October 15, 2020

Article at IBM Systems Magazine

View original

iTalk With Mike Pavlak

In this episode of iTalk With Tuohy, Mike Pavlak talks about the progress of open source on IBM i and some of the major game changers.

Paul Tuohy: Hi everybody and welcome to another iTalk with Tuohy. My guest today is the right honorable Michael Pavlak. Mike, how are you?
Mike Pavlak: Hi. I'm well, Paul. How are you?
Paul: I'm good. In these strange times, I'm good. So when we were chatting just beforehand, Mike, isn't it—I'm—I'm still sort of thinking maybe I've missed on in between but the last time I did an iTalk with you was 2016.
Mike: That sounds about right.
Paul: Yeah. It does not seem that long, but obviously a lot has happened since then. So back in 2016 Mike if I remember correctly, you were—you were working with Zend. I know you've worked with another company in between and you're now with—working with Profound. Correct?
Mike: I am. I am. It's been an interesting ride getting here but I'm—I'm definitely enjoying the ride with Profound. I'm over there as—as you know technical presales so I get to do—I'm learning a lot about the products, which are just amazing, and building demos and examples and I'm working with customers, which is really what I love doing. As you and I have kind of met over the years at various Summit conferences—
Paul: Yeah.
Mike: It's all about—it's all about working with people. I just love that stuff.
Paul: So Mike, I think it's—I think it's fair to say that you are probably at least one of the earliest open source people on IBM i, if not the original open source person on IBM i.
Mike: Definitely not the original but—but—but yeah, I was one of the handful.
Paul: Okay and you were def-okay-okay look. Maybe not one of the originals but definitely one of the earliest evangelists of it, as being out there speaking to people and convincing—convincing people myself and Jon [Paris] and Susan [Gantner] and other people that this open source was a good thing and not a bad and terrible thing.
Mike: And—and—and convincing Steve Will. I mean because I—I was—I was on the advisory council which—I know you're on the advisory council in Europe and I was on the advisory council in the states back in 2004-2005 and you know—we had made the transition from Mike Smith to Steve Will as software chief architect and—and Steve Will was like—he needed to understand this open source thing. So yeah, it's been—it's been fun being an advocate in this community.
Paul: Okay well so open source then: I mean since 2016 fair to say it hasn't really changed that much Mike?
Mike: Yeah, right [laughs].
Paul: Sorry.
Mike: Much like politics in the United States. Nothing new is going on here [laughs]. 2016 rings a bell. Anyway—let's not go there.
Paul: Okay.
Mike: We want to keep this conversation light. The—yeah I mean the—the—the—at a macro level in the IBM i space, Tim Rowe's team had deployed open source via the—a licensed program. It was 5733-OPS and that was kind of you know burning bright back then. And one of the things that you know I was certainly kind of you know herding the cats on is some issues within the open-source space—you know it was kind of like Christopher Walken and the cowbell. We were going back to Steve Will saying "we need more open source. We can't get it fast enough. I need more." And Steve Will you know had just recently I think or not too long after that minted Jesse Gorzinski as the business architect for open source, and one of Jesse's first big projects was "okay, Jesse." You know Steve Will told him: "go figure out a way to distribute open source faster than we're doing it." And that I think began the work that he and Kevin Adler and I'm sure several other people in IBM Rochester you know brought the RPM distribution model—early open source package management features that are available through ACS and all the YUM updates and that sort of stuff that's—that's available today in the PASE environment.
Paul: So—so I mean obviously the thing of the distribution has been a big thing, because I mean that was one of the killers I remember in the early days when things like Python and then Node and that were becoming available—having to do that through licensed product was a little of—well more than a bit of a pain, okay? And as you know Mike I only have dabbled with open source; I'm far from being an expert in any of it. But so apart from that, the distribution side of it, is there any other one big thing that you look at in the open source world and you sort of say yeah, this was another sort of game changer for open source on IBM i?
Mike: Absolutely. Absolutely and you know it's—the open source itself is a game changer, and then the distribution becomes a game changer because now we're able to adopt it faster. And one last shameless plug for the work the guys in Rochester are doing: is now they're—now they're starting to transition people towards ODBC as a connecting mechanism. You know there's—when I started with PHP back in the day, we had a native Db2 common interface model that was—was originally built for the UDB folks but you know Tony Carnes had ported it for IBM i and it was able to talk to Db2 directly. Jesse and Kevin have kind of remodeled that world saying "hey let's use ODBC" and Tim Rowe is like "yeah, sure" and part of the reason for that isn't necessarily because ODBC is so much better or faster than Db2, although arguably it's got some pretty good competitive numbers. It really has to do with when an open source product is brought to the IBM i platform, nine times out of ten it already has an ODBC driver interface for it, so—
Paul: Yup.
Mike: Now the guys in Rochester don't have to you know bring out their tools and shoehorn a Db2 adapter into the new open source thing and have to maintain that as well. So you know it's all about—you know and again I keep going back to this conversation that we had with Steve Will and Jesse Gorzinski about adoption. You know we—you know these guys are focused on bringing the tools to the platform and giving customers a choice, and that's—that's the beautiful thing that we're all after.
Paul: Yeah, so—so—so obviously Mike you know we—okay maybe 2016 the last time we did an iTalk but we've met many, many, many, many times in between, and obviously at the Summit where you—and thank you so much—keep coming back to talk at the Summit. Back in 2016 you would have been talking about PHP, and now you talk a lot about Python—
Mike: Yes.
Paul: Right?
Mike: Yes I do.
Paul: So-so-so-so have you now thrown PHP aside and you're now a Python convert?
Mike: Umm would that-would that be so—so—such a terrible thing [laughs]? The—you know what? The reality is that I really had no intention of like you know learning or working with Python, but Jesse Gorzinski called me up one day—you know I think it was like a week before a COMMON conference—and said hey, I you know I'm having trouble staffing a couple of these sessions. Would you be willing to take over a couple of sessions you know for the IBM i folks? And without batting an eyelash I said sure, no problem, and you know that was—this was all happening via direct message on Twitter, I think. And after I hit the enter key, I realized oh my God, what did I just commit myself to? And I fired a message back what did I—what did I just sign up for, Jesse? He said "oh we've got a couple of Python sessions." So I you know I literally went out, downloaded Python at my work station, started playing around with it, got a simple—a couple of simple examples working and understood enough of Python where I responded back just "yeah, no problem. I can do that" [laughs]. But what Jesse did was is he really—he kind of you know lit the fire for me and I you know have been doing PHP for a long time and I was really looking for something else to kind of you know—I wanted to you know, kind of dip my beak so to speak. And sometimes—as we all know in the IBM i community or as workers, as worker bees—you know until there's a reason, until there's a sense of urgency, we tend not to stray too far from our—from our roots, and Jesse created that fire for me. And it got me motivated and I've been obviously having a great time ever since.
Paul: Yeah so okay but you still didn't answer the question, Mike: Are you now going to throw—cast PHP aside and are you a Python convert?
Mike: No. I am—I am—I'm definitely a Python advocate as well as being a PHP advocate and I'll probably continue to advocate more and more open source projects as we go along here. I'm definitely still a big fan of PHP. I still teach it at the local community college when the classes actually run but—but yeah. It's—you know I think PHP is still—it's getting—you know there's a joke that Java is the new COBOL, right? There's so much Java code out there that you know is not going to go anytime soon. I think the same is very, very true about PHP. There's—
Paul: Yeah.
Mike: You know depending on whose statistics you look at anywhere between 50%-80% of the internet runs on PHP. It's going to be a long time before PHP goes away anytime soon and the community is still supporting and enhancing it—in fact PHP 8 is due out next year. So it's going to be pretty exciting times.
Paul: Yeah. So—so—and by the way you passed the test, Mike, because that of course was a trick question to see how you would answer it. Because of course one of the big things about open source is it's supposed to be about the right tool for the right job [laughs].
Mike: Yes. Yes.
Paul: So correct answer, sir. Well done.
Mike: Thank you.
Paul: Your certificate is in the post [laughs].
Mike: Oh lovely. All right. Oh oohh international yet. Okay, great.
Paul: So you—so actually something you touched on there, Mike, when you talked about teaching in a community college—and as you said, when your courses run. Again we were chatting about this beforehand and obviously because of—well due to Covid and that and some of these classes now not running, you found yourself with some extra time on your hands.
Mike: I did and you know what they say? Idle hands be the devil's workplace and the devil's had his way with me. I—I—and I don't know—I don't know how much you know about me personally, Paul, but I do—I was diagnosed very, very late in life with ADHD. And you know as I start to reflect on you know earlier challenges in my life, we start to learn things. And of course all my siblings were like "aha, yeah, sure. Okay, we believe it. Yup." So—so I get bored easily and so, like we were saying before—you know I have two classes at the local community college that were on deck and neither one of them ran, primarily because they're typically advertised via you know posters in the school and without students walking about the school the posters don't really advertise to anybody so—
Paul: Right.
Mike: So my classes didn't run this semester because of low enrollment—and so we're going to do some things to hopefully fix that up for next semester, certainly for next fall. So that leaves me with some free time on my hands which I've been channeling into an effort that I've been—I've been thinking about for a long time and that is I'm establishing a YouTube channel. And so you know by the time this—this message gets out there, hopefully I've got the channel up and running and we've got some recordings out there. And it's—it's going to be called i Mid Range, and my goal is to put together short—short videos—you know 10-12 minutes each is my target—just talking about things about IBM i. And it's really—you know I'm going to have several different kind of angles that I'm going after here. You know one for sure is the newbie, right? So people who don't know what an IBM i is, you know here's some short YouTube videos to kind of get you up to speed, some of the basic concepts and some of the ideas and stuff like that. And then you know go down the road with some—you know some veteran concepts, certainly some open source tract stuff and you know just basically I want to share you know what I know and what I think and—and what I believe, in certain cases, and see if anybody likes it.
Paul: So are you going to primarily aim this at well sort of the non-IBM i people, sort you know kids at college looking around?
Mike: I'm—I'm definitely going to have—I'm still trying to decide how to architect the channels and stuff like that, but the channel is going to be i Mid Range and I'm certainly going to have a tract or a series dedicated for that, yeah—
Paul: Yeah.
Mike: Without question because my—my—and also I mean part of what I want to do is you know trying to attract kids out of college is one thing right, but there's also people in existing shops, and this is what we find especially, as you and I know, with RPG. RPG is you know—I joke around that RPG programmers aren't programmers, right? RPG programmers are really business analysts who write code. And of course there's—there's—it's a varying spectrum out there, but the overwhelming majority of people out there who do RPG—see RPG is a great language because it's fairly easy to adopt. It's not—it's not terribly complex like a Java with all the different classes and inheritance and all that crazy stuff. I think RPG lets you focus in on the business, and so what ends up happening is people who have a strong business sense could end up drifting into the language as well. I mean I know more people in the RPG world who came in from accounting than came in through computer science, right?
Paul: Yeah.
Mike: And so you know—you know—creating that runway for folks to say you know I work at a company that has a IBM i. I wonder if that's something I might want to pursue—and you know give everybody an on-ramp.
Paul: Yup. Yeah, no, well that is—I'm delighted to hear that you're doing it, Mike, and I wish you the best of luck with it.
Mike: Thank you. I'm—I'm cautiously optimistic. My goal is not to get rich, although I'll take that [laughs]—but my goal is—as you know, my goal is beer money. If I can get some beer money out of this, I'm thrilled [laughs].
Paul: Yeah, I've tried that, Mike. People tell me it's cheaper to pay me a salary than it is to give me beer money [laughs].
Mike: Coming from the man who taught me the proper way to drink a pint of Guinness—
Paul: Okay.
Mike: And I am forever in your debt for that.
Paul: And I will never let you forget it, either [laughs]. So—so listen Mike, I'm—I think actually maybe talking about Guinness and beer money is a good way to leave this conversation. So thanks again and I will try and make sure that it's not another four years before we do this again.
Mike: That'd be fantastic, Paul. Thank you.
Paul: Okay, everyone: That's it for this iTalk. Tune in again for the next one. Bye for now.

Subscribe to IBM Systems magazine, Power-Systems