In today’s episode of the “I Am A Mainframer” podcast, Steven Dickens sits down with Paul Scott, Chief Architect Web Enablement Technologies at Phoenix Software International. On this podcast, Paul discusses his journey with the mainframe, advice for those just starting their journey with the Mainframe, and where he sees the Mainframe going in the future.
Steven Dickens: Hello, and welcome to the “I Am A Mainframer” podcast. Brought to you by the Linux Foundation’s collaborative project, the Open Mainframe Project. My name is Steven Dickens. I’m your host today, and we’re joined by Paul Scott, from Phoenix Software. Welcome to the show, Paul.
Paul Scott: Hi, it’s good to be here.
Steven Dickens: Paul, let’s dive straight in here. If you could tell us a little bit about your background. What you do for Phoenix? Just to get the listeners orientated.
Paul Scott: Yeah, sure. I’ve been involved with IBM mainframes for well over 35 years. I started out in this business doing conversions from one language to another and one operating system to another. It was always to IBM. For instance, I did a RPG to COBOL and I eventually specialized in Assembler to COBOL. I know that seems kind of strange, but we actually did that. From there, I went on to banking. I worked for a savings and loan for a while, on IBM mainframes doing Assembler coding. Eventually, I went into contract work, and I did a lot of conversions. Worked for another savings and loan, where we converted a MVS, I’m sorry, a DOS VSE system to MVS/XA. Did a lot of work with proprietary database stuff. Converting that, as well as converting things like microfiche readers.
I worked on an ATM system, hooked up to an IBM mainframe. I got a lot of experience with VTAM. And we converted BTAM to VTAM. Eventually, I started doing more things with GUI and PC work. I got away from the mainframe for a little bit but eventually came back to the mainframe, working for Phoenix software. I’ve been here now for about 17 years. Really enjoy it.
I’m working now with Phoenix’s flagship product (E)JES. I have worked on (E)JES web, which is a web interface, to the 3270 legacy product. Did a lot of work with Java Native Interface, Java, to make that work. From there, I’ve been doing a lot of web modernization. Building our portal for our customer support, using Node.js on z/OS. That connects to our Intel Linux system, which does the web serving as well as Z Linux, which has our customer support portal.
Steven Dickens: There’s a lot in there, from what you’ve just mentioned. What is your actual role today for Phoenix, Paul? What’s your title?
Paul Scott: So my title is Chief Architect of Web-enablement Technologies. Basically, what we’re doing, is trying to modernize our product line for web technologies.
Steven Dickens: What’s driving that? Is that just modernization, to keep the product current? Or is that customer demand that’s driving it? What’s really the driving force there?
Paul Scott: Yes. I think a lot of it is customer demand. Customers are expecting more web application base. They’re expecting GUI interfaces. Yeah, that’s what we’re doing, is modernizing it to a web-based GUI interface.
Steven Dickens: You mentioned you were doing work there with Linux on the platform. If you could just expand on that for the listeners? We’ve got a couple of different communities here. We’ve got the sort of traditional set of west space, but also as part of the project, and a lot of focus on Linux. If you could expand on what you’re doing in the Linux space, I think that’s going to be interesting for our listeners.
Paul Scott: Yeah, sure. We’re using Z Linux for a number of things, but what I’m doing currently is merging our traditional problem resolution database, which is GNATS. I’m making that work with our Intel-based web server. For instance, a customer will go to our web portal, and they will want to get a list of their problems. I use Node.js on the Intel platform, and Node.js on Z Linux, to build a communications channel between the two. It’s all secure because the Z Linux system is behind the firewall. You can’t connect to it from outside. I have the Z Linux system using Node.js connects to the Intel web server that’s internet facing. It does the connection, and it leaves a connection open indefinitely. Over that connection, we talk between Z Linux and Intel. When a customer goes to the web portal, ask for their problems, they’ll get a list instantly from Z Linux. They can update their problems via the web, communicating with the system on Z Linux.
Steven Dickens: Were you using that for the whole of Phoenix software’s portfolio, for logging issues? Is that the common interface across the whole lot there?
Paul Scott: For our customer support portal? Yes, absolutely.
Steven Dickens: Paul, you mentioned modernization. Can you say a little bit more about that for me?
Paul Scott: Okay, sure. What started this modernization ball rolling, was we had a licensed system that customers would use via the internet, via the web, your web browser. It was not a demand-driven system. It was not an event-driven system. The customer would put in their request for a license, that would be sent to the file system, which would then be using a polling system. It would get pulled from, let’s say, Z Linux. I’m sorry. It would get pulled from z/OS. A z/OS process would then run in the background, and deliver the license via email. This was a long process, a polling process, that could take up to 10 minutes for the customer to get their license. We wanted to modernize that. Using Node.js on z/OS, and using Node.js on the Intel platform, we were able to get the two platforms to talk together.
The customer makes a license request, and immediately using this event-driven processing that Node.js does so well, we were able to then get them a license instantaneously, instead of having them wait for an email that could take up to 10 minutes. That was a very successful project. The next step was then, “Well, what else can we modernize?” The next thing, of course, was customers had been asking to be able to see their problems via the web, be able to update them, and be able to interact that way. That was the next step. We have an antiquated system, GNATS, sitting on a Z Linux system, and we use the same interface we built for z/OS to the Intel platform using Node.js. We use that same system, but from Z Linux to the Intel platform. Now customers can get a list of all their problems, and interact with those problem records instantaneously using the same z/OS to Intel platform we built before. We’re looking to expand that in other ways, too. Customers want to upload files for support. We’re building the same kind of interface for that as well.
Steven Dickens: Interest in there, that you mix in both of the Linux and the z/OS side. Based on what you mentioned, call from me background, would it be right in saying that this project has meant you’re new to the Linux side? Is this the first time you’ve dug in, on Linux, on Z, on the platform?
Paul Scott: I’ve been using Linux for quite a while. I did a lot of contract work on Linux systems. I’m very familiar with Linux, but this is the first time that I’ve used Z Linux, on the Z platform.
Steven Dickens: Yeah. How have you found that? How’s that been, that first experience from using Linux on Z?
Paul Scott: Yeah. Linux z/OS is absolutely awesome. It runs on an IFL, so it runs full out. We get extremely good performance and we are running Red Hat Enterprise Linux 8. It’s a modern Linux, just like what we’re running on the Intel platform as well. Having experience on the Intel platform, it translates directly when running on z/OS as well.
Steven Dickens: Are you thinking of containerizing, or are you just running on Rell at the moment?
Paul Scott: We don’t have any plans to do any containerizing right now. We have expertise in-house to manage the systems that we’re running. We don’t have any need for that at this point, but we don’t know what the future might bring. We may want to go there in the future.
Steven Dickens: Okay. What do you think is next? It sounds like you guys are on a journey of transformation. You seem to have taken the learning from the first application, and applied that to the second case. What else do you think is next, as you look to transform, and take the Phoenix software interfaces on a journey? What’s next for you guys?
Paul Scott: The next big thing out there is Zowe. We’ve already started writing Zowe applications. We’ve got the Zowe CLI apps that we’re doing, and we’re going to be working on the Zowe GUI Apps as well. We’re going to be putting our (E)JES web product on Zowe Desktop. We’re going to be looking at what our other products can do, too. We’ve got some new products we’re looking at modernizing as well. The future is open. Lots of stuff we can do.
Steven Dickens: How’ve you found the Zowe piece? It’s been really fascinating watching that project develop over the last two to three years. What’s been your exposure and your experience?
Paul Scott: Currently I’m only working on Zowe’s command-line interface, the CLI, and it’s pretty rewarding. It’s a good standard interface. It allows the customers to find our product through the catalog. I’m looking forward to doing more with the GUI interface because that’s what I think our customers are going to want to do more of. The CLI is great for running ad hoc and batch-type work. The Zowe Desktop GUI is, I think, where people are going to want to spend a lot of their time.
Steven Dickens: You’d see that as the next evolution? Getting into that more GUI-based interface?
Paul Scott: It’s looking more and more like that is the next evolution, yeah. We’ll see how it goes. More and more people are starting to get interested in Zowe and adopt it. We’ll just have to see where that goes, but it’s looking promising. Yeah.
Steven Dickens: One of the things we spent some time talking about at the very top of the call, was your background, and how you’ve been on this platform for some time. We seem to have a lot of younger listeners on the podcast, here. What would be your advice? If you could speak to your 22-year-old self? You’re coming out of college, and you were able to go back and speak to yourself. What would you be saying around how to build a career in IT? What would those lessons be that you’ve learned over the years?
Paul Scott: I would say, “Never stop learning. Have an interest in everything. Don’t limit yourself in any way. Learn new languages. Learn new platforms.” People say the mainframe is a dinosaur. They’ve been saying that for the last 25 years, at least. 30 years. We still have the mainframe, nothing compares to the mainframe. When you look at RAS, reliability, availability, and the serviceability. The IBM platform is the best out. I say, “Learn IBM systems. Get involved in all that. Learn Gooey systems.” IBM has done a lot of work to bring a unique system services to z/OS, which is fantastic. It allows us to do so many of the things that we’re doing today.
Node.js wouldn’t exist without z/OS GUI System Services. There’s Java on the platform. IBM is putting Go. They’ve got Python. They’ve got all of these different languages that are now on the mainframe. These are on z/OS, this isn’t just Z Linux. This is on z/OS operating system. Whatever you can possibly learn, just keep learning it. Never put yourself into a little box. always try and learn new things.
Steven Dickens: I think that’s solid advice, and I obviously agree with that. One final question, Paul, as we look to bring this home. Where do you see the mainframe platform? We talked about the long touted demise of the platform there, over the last 20, 30 years, but where would you see the platform 3, 4, 5 years out from now? What do you think we’ve got coming?
Paul Scott: I’ve got some information, some technical disclosure stuff that I get wind of. I can’t really talk about that, but I see that the mainframe is moving more and more into areas that were traditionally done with PCs, and UNIX, and Linux. You’re seeing that picked up more and more on the mainframe. I would say the sky’s the limit. I see a progression happening here, with all the modern technologies. I think we’re going to see a lot more of artificial intelligence happening on the mainframe. Like I said, “The sky’s the limit.”
Steven Dickens: I tend to agree with you. I think that’s a fantastic way to wrap up. You’ve been listening to the “I Am A Mainframer” podcast. I’m your host, Steven Dickens. If you like what you’ve heard today, please click and subscribe. If you could take a couple of moments and give us a five-star rating on your podcast-listening platform of choice, that always helps. Join us again, next time for the “I Am A Mainframer” podcast, brought to you by the Linux Foundation’s Open Mainframe Project.