Skip to main content

Mike Pavlak on the Mainstreaming of Open-Source Technology

SoundCloud Widget

Charlie Guarino: Hi everybody. This is Charlie Guarino. Welcome to another edition of TechTalk SMB. As always, I am always thrilled to have these great guests and this month is no exception. I’m sitting here today with Mike Pavlak who I’ve known for many, many years, a very good friend. Mike has been working with IBM midrange solutions since 1992, and even before that working on mainframes. For plenty of years, he’s been doing RPG, CL, PHP, and was even an IT director for Tripp Lite in Chicago. Today Mike works for Perforce and he is a solutions engineer where he helps companies around the world approach the challenges of modernization and open-source technologies. In addition to that, Mike is also a volunteer in several capacities with COMMON. And finally, he is also a teacher at Moraine Valley Community College in suburban Chicago where he teaches classes on PHP and Python. So Mike, welcome to our podcast today.
 
Mike Pavlak: Thank you, Charlie. I’m glad I was able to fit you into my very busy schedule [laughs].
 
Charlie: Yes, I know. Well, I think I took a lot of time just on your bio, so thank you for coming. We’re out of time. Thank you. See you next time.
 
Mike: My pleasure.
 
Charlie: Thanks for showing up. Mike, we were talking before this—before we actually hit the record button here—and one thing that you of course are famous in the community for is your PHP and Python and open-source technology in general, and that’s where I want to focus this conversation on. Because this is a topic and we’ve spoken about this before—not just you and I, but many experts. This is one of the pet discussions because it’s just so popular and so pervasive in our community, but I think you bring some unique ideas and thoughts to this discussion, which is why I invited you on today. Thank you again for coming.
 
Mike: My pleasure.
 
Charlie: Sure. One of the things that we discussed and one of the things you brought my attention to Mike, which I wasn’t even aware of, is the annual report. Apparently Red Hat does come out with an annual report entitled “The State of Enterprise Open-Source” and there’s some interesting topics in there. So we’re not going to focus necessarily on that report but some interesting thoughts came out of it. But what can you tell me about the report in general?
 
Mike: Well, I’ve been watching this report for a couple of years. I think you know it first came to my attention when I was kind of following when Stephanie—forgetting her last name.
 
Charlie: Chiras.
 
Mike: Yeah, Stephanie Chiras. She left IBM to go to work at Red Hat and that’s when I noticed this report, and what’s interesting is that you know there’s a lot of anecdotal stories about what’s going on with open-source, but you know Red Hat actually took the effort to do some actual research and put some math and put some statistics behind it. A lot of the conversations I have with people even today with regards with IBM i and that kind of stuff, there’s a lot of people out there who still don’t trust open-source. You know it’s like oh, is it fad? It’s not safe. There are all the myths out there. So when I can grab a hold of a report like this—you know for example with Red Hat’s staunch credibility and of course Red Hat being owned by IBM now, I can bring some of that stuff like I said, you know what? 80% of enterprises out there are using open-source and you know so many, better than 50% have stopped spending money on proprietary software, and you know Software as a Service is growing. Those companies behind the scenes are using open-source and they have to use open-source because they couldn’t afford to buy proprietary software for ten million users and stuff like that. So you know the reality is that open-source is a real player and reports like this Red Hat report become very, very credible resources that I can share with people. It’s kind of a leave behind and they can kind of start to draw their own conclusions based on you know, some of the research data.
 
Charlie: You mentioned that there’s a lot of credibility in this report. Does that suggest that in the open-source community that there may be less credibility perhaps in the industry using open-source, maybe because it’s not necessarily owned by any one person?

Mike: I mean absolutely. But I think if we’ve learned anything over the last several years, or the last four or five years, is that credibility has way, way come under question. So I think having a healthy skepticism about everything is a good thing. There’s a Nobel laureate by the name of Paul Romer—he’s basically is a mathematics research statistician kind of guy and he used to do a lot of his research with commercial software. And the problem he ran into was that you know as a researcher, obviously you spend a lot of time researching, but the only thing that matters is getting published. Well you can’t get published unless you are peer reviewed and what Paul had discovered was that his expensive, you know statistical software—the only people who could peer review him were other people who had the same expensive software. So what he did a little research on this and he discovered that, you know by switching over Python and switching over to a Postgres database, he was able to get all of the things that he needed that he didn’t already have from the commercial software, and now the price of admission was zero so now he could actually put his research out there. He could put his data sets out there in the world and researchers could—they had no excuse. I mean if you don’t have time, you don’t have time, but certainly cost was no longer an issue and Paul’s observation was that you know, he anticipates. I’m kind of abridging his statements here but they’re easy to find on the net. You know he basically is saying is that he has a lot more faith in open-source than he has in proprietary software because with open-source software, you can see the code. You can look at the code and you can see what’s going on. With proprietary software you have no idea what’s going on inside that code. You know obviously if it works, it works, but you know what is it doing and do you care? At the end of the day I think you should be skeptical about anything that you’re using from a technical perspective, and it should be a healthy skepticism. It shouldn’t be prohibiting you from embracing it, but you should definitely do your homework.
 
Charlie: You know Mike, it’s funny you say all of that because there’s a great corollary right in the first page of this report by the president and CEO of Red Hat, Paul Cormier, and the quote in here is quite fascinating to me. I’ll just read it out loud. I’ll try to read it out loud. I’ll abridge it anyway.
 
Mike: Yes.
 
Charlie: There’s so much to digest in this report but I hope what you take away is while the open-source development model may have started in the playground of developers, hackers, and visionaries decades ago, we’ve moved so far past that. It’s now a mainstream part of commercial software development and the engine for consistent innovation from the server room to the public cloud to the edge and beyond, so I think that only validates what you’re saying. So I mean—you completely subscribe to that. You know what he’s saying right there?

Mike: Oh yeah and in fact in you know 2002 when I was first getting exposed to open-source, I was that passionate about it. I had that much of a belief system about it. I still remember the good old days when people couldn’t trust Linux. Oh my gosh. How could you run your business on an operating system that everybody can see the source code for? You know we’ve come a long way and if you know, arguably depending on whose statistics you look at, if you look at the z, System z platform, the only—I shouldn’t say the only—one of the main reasons why that platform is still relevant is because they embraced Linux at the right time. They made that a real-world class player on the platform and so—because they saw general attrition away from the monolithic programming environment. Now does that mean the mainframe is dead? Hardly. That’s like saying the IBM i is dead. No it’s not but the reality is that people started moving away from the monoliths and moving more towards modular and interchangeable components and that sort of thing. And what IBM did with the z is they said okay, you know they put Linux up there. They have the credibility of a world class player and now those components that may have off the mainframe, the z side, they just move them over to the Linux side, but they kept them on the hardware. They kept them on the same environment. They kept them in the same infrastructure. Now when IBM i tried to do that, when the Power Systems brand tried to do that, I think they were a little late at the game. It wasn’t resonating as much in that kind of space and there’s a lot of other, you know discussions about why or why not Linux is exciting and that kind of stuff. I don’t want to bite on that right now but I think you know, IBM embracing Linux on the mainframe brand gave open-source a tremendous amount of credibility.
 
Charlie: You what, Mike? You are well-qualified to make that kind of statement because you’re in the know.
 
Mike: [Laughs]
 
Charlie: No but I’m serious about this. You’re in the know, but sadly I think many people are just sticking to their old perceptions and that’s where they are. They hear mainframe, or even IBM i or dare I say AS/400, and you know where does their mind go immediately? They think about old legacy systems, and you and I will speak until we’re blue in the face correctly so how open these systems are, but it’s a perception issue here that still exists. I was just looking at the IBM z16 mainframe website actually, and that machine is amazing. First of all, the mainframe—that whole platform is amazing what it’s capable of doing, whether it’s going into AI and now even quantum is in the discussion there, but it’s an interesting discussion. How do we break that perception of—you know is open-source—is that what’s going to bring us there to ultimately break that perception of these larger systems?

Mike: You know I will share with you something that I was taught by my CFO at Tripp Lite who I have a lot of respect for, God rest his soul. But he—you know as I was being brought in as an IT director and he knew my personality. I’m kind of you know, I don’t want to say a wingnut, but energetic. You know I want to explore new things. I want to charge ahead, and he put the reins on me. He basically said here’s what I want you to do, Mike. He says you’re going to walk around here and you’re going to see all kinds of opportunities, all the things you want to change. He said promise you’ll do this. Help the people who want help and—you know so even though I might see something that just absolutely needs to be remedied or is an issue here, an issue there. He said help the people who want help. Build your credibility with your community by working with the people and delivering for the people who actually ask and/or need the help. And so you know, you’re right, Charlie. You know you and I are going to run into people who are still coding in RPG III, who are still using SEU because they’re faster in SEU than they are in RDi, and you and I both know the myths of that, but you can’t convince them. But there’s someone standing right next to them who just installed RDi and they really want to get going with it. Who are you going to help? Who are you going to help, Charlie? You’re going to help the people who want help and you know I’m out there just trying to do the sales pitch, trying to ease people down the road—you know quoting the statistics like from this report and from others and stuff like that, and the people who come across the field, I grab them by the arm and I drag them. I say let’s go [laughs].
 
Charlie: Interesting. You know one of the things that we talked about prior to this podcast Mike, you made an interesting point in our pre-discussion and it was about trends of open-source, and more specifically about how there seems to be a melding of sorts of different languages and how they’re being used. For example, one thing that you and I’ll just repeat—paraphrasing because I don’t remember—it was said so eloquently and I probably won’t even do it justice here, but I’ll just repeat it as I remember it. The trend moving forward is not just one language, but a suite of them. And in your notes that I copied here, we talk about how you identified three particular technologies—Python, Node, and PHP—and each of those you had a different use case for them, for example.
 
Mike: Sure.
 
Charlie: And I’ll just give you a quick—you know, refresh your memory: Python for DevOps for systems, Node for stand-alone horizontal scaling, and PHP for enterprise web. But so is that where we should strive to go? Should we strive to really be using a suite of them? I know a topic you love to bring up all the time—or at least a theme you bring up—is well, it’s religion. You know I’m on one language and I can make this thing do anything so how can you—isn’t there an issue with those two? Aren’t they counter to each other with religion in that context or using a suite of languages?

Mike: Yes and you know this is a topic we can certainly spend the next two hours talking about. The answer is it depends. Now if you look at a general trend when you’re looking across many organizations, many businesses, many industries, you start to see a smoothing factor, and the smoothing factor is that no one is hanging their hat on one language anymore. I mean even if you think way, way back to the 80s, the early 80s, right? COBOL was the language of business, end of discussion. Nobody argued that point in the 70s and the 80s. RPG right, what does the R stand for?

Charlie: Report.
 
Mike: Yeah. RPG was brought in as a utility language, or as Dr. Frank Soltis liked to say, the macro language, and they weren’t talking about RPG I and RPG II, right?

Charlie: Right.
 
Mike: This language was brought in because writing reports in COBOL—you know, building those picture clauses, was time-consuming. I could do in 20 lines of RPG code what takes me 200 lines of COBOL, and it was about productivity—and really what was it about? It is about using the right hammer on the right nail, okay? I could write reports in COBOL. I’m good at writing reports in COBOL. I’m faster at writing reports in COBOL, but if I’ve got someone who is comfortable with RPG, and back in the day it was RPG II, they’re going to be a lot faster than you are. Even if you’re cloning that COBOL report, this RPG guy is going to be a heck of a lot faster. So you know you know that’s where it began way back in the 70s and the 80s. The stuff that I’m talking about is not new, but we’re starting to see it. It’s encroaching upon the environments in which we operate. So you know IBM i customers and RPG developers—I’ve run into a lot of them out there who you know—and I’ll say the graybeards, right? The folks who have been doing this stuff for a long time. They remember the old IBM SEs and how the SE would come into the shop or they’d come visit—physically walk into your shop, talk to you and walk around. You know shake hands with people and meet with your VPs and your owners and that kind of stuff and the SE would tell you what to do. I mean that was the SE’s job. They would guide you down the path. They would sell you a System/3 and they would sell you COBOL, then they would sell you RPG. It’s like hey, well RPG is here too and look what it’s going to do for you. Then you know down the road, they sold System/38s, and down the road they sold them AS/400s, and then the SEs went away, right? But there are still people out there who have that mentality that IBM is going to tell them what to do and going to lead them by the hand. That changed, I would say, probably late 90s/early 2000s. You know a funny story: I had a customer I worked with in New Jersey, and you know they were an RPG shop, doing amazing things with RPG, and we’re talking to the customer about going to PHP, saying this is the next logical step for you because they were into RPG/CGI. And the guy’s like, you know he’s hemming and hawing, hemming and hawing and he says I’m going to IBM Rochester in a couple of weeks and I’m going to meet with Steve Will, and Steve Will is going to tell me what to do next. I laughed. He’s like why are you laughing? I said Steve Will won’t tell you what to do next. He will tell you about all the things you can do next, and that’s because IBM is not in that business anymore. They’re not telling you to use this or use that. They’re creating that palate with so many more colors now and so many more opportunities, and now when you get into modern application development, especially in DevOps infrastructure around open-source, it’s no longer about how fast I can write a piece of code, how fast I can do this? It’s about where is the code in the open-source space that I can bring into my shop and implement quickly and support. So you know the idea of saying we’re an RPG shop or we’re a Java shop or we’re a PHP shop? Well, that might be the principle you’re using for maybe 60-80% of your bread and butter, but the reality is you’re probably using other languages, too. Like for example IBM i shops take it for granted. Anybody out there using CL? Of course you are, but you just assume that you know that’s there because that is part of the infrastructure of the machine. But RPG has been our flagship for so long, primarily because for so many years there weren’t really many options. It was RPG, it was CL, and it was COBOL. Now open-source comes along and now we’re getting parity with some of the more pure open-source shops out there. Like you know I like to pick on Netflix as a good example of a pure open-source shop. They’re one of the few companies out there that actually puts their code out on the net for people to see. So their actual applications, the things that you use under your Netflix application is in the open-source space, but they’ll tell you about how they are using Node.js for this and they’re using Python for that and they’re using Ruby for—I don’t think they’re using Ruby but you understand what I’m saying. They’re using a variety of technologies because they’re using the right hammer on the right nail.
 
Charlie: Well that’s all true but I mean it’s still a generational thing. I still think this dichotomy exists where we have the—gosh I hate to say it but you said it first so I’ll just repeat your words—the graybeards, which is obviously some—a bit condescending perhaps but it—
 
Mike: [Laughs] I’ve got a graybeard, really. I can completely tell you I’ve got a graybeard now so—
 
Charlie: All right but in any event sometimes this mentality exists where we have proprietary systems vs. open. It’s almost an us against them mentality, right?

Mike: Right.
 
Charlie: I think anybody who is more progressive understands that you need—and you’ve said it already—you need to really adopt both. But there are some shops even today that will dig in and they simply will refuse to accept open-source and then they’ll give you a whole score of reasons why they shouldn’t do it—for example well, security. That’s big red punching bag, right?
 
Mike: Sure.
 
Charlie: And that’s a big one. So how do you address objections like that—especially in security, because I mean that’s another two-hour discussion, if not longer.
 
Mike: I would agree but let me throw something else at you. I’m going to go down with curve ball really quick and then I’ll come right back to security. I think there is a valid point to be made, there’s a valid concern to be made for companies who are understaffed, right? That’s a big issue and I see this a lot in the IBM i space. It happens in every industry. It doesn’t just happen in IBM i, but you know I certainly see it a lot in the IBM i space. We don’t have time to learn this, we don’t have time to adopt it. We don’t even have time to learn RDi let alone learn about open-source, and that kind of stuff. And when you peel the onion back a little bit—in some cases that’s myth because they probably have the time if the management team can make it—but I was sitting next to a guy yesterday at a dinner meeting and he remembers back in the day which was like ten years ago, they had 15 people in IT. Now there’s two and he’s sitting there at a presentation about PHP stuff, right? Because he’s going to get in that hour, he knows he’s getting a lot of good material and he should take that back to his shop because they’re using a little PHP. What’ s the chances of him adopting Python? He probably doesn’t [have] the cycles in the day to do that kind of stuff. So I think staffing is one of those things. It's management priorities and that kind of stuff as well. Now I promised I’d come back to security.
 
Charlie: But now I want to just interrupt you for a second.
 
Mike: Sure, please.
 
Charlie: But is that issue singular to IBM i shops? I don’t think so.
 
Mike: No. No.
 
Charlie: I mean there’s always pressure on, you know, on IT directors or companies in general to not have to hire additional resources, things like that so—
 
Mike: Right.
 
Charlie: So, I mean this is our world. Our world is IBM i. Maybe we can best speak to this.
 
Mike: Exactly. Exactly but it’s not and also another place that’s feeling this pain is Microsoft shops. The C# shops, the VB shops and that kind of stuff, because what’s happened is is that people who hung their head on a single technology, they only have a hammer, right? The people in the open-source space, people who have grown up with open-sourcelike there’s a fellow that I work with over here at Perforce. He’s very flexible about the technology because he has grown up with that world. He has grown up with the assumption that you’re not going to hang your hat on one technology. So you’re talking about a couple of things here. One is, you know, culture shock for the shop, and then there’s also culture shock for the employees, right, and how do you diversify? How do you get people, you know, comfortable with using more than one technology, more than one language? I’ll tell you management will respond very, very quickly when they discover that well, if we just use RPG to get that done it’s going to take us six months. But the open-source guy over here, he can get that done in three weeks. And management is like, why are we looking at RPG? I’m not disparaging RPG. I’m using that as a—you know, pick your technology. It could be PHP vs. Ruby or something like that. These kind of conversations go on out there but the reality is in the world of open-source, there’s a mentality that we’re not going to reinvent the world. We’re going to grab code from other places, we’re going to assemble code and we’ll tweak, tweak, tweak, tweak, tweak. And that’s okay to do that kind of stuff. You know again, the right hammer on the right nail sort of thing.
 
Charlie: Let’s go back to security. Again that’s like the pet excuse, I suppose, is security. So I mean I’ve done enough research on my own to recognize that security is probably not an issue but there—you know as people hold onto their old biases—that is still a concern. So talk about security in the world of open-source.
 
Mike: So, I mean yeah, where do you start? Let me start by saying—I’ll parrot things that Steve Pitcher says. Your IBM i and your RPG is not secure. Why are you whining about open-source not being secure, right, because he walks into shops all the time that they think oh, my IBM i is safe; it’s impregnable. It’s not going to you know, fall victim to this. And Steve finds a dozen vector points without even breaking a sweat, you know and that kind of stuff. So what we need to do is we need to remove the myth that anything is more or less secure than anything else, and we have to understand that security is—it’s cliché but I’ll say it: security is a journey, it is not a destination. RPG is not any more or less inherently secure than Python. End of discussion.
 
Charlie: That’s a pretty bold statement to make by the way.
 
Mike: Well I’m happy to make it [laughs].
 
Charlie: Okay.
 
Mike: Now if you’re running an IBM i, or better yet you’re running an AS/400 with nothing but dumb terminals and no network interfaces to it, that’s a darn secure machine. You’re not going to be able to—that thing won’t get hacked. You’re not doing that, though.
 
Charlie: But many people have the perception that they still are.
 
Mike: That’s the problem. It’s perception vs. reality, so now we need to level set the perception. We need people like Steve Pitcher out there and other folks out there like him, you know talking about this stuff and making sure that the people that have all this hubris about their IBM i being—you know, the most secure platform on the planet—are brought down to earth. Now we can have a level conversation about technology in general and open-source and that kind of stuff. Now is the security model of open-source different than it is for, let’s say, RPG? Absolutely. There are going to be differences. There’s a learning curve involved and that’s one of the challenges that we run into with this kind of stuff, but here’s the cool thing: The kids coming out of college, the kids coming out of school, they are all being trained on using open-source tools. They’re going to walk into your shop and you’re going to say hey, I need you to work on the network and work with this managed switchover here. Guess what? They’re going to pull out an open-source tool and they’re going to hit that switch with an open-source tool because why? Well there’s commercial tools they can buy, but an open-source tool, they don’t have to go to management and get a PO signed off. I can start using it today and I know this tool. I learned it in school, I trust it and it’s going to give me the information I need. And okay you know there’s a huge barrier to entry that gets lowered when you start looking at open-source solutions and like anything else, you know you have to maintain your open-source. You have to keep current with certain security features and security patches. That Apache server on IBM i is an open-source project under the covers and IBM is constantly updating it. And how do you update it? Well IBM makes it easy because they just make it part of the PTF package. You do the group for HTTP and now you’ve got all your updates for Apache. You don’t think about it but that’s how it works and for that customer out there who’s running V7R3 and hasn’t loaded PTFs in seven years, they don’t have the updates so they’re just as insecure as somebody else out there who’s not updating their open-source code.
 
Charlie: But Mike so you’ve talked mostly—I think primarily—about infrastructure, but I haven’t heard a lot of discussion about business apps. Some people call it meat and potatoes, you know the real hard core back-end processes. So is that arena still reserved for—I hate the word proprietary but is that the ultimate landing space for those languages? I mean again I hate to even use the word, but in a perfect world if you could redesign the world tomorrow, would that be your design where open-source maybe is doing infrastructure and maybe some of the periphery front customer-facing stuff and the back-end hard core mathematical processes is COBOL, RPG, etc.? Or can you do it all in any of them?

Mike: And the answer to that is the latter. You know so the reality is well, that people are going to use what they’re comfortable using. People want to leverage their investments. So I mean we hear Steve Will talk about this all the time, and maybe to a fault with the IBM i platform. You know companies have made investments in the systems that they’re running, right? And that could be RPG that’s 40 years old, it could be PHP that’s 15 years old, right? And you know to walk away from that investment is expensive, risky, time-consuming—
 
Charlie: Foolhardy?

Mike: You know foolhardy, it depends, and you and I could debate that subject because if the system that you have is harming your business, then you have to get away from it.
 
Charlie: Well yeah.
 
Mike: If your industry changed so much—and think about some of the industries out there that have changed so much. We were talking at a conference last week about you know, the tax situation with the municipalities in New Jersey when they got overwhelmed, their systems got overwhelmed and now their system has become a liability. You know their greatest asset, all that investment for the last 60 years now becomes a liability. You know the IRS system which was monolithic system, and I think ten years ago they started breaking it apart. Now it’s 140 monolithic systems. So you know systems can become a liability but again if it’s still an asset, if your business logic is an asset and it’s in RPG, I’m a big fan of leverage that moving forward. You know and we were chatting earlier about the whole hybrid cloud concept, and I think IBM a couple of years ago finally got that. The light bulb went on and it’s like you know, leverage your assets wherever they are and if that asset becomes a liability, then you have to replace it. You have to change it out. You have to you know replace that to become an asset again. So I would say—and what we’re seeing in the general open-source trending out there right now, companies, startup companies are doing everything online. They’re going everything virtually. Everything is Software as a Service. They don’t even have a database on-premise anymore, so first of all the thought of buying software is completely ludicrous. Nobody is buying anything anymore. We’re using Office 365 in the Cloud. We’re using Google Docs. We’re using, you know, all this other stuff. No one is paying for software anymore so—
 
Charlie: Well, let’s not—is that true? I mean you’re paying subscription.
 
Mike: And again, it depends. You could literally be running an entire business on Google for free. You don’t have to pay one thin dime. Now I’m not saying that’s a smart investment or that’s the way to go, but if you’re a startup with three people, with three employees, everything you need is in the free versions of the Google document strain. If you needed a desktop, you know there’s LibreOffice for the people who are adventurous, and it’s not hard. It’s not hard. I was on an advisory call with a local community college and we were talking about, you know the new version of Microsoft Office 365 is going to have all these new features. When should we roll that out to the students? When should we push that out and make the students start using it? I jumped in and I say hey, hold on. My niece is about to go to this school next fall and she is apprehensive because you’re telling her she has to use Office 365, but her entire K-12 experience was on Google Docs. Now she has to learn a whole new technology just because you are forcing her to. I said why don’t you guys embrace Google Docs as well as Office 365, and then of course the groupthink in the room, oh everybody is using Office 365. And then a lawyer, a paralegal, raises her hand and says no, no, no, we’re using Google Docs for everything and we’re a law firm because someone had said something like Google Docs is insecure. She says no, Google Docs is very secure. You can secure everything that way. So again, whether you can or whether you should is a different story. Obviously the three people in that fictitious, you know, startup company—if all of their background was in Microsoft Office, they’re probably going to buy an Office 365 subscription because that’s their comfort zone. That’s where they had their experience, that’s where they have their investment, right? But I think what you’re going to find with the younger crowd out there, they’re very, very willing to embrace new technologies. So they’ll look at Google Docs especially if they can save, you know, 40, 50, 80 bucks a month.
 
Charlie: We’ve talked about so much Mike and yet I’m not quite sure where we—it’s so—anytime I speak with you I think it’s so fascinating. But you said two points that I wrote down while you were speaking that I want to just expand on briefly. The first one is about your niece—or the law firm rather. Excuse me the law firm how Google Docs is not secure because—I’m sure that’s a very closely held bias that they have. Doesn’t this go right back to the mainframe again, how we started the conversation about the z System is an old ancient system running these legacy applications? Meanwhile it’s one of the most advanced systems on the planet—
 
Mike: Yup.
 
Charlie: Right but there’s a larger question here. Are we all guilty on some level at how our biases are really affecting our decisions whether to use open-source or proprietary or anything like that, and maybe to a fault to our companies that we’re doing work for?

Mike: Absolutely.
 
Charlie: Because we’re imposing our biases on them and maybe not doing what’s best for them.
 
Mike: And so what you have to do is you have to give yourself a little bit of slack, right? So a couple of things I’ll point out there: I’m a big fan of the line that says I think religion is okay as long as you’re honest about it. People invite me to a conversation, they know what they’re getting. They know they’re going to get a health conversation that’s going to be you know, pro open-source, it’s going to be pro IBM i. But I’m not a zealot. I’m not going to sit here and say that IBM i is the only solution for everything out there, right? Because I think that’s myopic, especially in today’s day and age. There are so many other things out there that can add to the equation. So you know you want to have an open mind and that sort of thing. So I think religion is a reasonable thing to have as long as you’re honest about it, as long as you recognize it. You have a conversation with someone who is an RPG III developer. They say RPG III is the best; that’s my religion. At least if they’re honest about it; I can respect them. I may not agree with them, I don’t have to sanction their position but at least they’re being honest about it, right? They say this is what I like, this is what I do. I’m like, okay great, that’s your belief. The thing that you have to do you know from an open perspective is you have to look at things from a higher level,, and in a lot of cases and that’s where we kind of get into like corporate strategy and that sort of thing. You know this is where businesses—if you as the staff, if you don’t push your ideas up forward, if you don’t push your agenda forward, your CIO is going to push it down, right? We’ve all heard the stories you know this company got a new CIO and now they’re moving everything off the IBM i to Microsoft. Is that a right decision? Is that a smart decision? I don’t know. We’d have to analyze that and really tear it apart and you know find out why he’s doing what he’s doing. Now it’s probably based on the fact that this new CIO is 40 years old but he’s spent the last 20 years working with Microsoft technologies and he knows they work. He knows they can work so we’re going to move everything to Microsoft because that’s his religion and you know—
 
Charlie: And bias.
 
Mike: And bias. Can you make it work? Yeah, you can make it work. Is it going to be a smart investment? We don’t know until after we’ve watched them for ten years try to struggle and move over [laughs].
 
Charlie: The other point I want to just bring up before we are kind of wrapping this up was the two key words I clinged [sic] onto here is liability and asset, because that is one of the metrics of deciding when to switch over from one system to another. And that can go either direction of course, but we talk about liability—and you mentioned like the New Jersey tax and the IRS, how their monoliths became you know, a whole plethora of monoliths, which is a problem obviously—but it’s not easy to walk away from an enterprise solution that’s so deeply embedded not only into the enterprise but into the people itself.
 
Mike: Correct.
 
Charlie: It’s so hard to walk away and it’s understandable I guess why these liabilities continue to linger on, because people don’t want to switch that quickly—and the same thing with the asset. So just familiarity being why you don’t want to switch, what are some of the liabilities? Tell me is there a black and white definition of oh my gosh, that is such a liability because, and this is why we have you make a case for switching for example.
 
Mike: I’ll throw one on the table that I think we’ve all lived through. And Charlie, I want you to push back on it because I know you will—I know you can, I should say. I shouldn’t say I know you will. I would say that the green screen is a liability, right?
 
Charlie: That’s a blanket statement you’re saying that?

Mike: Yeah, yeah. The green screen is a liability—
 
Charlie: Okay.
 
Mike: And here’s why. Anybody who is coming out of high school today has never seen a text interface. Unless they’ve been in a programming class, they don’t even know what a command line is, right? So you know that perception right there. You immediately as an organization, if you’re going to hang onto the green screen—and again I love the green screen. I fed my family for 20 years on the green screen development. You know it’s fantastic, but the world has evolved. The world has gone past that to not only GUI, but touch, right? It’s all about the touch interfaces now, so if you’re not developing UIs that are graphical touch-oriented, you have to own a learning curve. Now if that learning curve is worth it to you, then do it. You know if the green screen learning curve is still an asset to your organization, that’s fantastic. I’ve got four employees. I need to hire a new employee. It takes me maybe six weeks to get them comfortable with the text interface. If that’s an asset, then leverage that asset. But then the boss comes in and says yeah, but you know that order entry program? We need that on the web. Can you put that green screen on the web? You and I will be kind of laughing at that kind of a suggestion, and it’s no. We’re not going to put the green screen on the web. You’re crazy, you know, but he says I know but I need my customers to order their own stuff. Can you do that with the RPG and the IBM i? The answer is I can but maybe we should be using the IBM i and some PHP for that, or maybe we should be using Db2 and Node.js. You know there’s other technologies that may be more appropriate for web development. I can do RPG/CGI, and we were talking about this last week. I said yeah but think about that. All right. We already hear a lot of people complaining, oh I can’t find RPG developers. Well okay RPG/CGI is a great technology, fast. You’re not going to find any faster way to run web work on the IBM i, hands down. But what about the human resource aspect of that? You need to have someone who knows RPG, knows your existing business logic, knows HTML, knows Javascript, knows web technologies, knows stateless vs. stateful processing, all in one body. That’s got to be one body—oh and by the way if this is an important system, you need two of those bodies, because one of them wants to go on vacation and that kind of stuff, right? So go find that unicorn somewhere. There’s a couple out there.
 
Charlie: This is the unicorn you’re talking about?

Mike: Yeah.
 
Charlie: I suppose—I mean I guess I’ve seen a couple in my life perhaps, or my perception of somebody might have been a unicorn, right? So what you’ve described right there is a liability, clearly.

Mike: Yeah.
 
Charlie: Okay and one thing that comes to my mind, just you know, another liability of course is we need to roll this out by next Friday.
 
Mike: Yeah. Yeah. You know that’s a big deal and you know one of the other topics that you and I were kicking around was the idea of DevOps and that kind of stuff, and getting companies to the point where they can realize that they can’t be doing waterfall development anymore, right? That’s a huge liability because what’s happening is your competition is eating your lunch. If your competition is doing Agile, they’re rolling out new features and functionalities to their application sets weekly, daily, right? Amazon being the overwhelming—they’re the exaggeration. They’re rolling out new features 3,500 times a day, right? They’re constantly rolling—their infrastructure is that nimble. You don’t need to be that nimble. In fact, if I’m working with an IBM i shop and they’re rolling out new features weekly, I’m thrilled. That’s great; at least you’re moving down the road towards that whole Agile kind of mentality. Yeah, and so now you’re turning your application development into an asset as opposed to a liability in that scenario, because you could be more responsive to your customers. Because the ones who are going to be more responsive to their customers, they’re the ones who are going to get the business.
 
Charlie: Wow. Mike, we’ve talked about it all. As is often the case, once we get rolling, we tend to get on a roll, which I definitely love but we have to kind of leave it there. We’ve been going for quite some time and I really just wanted to thank you as always for such an engaging discussion. It's always amazing when we get into these conversations. So thank you very much for what you bring to the table and your welcome knowledge in this space, and obviously once again for all you do for the community. It’s really wonderful so I want to thank you very much and it’s very apparent to me why you’re such a great IBM Champion.
 
Mike: Well thank you, Charlie. I’m going to kind of throw this right back in your face as well. I mean one of the reasons I think that you and I get along so well is that we have a certain amount of confidence and trust in each other from a communication perspective, and we can push each other around a little bit, you know, and explore some new ideas. I’ve tried and I can’t seem to offend you, so that’s a beautiful thing. I’d like to see more of that our society of course, but I think that’s one of the keys to freedom of communication is you know, trust in the ability to explore other ideas whether you agree with them or not, right? You might think I’m totally half-baked on some of the ideas I have out there and vice versa, but you know what? We respect each other enough and I think that’s what we need more in today’s day and age. Thank you Charlie for creating that environment for me.
 
Charlie: Well, you’re very welcome. As a safeguard, we’re going to turn off all comments to this podcast so [laughs].
 
Mike: Mike Pavlak said RPG is dead. No, I didn’t say that. Come on. Green screen maybe, not RPG.
 
Charlie: Okay. All right. Well, that’s a great way to end the conversation. Mike, thank you very much. To anybody who is still listening to the podcast, thank you very much for hanging in there with us. I think you were well rewarded with some great thoughts from Mike, so thank you again Mike. Be sure to check out other offerings on TechChannel. They have a whole wealth of great content on there and you won’t go wrong doing that. Anyway, until next month everybody, this is Charlie Guarino and thanks for listening. Take care. Bye now.