Z DevOps Talks Season 1 Episode 5 with Bruce Armstrong Part 1:
Z DevOps Talks Season 1 Episode 5 with Bruce Armstrong Part 2:
Hey, welcome back to the show, thanks for joining us! In Episode 5: The Mainframe Strikes Back, we tricked Bruce Armstrong into joining us for an episode. Bruce is an IBM veteran, Offering Manager, and serves on the leadership of the Open Mainframe Project’s Zowe™ Project. Bruce sure lived up to all the hype; he’s a wealth of knowledge when it comes to Zowe™, the mainframe in general, and the Open Mainframe Project. In our latest episode we covered so much ground that we had to split it into Parts I and II. Here are some of the topics we covered:
- Zowe™ – what is it and how it was framed
- The Open Mainframe Project
- Zowe™ as the infrastructure’s plumbing
- VS Code and Eclipse Ché making use of Zowe™ services
- Wrapping the Zowe™ framework in restful APIs and how that contributes to opening up the mainframe
- Use cases for Zowe™
- The cultural shift in the mainframe community
- How old dogs can teach young pups new tricks
- Making the mainframe less remarkable
Care to learn how you too can join the Zowe™ revolution? Bruce recommends the following resources:
- The Open Mainframe Project
- What is Zowe™?
- Git Hub
- Zowe™ zTrial
- Getting started with Zowe™ Tutorial
- Getting started with Zowe™ WebUi
- Zowe™ Badge
- Zowe™ on Slack
And in case you haven’t heard our other episodes, check them out here. Happy listening!
Chris H: Welcome back to zDevOps talks. We’re here with Bruce Armstrong. Hi Bruce, thanks for joining us.
Chris H: I’ve got a little background on Bruce here – nothing too crazy.
Chris S: The rap sheet.
Bruce: I was gonna say – I’m not sure I’ve screened this.
Chris H: I have to tell you, it’s a lot easier when we interview Millennials and people in their 30’s, because they’re a lot more open with social media and posting things online.
Chris S: Meaning we have to dig a little bit more. But I’m sure that you probably would prefer that, right, Bruce?
Bruce: The statute of limitations of things I’ve done in my past have probably run out, so I’m ok.
Chris H: Bruce Armstrong is a pretty popular, not a common name, I don’t know if you’re the same Bruce Armstrong, but I had to look back at some of these NetWorld and Computer World archived magazines and there were some mention of Pentium processors and some SQL stuff.
Bruce: So what you’re really saying is that you’re an old guy and I had to do a lot of digging. That’s kind of the summary, right? And for the record, as of next week – and I’ll show my age because I listened to Rosalind’s broadcast and she was very proud of her age. I actually have a little more years of service with IBM than she did. As of next week I’ll have 35 years.
Chris S: Wow, slow clap for that! That’s amazing!
Bruce: Just to show how much older I even am – IBM was not the first company that I worked for. I actually worked back in the day for Boeing Computer Services for 5 years. So I can say I have forty years in the IT industry. And I think, ‘Oh my God! What happened?!’
Chris S: Really? That’s amazing.
Bruce: And because I know your audience is primarily Millennials, I’ll just say “Time flies when you’re having fun”.
Chris S: Absolutely.
Bruce: That’s the net of it.
Chris S: I like that – that’s a very positive message for our listeners.
Chris H: Two of my sources had you at somewhere between 31 and 33 years, and one mentioned additional years in the IT industry with a large Aerospace company, Boeing, so that takes care of that.
Bruce: And I actually worked my way through college at an IT company, NCR, which was known for cash registers. They made a line of mini computers and while I was going to school I was actually a second shift operator. So maybe there are those of you out there that can relate to feeding punched cards through a card reader and hanging nine track mag tapes? Yeah, so I’m an old dog.
Chris S: He knows a thing or two.
Chris H: The programming language of the day would that be ALGOL?
Bruce: I started out with Fortran. I’ll be quite honest, I didn’t do a lot of COBOL. I did a little assembler. PL/1 was big back in the day. The interpretive language was Rexx, which is still around.
Chris S: Still very much around, yeah, for sure.
Bruce: Anyway – we digress.
Chris H: Man, you’ve done a lot. So current focus is mobile and System Z.
Bruce: Yeah, the mobile stuff was a few years ago, but go ahead.
Chris: Expertise in integration and System Z and enterprise mobile strategies, practical
application of cross-platform network and systems management products and processes. I think we covered a lot of other stuff. The other thing is that you are on the Zowe™ Leadership Council. The Zowe™ Leadership Council oversees all code development and release management for the Zowe™ project. In particular the ZLC has the following basic roles: Receive proposals for sub-projects leveraging Zowe™ technology and will decide via a vote on whether the subproject is accepted – and a whole bunch of other stuff. Sounds like a pretty important deal. I don’t know what Zowe™ is so talk to me more about this.
Bruce: So my role inside of IBM. I have the title of Offering Manager – which a lot of other companies call them product managers – but let’s just say I’m Chief Cat Herder. And in the case of Zowe™; this is an open source project, and I am an IBM Offering Manager assigned to the open source community. We can spend a little more time on how this came about, but the short version is that three companies came together, late in 2017, to contribute a set of technologies for helping to speed modernization of the mainframe
This Zowe™ leadership committee kind of oversees the volunteers, from not just the three founding members, but from other companies, on what code gets contributed and what we’re trying to do with Zowe™. And the main message is that it’s for the good of the platform. We’re trying to make z/OS appealing to the next gen system programmers, developers and such, and again we can talk about it some more…. but also I’ll say, to cure some ills that have grown up over the last 50 years of the platform, that make it confusing and difficult for people to use and consume. So those are kind of the two main themes of what the Zowe™ leadership committee is trying to do. How do we make this platform attractive for next-gen? How do we make it even more appealing for everybody, not just next gen?
Chris H: So while you’re talking, I have a slide deck here that looks like one of those investor slide decks. You said that three founding members are IBM, Rocket Software and CA Technologies, right?
Bruce: Correct. And CA Technologies has since be purchased by Broadcom. So depending on the age of the slide it will reference or say either CA Technologies or Broadcom. CA Technologies is a Broadcom asset.
Chris: So you mentioned it’s all open source. So this is all licensed under the Eclipse Public License.
Bruce: Correct. EPL2.0 is the jargon.
Chris H: It’s the open source project under the open mainframe project.
Bruce: Yes, and this is something that I’ll say for us ‘old guys.’ You know, z/OS is clearly an IBM platform, and we’ve survived 50 plus years of evolving. Open source is a little different beast. So this is really kind of merging of the two worlds. The EPL 2.0 license, just in layman’s term, says ‘these three companies contributed this technology and anyone can use it – whether that be a customer, an ISV – whoever. And the idea here is, going back to my theme of good for the platform; we three companies at the same time we contributed this, we said ‘and we’re going to use this technology in our products to help further this along.’ People think open source is kind of the answer to everything, but really, open source is just open source. It’s what you do with it that matters. So we’re kind of seeding the IT industry to say, ‘Hey, here’s a set of technology to kind of rally around and oh, by the way we’re going to use it too, and we’re solving real-world problems with it – come join us.’
Chris H: Got it.
Chris S: It’s a good message.
Chris H: I think a good example is what Red Hat’s doing with Linux. I mean it’s open source, but they’re rallying around that and providing support around that.
Bruce: Exactly – and there’s a good analogy there. In the Linux story, Linus contributed this, and in our case, you have the three founding members, who know z/OS, so we contributed the open-source. And you know, the issue of support and all of that are things that we’re working on in the community, just the same way Linus kind of planted the seeds and you had Red Hat or Ubuntu or whoever, kind of pick up that open source and turn it into stuff that was enterprise-ready. So there’s definite analogies here to what has happened in the Linux world.
Chris H: So Zowe™; and tell me if this is still true -‘ is an extensible framework for connecting applications and tools to Mainframe data and applications.’
Bruce: That’s a mouthful, but yes, that’s what it is.
Chris H: I feel like these were at one point technical statements that were turned into, or attempted to turn into, something that was more consumable by wider audiences. So Zowe™ ‘aims to make the mainframe an integrated and agile platform.’ I think agile and DevOps – we could probably add that in there as well – ‘within the changing IT architectural landscape’. And this is one more thing; it’s pronounced Zowe™. And it’s pretty interesting how it’s not an acronym. That’s the second bullet here – ‘it’s not an acronym; it’s just a simple, fun and easy name.’
Bruce: And it drives people crazy. Because they look at it and they say, well Z is obviously for System Z, and the ‘O’ – well maybe that’s for ‘Open’ and the ‘WE’ – well, maybe we could say ‘We’re doing this’ or something like that.
Chris H: I actually thought about this, and this is funny because – who’s the episode we had on ’embouchure’? Was that Suman?
Chris S: That was Suman’s – yes. That was the episode before this one for all of you listening.
Chris H: And we should just pause here for a second to recognize that Suman was just recently promoted to STSM, which is Senior Technical Staff Member. So congratulations to Suman.
Bruce: For those of you who are not in sync with all the IBM terminology – that is a BIG step in one’s career in IBM.
Chris S: Oh yes – it’s huge.
Chris H: I believe the next step is Distinguished Engineer?
Bruce: Correct, yes.
Chris H: So that’s huge – that’s a big deal.
Bruce: And it was well deserved.
Chris H: Ok, so I thought Zowe™ was Z – z/OS obviously, ‘O’ for open source and the ‘WE’ for web – because there is a component to Zowe™ that is the web interface, right?
Bruce: That’s right and an excellent point. You know it’s almost fun to say, ‘make it whatever you want it to be.’
Chris H: Oh geez. It’s like that movie, Inception, where you not really sure about that little top thing – what is it called?
Chris S: The dradle.
Chris H: That’s a portmanteau, right? Not an acronym, a portmanteau; when you combine these different words together and create a new thing.
Bruce: Oh, I learned something.
Chris H: It’s French.
Chris S: Learning French with Chris and Chris.
Chris H: So I think it’s not an acronym, but it is a portmanteau.
Bruce: There is a little side story on how this came about. It’s crazy – one of the hardest things we do is name stuff. This started out as “Zoe”; Z – O – E. And it was one of those – you threw it on a wall and it stuck kind of things. It was easy to type – so as you’re creating a directory structure, whatever, three characters, that’s good enough… It is when we went to the Open Mainframe Project, that is part of the Linux Foundation, and this ties in with how is open source, what’s different about this, and how we are merging the world of open source and z/OS.
So think of the open mainframe project and the Linux foundation as a big, old, not for profit organization; again, not only for the good of the platform, but the good of computers throughout the world. They deal in trademarks and the idea here was that “Zoe” was to become a registered trademark of the Linux Foundation. So when the lawyers looked it up, it turned out that “Zoe” was already trademarked. We had to come up with something that made it unique, so we stuck the W in it.
Chris H: Nice; the truth comes out.
Chris S: I’ll admit, whenever I go to these shows – which I’m going to miss SHARE in Fort Worth – I’m very disappointed about that – however, it’s because my first kid is supposed to be born in the beginning of February, so there’s never, ever, been a better excuse.
Bruce: Your world will never be the same.
Chris S: But every show, someone will come up to me, and ask, “What does Zowe™ mean?
What does it stand for?”
Bruce: My goal in life, in 20 years from now, is that in Trivial Pursuit there will be a question asking, ‘why is Zowe™ spelled with a W?’
Chris H: And if you didn’t listen to this podcast, you’re going to lose. I hope after this airs, that Wikipedia is updated. That Zowe™ is in Wikipedia.
Chris S: Checking Google now and there is NOT a wikipedia for Zowe™. We need to fix that.
Bruce: Alright, I’ve got an action item from this podcast. I’ll take that ‘to do’ and I will take it to the ZLC for a vote.
Chris S: We’re learning French, we’ve got action items – what is this? (all laugh)
Chris H: Ok, the next topic is – what makes Zowe™ unique? And from what I can understand, cloud interfaces is the biggest one – and nested underneath that are application framework, command line interface and then API mediation layer. Right? So those are some big topics.
Bruce: Right, so let’s break that apart a little bit. It kinda goes back to our wordy marketing bullet that sounds good but doesn’t mean a lot to a whole lot of people.
Chris H: It’s inspirational though.
Bruce: That’s true.
Chris H: It’s provocative.
Bruce: My simple description of what Zowe™ is; is that we are ‘wrappering’ z/OS services with interfaces that make those services easier to use. So let me give you an example. REST APIs. Zowe™ provides some REST APIs. A lot of the REST APIs are actually coming from the operating system itself, through a service called z/OSMF or the product teams. And when I say product teams, I’m talking CICS and IMS and DB/2. And those REST APIs think of Zowe™ as being a catalog of capturing all those REST APIs for some minimal API management. Is the API up? Is it down? What version of the API? Can I load balance across multiple APIs? That kind of stuff. That’s what’s in the API mediation layer. So think REST APIs when you think Zowe™.
The other piece that you’ve touched upon is, well how do we use these APIs? Well, you could use them through a web browser or you could use them through what we call a command line interface. And think of those as two visualization techniques, right? If I love browsers, I can go interact with these REST APIs through the browser. If I’m a command line guy or gal, I type a command line. I’m still talking to the same set of REST APIs in the background, but the information is presented to me in a different way.
Chris H: So correct me if I’m wrong; that is almost exactly like IDz, right? In that you have an emulator for 3270 or you have the Eclipse GUI. Is it similar?
Bruce: Yes, it’s similar. And I’m glad you brought this up. I know your target audience is mainly a DevOps interested audience. I think of Zowe™ as infrastructure; as plumbing. What you do with that plumbing, has a full spectrum of possibilities. Now in my mind, IDz for example, is a product implementation for a developer. It happens to be Eclipse-based. But I don’t think it’s a big secret to say, that Eclipse client making use of REST APIs seems like a natural progression for where we’d want to go with that.
The best example of the merge of Zowe™ open technology and an open IDE is something like Visual Studio (VS). They’re making use of some Zowe™ services. Probably Eclipse Che. And that’s where I’ll point you back to Rosalind as far as all the permutations and combinations of what’s going on with that technology.
Now that’s for a developer. Zowe™ could equally be used by a system programmer. By operations people. By any number of people that can use REST APIs, a command line interface or a browser interface. So I think of it as plumbing and what you do with that plumbing is kind of, whatever the business needs. Whether you’re a software vendor that wants to use this plumbing. Whether you’re a customer that wants to write your own application and use this plumbing. Or wether you’re an IBM product team that wants to use this plumbing.
Chris H: So maybe what Zowe™ is trying to do – and this is in my head – is if you have an enterprise that wants to do a thing. So enterprise, specifically mainframe related – wants to do a thing. So if we relate it to your plumbing analogy; they say we need this valve, we need this flange, we need these pipes, we need this type of schedule, pvc, this, this, this and this…and we’re going to go to this vendor, that vendor, etc.. So six months later, we’ve got everything, now let’s go ahead and put it all together. And that’s traditionally the way it’s being done. But essentially what you’re saying is that you have an enterprise and they can basically turn to Zowe™ and say ‘this is what we want to do.’ And Zowe™ will say, ‘ok cool – we already have all that stuff for you, so let’s go ahead and set it up.’
Bruce: That’s a good analogy. And rather than going and shopping a bunch of technologies from a different set of vendors, for example, wouldn’t it be nice if those vendors agreed on a set of infrastructure? And that infrastructure was consistent. It was secure. And rather than the customer having to deal with a whole bunch of technologies, from a whole set of different vendors – what if the vendors agreed to use Zowe™ as an integration platform? Granted we’re talking browser, we’re talking command line, and I’ll let you in on a little secret…there’s a proposal coming to the ZLC on what are we going to do about mobile. So imagine this technology as something that is consistent on z/OS for anybody to use.
One of the other bullets… we’ve mentioned cloud, we’ve mentioned agile in some of this. The overriding theme here is that you wrapper all of these z/OS services; with REST APIs and these user interfaces – and suddenly that platform is not this weird uncle in the closet. Right? This is a set of REST APIs. And if you know how to work with REST APIs, you could talk to cloud, you could talk to z/OS and really you don’t see a difference. I usually point out that you still need somebody that understands what JCL is and where the comma goes on the job card and all that kind of stuff. But for a vast majority of your system programmers and developers, and operations and devOps – whoever; just being able to call a REST API and get information back from the z/OS platform should be no different then doing it from a cloud environment. And that’s kind of the great equalizer that we’re trying to do here.
So I’ll go back to my two themes. We’re trying to make the platform attractive to millennials by putting this technology as the interface of choice on how you interact with it. And when I talk about ‘correcting the sins of the last 50 years’ – you know the good news is we’ve got a lot of interfaces on z/OS – the bad news is we’ve got a lot of interfaces on z/OS. So imagine trying to converge all of that to a set of REST APIs, an API catalog and these different ways to make use of those REST APIs. There’s everything you ever wanted to know about Zowe™ in 30 minutes.
Chris H: So, the APIs support the open standard specification, right?
Bruce: Correct. The short answer is yes. There’s actually a couple ways in Zowe™. If you don’t have an open API spec, there is a way in the API catalog of Zowe™ for ‘faking it out.’ So, you have the description of of the AP, you actually code it into the catalog, but the net of it is, yes, those APIs should look just the same as any other APIs. And what that means is, you’ve got a set of get, put, post, delete, kind of operations and you’ve got a set of Jason variables that you use to interact with those backend z/OS services. So no different than any other API.
Chris H: Also, the architect on my team told me that most people just copy and paste their JCL.
Bruce: That works until something breaks. Then you do need your weird uncle in the closet to come out and say “Oh, you’ve gone past column 72 and that’s a No-No.”
Chris H: If you want to make Zowe™ attractive to millennials you’ll need to position it and advertise it in meme format.
Bruce: Ok, I’ll take that as a to do, too.
Chris S: So what are some use cases that you see for Zowe™?
Bruce: One of the problems I have with that, is that it’s a little like saying ‘so what’s the use case for an automobile engine?’ In the sense that it kind of depends on what you want to do. If you go back to REST APIs; you know that you’re only as smart as what the REST API allows you to do. So the idea of submitting a job, getting the results back, editing a file; those are services that could be used by any number of people. In design thinking we spend a lot of time= on personas and who is your real end user and all this kind of stuff. We’ve struggled with the various personas that could potentially use Zowe™. So I guess I’m wimping out a little bit on giving you a straight answer, in the sense that it’s a set of services, we refer to it as an open framework, and it’s kind of whatever you want to do with it. I tend to talk about developers, system programmers and operations people, and so on. But if you had somebody that wanted to do something bizarre, as long as you had a set of REST APIs that allowed that – Zowe™ is kind of the funnel for being able to do that.
I would like to make one comment. We talk about REST APIs and I do want to draw a distinction between what I’ll call ‘back-office’ REST APIs. What you’ve heard me describe is editing a file and submitting jobs and that kind of stuff. There is a set of REST APIs, or products, that allow you to REST-enable business applications. These are the CICS, IMS, DB/2; the things that run a business. That is not Zowe™’s domain. We have other products that want to give you REST APIs into your payroll application or that kind of stuff. So I would put a caveat on the use cases as being kind of the back-office type of operations and not customer-facing, if that makes sense.
Chris H: The customer-facing would be z/OS Connect?
Bruce: You’ve got it; correct. And I’d also like to further elaborate…I talked a little bit about API management from Zowe™ on z/OS. Keep in mind that this is just for a z/OS set of staff and programmers. The API management that we have in Zowe™ is good, but it is not intended to be what I call an enterprise API manager. For that, IBM has a product called API Connect, that I view as being five or six years ahead of where we are with Zowe™. So keep in mind; use the right tool for the right job. The Zowe™ API management is for the z/OS community and
something like API Connect is for the enterprise. So keep that in mind.
Chris S: Nicely done. So the next question on my list is ‘How is the Zowe™ open source project organized?’
Bruce: We’ve already mentioned a little bit about the ZLC. A lot of people say ‘Oh, so it’s one big blob of people working on stuff.’ That is not quite the case. We have what we refer to as ‘squads’. It’s teams of people. The Zowe™ leadership committee are kind of overseers; we are not dictators. The squads are free to do what they think is best for their area. I mentioned the command line interface. There’s a squad that focuses on command line. I mentioned the browser and we refer to it as the application framework, because that’s what it is. There’s a squad that focuses on that. We actually have one squad referred to as the onboarding squad. These people are committed to working to bring either customers or ISV’s on board to Zowe™. To get them started, to point out what we’re doing and where to go for information. So think of it as a collection of squads or ‘guilds.’ It’s a bunch of people and the ZLC is kind of ‘Chief Cat Herder.’
And let’s say that somebody wants to donate new technology that takes Zowe™ in a whole different direction than what we’re currently on. Mobile is an example of that. Well, we expect them to bring on their squad and a set of committers – which is another key concept. You have people whose job is to maintain the source code and in these squads there are the committers. Anybody can join a squad, but you have to earn the right to be a committer, because you have the authority to change the source code. So I fully hope and expect that we have more squads in the future and we take Zowe™ where the open community wants to take it. There’s technology out there that we probably haven’t even thought of yet. Bring it on. We want Zowe™ to grow in that way and not limit it to the three or four technologies that I’ve described today.
Chris S: What’s it like to be Chief Cat Herder for the ZLC? (all laugh)
Chris S: I’m a big fan of Formula One racing and I’d like to compare the features themselves, what each squad is responsible for, to the car. The committer would be the driver, and then the remaining squad would be that massive support team that each race crew has. Tons of people that are monitoring tire pressure, rebuilding engines between runs, making sure that all the telemetry is good, they’re checking hydraulics, ready to change the tires when necessary. But the committers are responsible for the win. What do you think?
Bruce: It’s an analogy that we could draw some parallels to. The thing that I would point out is that the open community is staffed technically by volunteers.
Chris H: So it’s more like a pinewood derby then?
Bruce: This is what makes the open community fun. So you ask about what it’s like to cat herd? These are willing cats. So that’s one advantage. But, technically they are volunteers. The three founding members – IBM, Rocket and CA Technologies/Broadcom – have volunteered their people to work on this. But there can be challenges when Squad A needs something from Squad B, and you’re trying to close on getting a release out the door, for example. We have a great track record so far, of getting releases out the door. But a given feature or function might slip a release. But if you’re putting a release out every 30 days, that’s not the end of the world. So I guess that’s another point that I would make in this. Yes, it’s volunteers, but we are living and breathing agile development process. We have, in effect, trains that we are building code on, and we put something out once a month. Whether you get everything you want in a given month, if it misses that train, it catches the next one.
Just a little bit of commentary. This is a weird world for people that grew up on z/OS and mainframes. “What? You’re putting something out and you can’t necessarily tell me exactly when widget ABC is going to be complete and regression test and all that kind of stuff?” So this is a cultural thing as much as anything else. I have learned from the open community and I would like to think the open community has learned something about how the mainframe ‘runs the civilized world’ as I like to put it. How those two worlds come together. Because the mainframes need to get more agile and integrated. But I’ll also say that there are still some things this old dog can teach the millennials about how important it is to do regression testing and worry about breaking changes and so on.
Chris S: One thing that struck us from our previous guest is something that I’ve been hearing more often. Is that we want to, while doing that, make the Mainframe less remarkable. Which touches sort of on your discussion about how Zowe™ and it being plumbing and infrastructure, means that you should be able to touch different platforms and different types of things, without really noticing it up front. In other words, it doesn’t matter what it is. So what do you think about making the Mainframe more modern, bringing in more features, more APIs, this and that, but still making it less remarkable? What’s your opinion on that?
Bruce: On one level I completely agree. But as we’ve also talked about, there’s 50 years of history there. There’s stuff in that box that we have made remarkable in those 50 years. So I don’t want that to be overshadowed by this message that it’s ‘just another platform.’ You know, I’ll pick on security. The platform has a huge story with the encryption capabilities that we have and that kind of stuff. At the beginning of the conversation I would say, yes, that REST API looks like any other platform. But we also need to get to the point that that REST API may be exploiting technology that no other platform has. You almost need a way to say ‘my API has that baked into it,’ and you get that ‘for free.’ Now, there’s no such thing as free. But, if I have an API where I could store my data on platform one or platform two; I need a way in my REST API to say and my value on the mainframe is that that data isn’t going to be hacked…even if you’ve got some bad guy trying to use a quantum computer to do it. Right? To me, we haven’t figured that out yet. In my mind, my REST API catalog should have a piece of metadata or something that says ‘yeah, but I’ve got this value-add,’ so think about whether you want to make use of my services or not.
Chris S: Interesting.
Chris H: So one of the things Bruce and I were talking about as we were walking over here,
was that I just saw this Diego Bravo article about ‘VS (Visual Studio) code and Zowe™ for z/OS systems programmers.’ That’s about as far as I got with the article, because I just I literally just saw it this morning. I thought, Bruce, maybe you know something about this or if you could talk a little bit about why it’s important.
Bruce: The VS code story is a good example of this convergence of open source and making use of the mainframe. In my mind, VS code, open source, IDE, blah blah blah, are used by a large install base out there. And it was actually started by the command line interface team; writing an extension to VS code (it’s available in market place; you can Google it), that in effect is making use of, I’ll say a CLI infrastructure – just to oversimplify it – working with the z/OS based REST APIs. And guess what? In the Visual Studio tree structure, over to the side, you can have z/OS based source files, data, whatever, show up in that tree just like anything else. So it’s an example of kind of the equalizer of now that VS code developer can go look at and submit and do things that five years ago would have been alien. Like you have a Microsoft IDE that’s talking to the mainframe? That’s kind of weird. So we’re bringing those two worlds together. That’s kind of an over-simplification of what it is, but that’s also a message on its significance of bringing these two worlds together.
Chris H: Is the web interface called zLUX?
Chris S: Another name…
Bruce: I think we have settled upon Application Framework.
Chris S: Not zLUX at all.
Bruce: I’m trying to remember. zLUX was an acronym. It might have been lightweight user experience or something like that. Which, the story was – I just point a browser to my mainframe and it gets populated with a bunch of web apps there. But I think we went to App Framework because we could provide web apps that run in there, ISV’s could provide web apps that run in there and customers can provide web apps that run in there. One of the narratives that we tell about this is; launch in context between the various web apps. So we’re not just wanting to put a bunch of icons on a browser. We want to break down the silos that have grown up over the last 50 years, so that from one app, I could launch in context to another app. That’s part of what this environment gives you. And that’s much more App Framework concept than it is zLUX concept. It’s the subtleties in the naming that you need podcasts like this to explain what the heck is really going on.
Chris H: Ok, so we talked about guilds, squads, Wikipedia, memes, the ZLC, committers…so
how can I join if I want to get in on this? I want to get in on the ground floor.
Bruce: Great question. First of all, just go to Zowe.org. Joining means different things to different people. So out there are some simple tutorials; just to understand what this technology is and what it’s about. So that’s kind of step one. I’ve described some of the lingo that were using of having squads or whatever. If you go to Zowe.org, IBM has put an instance of Zowe™ up on Z Trial – if your audience is familiar with Z Trial? And that is a test drive. It’s a scripted, kind of get your feet wet type of stuff.
Chris H: That’s pretty cool that they have access to that.
Chris S: Yes, I’ve actually worked on a couple of the different Z Trial experiences and you described it exactly right. It’s basically kicking the tires. It’s essentially an all inclusive remote desktop image. An environment. All products configure. A little test script. You can walk through it and see what the product looks and feels like without having to commit.
Chris H: That’s cool because, like, what are your alternatives to accessing the mainframe?
Bruce: So we could probably do a whole podcast just on this. There are some tutorials that are sponsored by the same organization that leads Master the Mainframe. Somewhere on Zowe.org there is a link to a tutorial if you want to learn how to write a web app with this launch in context capability that I just described. That’s one way to get involved. Another way is…you know, we’ve talked about open-source, but the source is just kind of the necessary evil. What you really want, is an executable of this compiled source that does something. And we refer to that as a convenience build. So when I talk about putting a release out every month; we’re updating the open source, but what were delivering is a downloadable executable for you to play with.
So you can download that. You can install it. And then this is where you’ll discover…how do I write an extension to the CLI? how do I write a web app? how do I get APIs populated in the API catalog? Those are things that you can do in your own shop – assuming you got a dev test environment where you can play with this stuff. So that, in my mind, is kind of the next degree of getting involved.
Now we talked a little bit about committers; but all committers are doing is submitting the changes. We need testers, we need reviewers of documentation, we need people to help debug – even though they may not actually change the source code. If we had someone to research a given bug and do a pull request in Github – all of this is in the open, we’re using GitHub and I mentioned we’re following agile principles. You can earn becoming a committer by learning the code and helping the existing committers do the right thing going forward. Customers can get involved. I’m also very encouraged that we’ve got universities that are wanting to teach Zowe™. So the idea is that people coming out of school would have some Zowe™ knowledge. For people in universities – go to Z trial. There is a way to earn a Zowe™ badge. Put that on your resume and it will help you get attention, because you know the technology and how it interacts with z/OS.
Chris H: What are the colleges?
Bruce: I don’t have them memorized. I don’t want to say because I don’t want to get the names wrong. We have a great working relationship with Marist. Definitely Marist, and we’re beginning a conversation with Virginia Commonwealth University – we’ve got a strong advocate there. Now when I say Marist College – they are actually delivering our build environment. Keep in mind that this is not an IBM thing, a Rocket thing or a Broadcom thing. This is an open community and we try, as best we can, to be vendor-neutral. Even Z Trial… is IBM is donating infrastructure for people to be able to test drive this. But I’d really like Z Trial to be running at a university somewhere, that is vendor-neutral, because that’s the philosophy of open software. Again, this is where we’re kind of finding that balance between a long-standing proprietary platform and trying to do this in an open community. The Z Linux organization jumped out ahead by providing infrastructure for the community to play with. And there’s my mission to take over the world.
Chris H: I think that you summed it up. That’s a great the great ending. Do you have any words of advice or wisdom that you want to share with the listeners? It can be generationally agnostic and doesn’t have to be limited to millennials.
Bruce: I want Zowe™ to be driven by more than the three founding companies. We’ve had other
software vendors get involved – Phoenix Software was one that has provided REST APIs. I want to give a shout out to them. But we’re kind of in the first year – depending upon how you want to count it. This February will be the one year anniversary of Zowe™ being 100% open source – so for anybody to be able to get involved. I need participants. Whether you’re a customer, a software vendor, a system integrator, a university. I want a movement here. Come join the party. Figure out what we’re doing. Tell us if it meets or doesn’t meet what you need. If it doesn’t meet what you need – come join and take us in the direction you want it to go. So I guess that’s my message. We’ve used the term product a couple times, but this is not a product in the same way that Linux was not a product when Linus was putting all this code together. It turns into a product when somebody takes that code and says “I’m going to provide support for this; call me and blah blah blah.” We’re at that stage where we’re not a product – we’re a movement. And I need more people to join the parade. So there’s my advertisement.
Chris S: Bruce, thank you so much for being on. Everybody go check out a Zowe.org. As Bruce mentioned; when you go and navigate through the site, you’ll see links to Z Trial, and Github and Slack channels. You can find the experts that are out there on the slack channels. And that’s where you’ll find your answers and be able to contribute to the community.
Bruce: Thanks so much for having me, and in another 6 months we’ll come back and do this again and check in on the movement.