makes a high-pitched whining sound, they’re keening. But in Canada a keener is the person who does the extra work when they’re not going to get extra credit. It’s like you’re already guaranteed 100% and you still go overboard and do the extra work. They’re the person who does extra stuff just for the sheer enjoyment of doing extra stuff, and so I’m a mainframe keener. I’m a mainframe legacy keener. I am just all about this, and yet part of that is recognizing that the mainframe absolutely has a future, but it doesn’t mean it has a future I might choose for it as somebody who is really sold on the old interfaces. The future of the mainframe may very well be—
Joe: Dude, you always welcome. You can always bring—
Reg: You know, putting those interfaces into a museum.
Joe: You can always bring your cello and ISPF saxophone and you can always jam with the cool kids, but I’m going to bring my keyboard. Sorry, anyway.
Reg: And I love how it comes back to a keyboard either way. You’re going on and on and on about the mouse, and yet it all comes to back to a keyboard. Before you know it, we’ll be talking about cards.
Joe: Oh touché. Touché. The wimp. The wimp in the room, the wimp that came with the keyboard. But no, I’m actually talking about my piano keyboard, but goodness. I mean we’re just looking at the screen and blinking and you know it’s going to be—I mean Siri. I mean gosh, have you ever had these conversations with people? Well like, you’re next to somebody—my wife does. She talks to her sister and she gets Siri to say a text that gets sent to her sister, who then uses Siri to say a text that goes back. And I’m like, you can just push a button.
Reg: I can’t do that. I can’t do that. I’m a compulsive speller.
Joe: You can just talk to her. You are?
Reg: I desperately need to spell words properly. It’s a probably a personality flaw, but I have literally had to learn to forgive my friends for sending me texts that Siri put the wrong words in and I have to read them out loud to figure out what they were saying.
Joe: Well, they use you’re and your correctly.
Reg: But it’s just—they’ll have literally three words that sound exactly the same as what they are trying to say but mean completely different things, and you have to say those three words together to hear what it is they are trying to communicate because that’s what Siri heard.
Joe: Okay, so I’m not a bigot. I understand that ISPF has a place, but I also recognize the fact that—so I was at Imperial College in London recently at an IBM Z Datathon, and one of the big successes that Zowe has had is the fact that the professors at universities and people teaching mainframe skills have it as a part of a curriculum, right? It’s a more welcoming and pleasing interface. At some point in time—I was going to say when the shit hits the fan, but I’m not because I couldn’t swear on this call—you probably do need to crank up a 3270 emulator and go and edit a PARMLIB or a PROCLIB using ISPF. I get that. When the car breaks down by the side of the road and you can’t patch it, you need to call for the repair. They’ve got a better kit available, right? They’re more trained. I get that, but that shouldn’t be your day one experience. Is it you have to get into a time warp your way back and understand how to allocate a PDS fixed block, select what kind of tape drive—I mean that stuff belongs in—yeah. Anyway, so I just see ISPF as part of that whole kind of like, headwind, and there’s also massive bigotry in the mainframe. I mean it’s not a diverse enough community. When I go to conferences, I see change happening and sometimes change is forced, but I’m actually a huge proponent of forcing change. Whenever I do submissions for conferences, I love putting more diverse people, diverse backgrounds, whether in all sense of the word—you know, gender, race, every possible sense that that word can mean, and I’m not discriminating against older people as well. I recognize that also sometimes it’s a diverse backdrop but the mainframe world is still a shrinking world of introverts who think that they and they alone are waiting to save humanity because people will come to them, and that’s not the case if you go to a non-mainframe conference. There’s a lot of very smart people out there, and we need to hybridize and blend with some of that intellect at all levels, right?
Reg: Joe, I knew we were going to go over 15 minutes. We’re actually at about 27 ½ minutes now and—
Joe: Ah. Just delete all the swearing I did. I’m sorry for—
Reg: The first thing I wanted to say—
Joe: And saxophonist stuff.
Reg: But that said, it’s interesting you say sax-oph-on-ists rather than saxophon-ists. I’m not sure which I would say, but anyway, here’s the thing. I want the call to action for this to be, go see Joe. Joe is presenting at every different mainframe conference you can think of and he’s getting five stars out of five ratings for everything he does.
Joe: Oh, I have had some terrible ratings.
Reg: You can tell the guy is on fire, and so I’m happy to know him because just having a conversation with this dude is fun, and his presentations are also really fun. So that said, with the little time that we theoretically have left in this call, what I’m going to do is what I typically do. I’m going to ask you to paint a picture of the future of the mainframe. I like to use the term that I have may have been the one to coin, called “prodiction”—and they say that the best way to predict the future is to choose it, to proactively choose that future—and then say here is what I’m going to do or here is what is going to happen. So as you think about the future of the mainframe and what you intend to make it to be, what does the future of the mainframe look like—you know, 5, 50, 500 years from now?
Joe: So I’m going to sound like a IBM salesperson here: It’s a very secure mainframe computing platform. Everything is always a blend. In genetics for example you have a blend between single-celled organisms, which is basically bacteria and yeast, and they can evolve very, very quickly into huge multi-cellular organisms that tend to suffer when the climate changes. You tend to get mass extinctions and unfortunately, we’re one of the multi-cellular ones, right? So we will be outlived by bacteria and yeast, but we can have more fun while we’re here, so let’s have some fun while we’re here. So I do think that as the world gets into more perilous areas to do with, you know, the AI algorithms and the truth you know who really—is this really genuine? Where’s the ledger that this came from? All of the things where secure computing, people’s personal data being used for impersonation and fraud and blackmail and all this. I do want the mainframe computing values—which is what-seven, eight 9s up time?—to always remain true. The Db2s, the transaction processors of this world, I think they’re fundamentally squarely there. What I’d like to see is the platform shouldn’t have such a steep learning curve to get to it, right? It should just be easier for—I was at Imperial College London just a few days ago. It should be easier for people to become proficient, faster, to be able to basically use the platform for what it’s very good at doing. It should be open, simple, and familiar. Hey ho, those are the three tag lines for the Zowe project that I happen to work on—but that’s what it should be. So when I go to a mainframe conference, I’d love you to come to one of my sessions and give me five stars, which is great because Reg, Reg ordered you to! So you have to now.
Reg: There you go. That’s not your call to action!
Joe: I’ll be playing saxophone music on my keyboard just to annoy you, Reg, because it will be like squeaky and beepy—but I want to see a more diverse community. I want to see more debate. I just want to see more healthy debate, and I don’t want to see people talking about the past. So when you ask me about ISPF, you just really grated against the thing—it’s not that. We shouldn’t be talking about the past apart from to basically realize mistakes that others made, possibly ourselves, but we are not that person. We are not the sum of our worst mistakes that we’ve made in life, right? We should be completely looking out the front window. So I don’t even like mainframe conferences. I want to have IT conferences. I want to get other people in the room. I want to get people who are building cool things with mobile phones, who are building cool things with AI, who are solving difficult problems to do with predicting the weather—you know, when to harvest the crop, when to plant a crop, what strain to select—very complex things that have to be solved for the benefit of humanity, and the mainframe is a computing platform that excels at doing all of that stuff. Amen.
Reg: Well, I’m not deliberately taking the last word on this, but it seems like I might because I have to sort of finish up.
Joe: Any sentence that says I’m not deliberately is like, I think we should be insulting paths. But go ahead.
Reg: Okay, so let me do this. Before I do my finish-up blurb, let me just say that you and I don’t think the same way about ISPF. Because on the one hand, I absolutely agree with you that things like Zowe are the future of the mainframe, but I don’t think that ISPF is exclusively the past of the mainframe. I think that it’s a very special environment that does things—for me, I mean 3.4 and REXX macros are always going to be a very essential part—
Joe: I’ve had this discussion before. Editor of choice, freedom, everybody should be able to be whoever they are. I will welcome everybody. I will welcome any debate with anybody. There are a few people I won’t debate with who have real exceptions. There’s a few things that I think are just like no, you should be locked away and humanity should just stick you in a dark room. But for the most part—and so I’m happy to debate ISPF—but if you go into an organization, what I don’t want you to be doing is using proprietary source code managers, proprietary tools—you know, weird, strange bits of Assembler that only you understand, right? So DevOps is all about whatever editor you use—I don’t care, configuration as code, this is incredible north star about the fact that they touch live systems, touch the definition of a system and that’s already done by the distributed community—which by the way where did the word distributed come from? Mainframers talk about the distributed community. I worked in Windows and Linux for a year. I never thought I was part of a distributed community. I get it, because you’re saying we’re centralized, and they’re distributed and somehow—it’s just like all these weird words. It’s like anyway, I’ve stolen your last word from you.
Reg: That’s okay, good. Not only could we, but we will keep talking. But, I think, probably, having blasted right through 30 minutes, I think this is the longest one of these podcasts that I’ve done yet—
Joe: I’m so sorry.
Reg: But that said, let me—
Joe: Don’t worry. Nobody is listening to it. Nobody is listening, Reg. They’ve already left.
Reg: I think this will be one of my most listened to, honestly. But that said, I intend to be back with another podcast next month. In the meantime, check out the other content on TechChannel. You could also subscribe to the weekly newsletters, webinars, e-books, Solutions Directory and more on the subscription page. I’m Reg Harbeck.
About the author
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.
See more by Reg Harbeck