 Welcome to Mainframe Matters, and today we have with us Louisa Sears, Chair of the Galassa Technical History Committee for Open Mainframe Project. Louisa, it's great to have you on the show. Thank you. Yeah, and as today, we are going to talk about Galassa Project. I want to understand, first of all, it's a new project. Talk a bit about what is this project about, what problem is trying to solve, and also if you can tell me what is the story behind the name. Okay, so the name is probably the easiest place to start. So Grace Hopper is, I believe, she's the kind of founder of Kobol, and she came up with the fact, the name The Bug, because she came up with the first bug, I believe. And so when they were putting together the Galassa name, I believe it's the latter name for a moth. It's, if you Google Galassa Moth, then some pictures will come up. And so at the time, the team were looking for a name that looked a bit weird, looked a bit different. And so there's a loose connection with Grace Hopper, which is quite cool. It's about four years old, but we are just releasing it, of course, into the Open Mainframe Project. So at a kind of a high level, the problem we're trying to solve is the manual testing that users and testers have to go through every day on the mainframe, because ZOS is quite a unique platform to kind of develop and push through changes. So at the moment, from my understanding, testers will have to go through many manual steps. So for instance, if they were creating a new banking account on the system, then you would have to log on to a web interface. You would then have to go into the back end, for example, Kix or IMS, and then into a database at the back. And at the moment, it's very difficult to make those changes through testing it end to end. So at the moment, a tester would go in and look at their spreadsheet of individual kind of keyboard inputs into a 3270. And they would record those screenshots. Test results would be everywhere. And it would kind of be organized chaos is how I would probably describe it. So Galassa is an attempt to automate that, make it seamless, and make it really easy for testers to kind of take back control of that ability to make changes. And so you're able to run that test that go from the web interface all the way to the back end and back to the web interface again in one automated kind of suite, which is really nice. And that's where Galassa kind of fits into the kind of new mainframe modernization agenda. And as you're saying, the project has been around for four years. So my question to you is that, was it an internal IBM project? And what made you folks release this into open source? And also if I'm not wrong, it will be officially announced at the open mainframe summit, which is co-hosted with the IBM exchange event. So it was originally developed to solve the testing conundrum within KIX, the customer information control system within IBM itself. So its own development problems. And they have their own testing facility. But the challenge they were having was exactly as I was describing. The ability to do the end-to-end integration tests were becoming much more difficult. And at the time, we have kind of yearly forums where customers can come and talk about their issues and talk about things that are going on. And one of the big questions that they always asked was, how do you test these mainframe applications? How do you do it internally? And we kind of would tell them what was going on and how we would fix it. And then they would say, great, can we have that solution? And that ended up being Galassa. So the way that it kind of evolved over time into the open source world, I think it's been open source for about two years. And the way that it kind of operates in terms of being able to extend Galassa, which is where the open source really comes into play, is that you have something called managers, which are these kind of plug-ins to Galassa, where you can plug in the KIX or the IMS or if there was something from another company, you would be able to create your own. For example, Azure would be a good example of that. You'd be able to plug in the ability to talk to Azure all within the same framework, within the same test. So the open mainframe project was a way for us to take it from its inception as an ability to do tests on internal IBM products. And then it branched that out into other cloud providers. And some users, some testers have very unique tools. And so Galassa is hoping to be able to accommodate to all of those different types of applications that you might need for an integration end-to-end test. When we look at open source project or open source communities, oftentimes it happens that there are projects which might overlap the functionality. And then sometimes there are gaps that you're trying to fill. If you look at open mainframe project, of course, they're not specifically for testing like cobalt check is there, test CI is there, which do the whole deployment, but it's also the testing. So when we look at this project, can you talk about where does it fit in, where it's like, we are kind of building the whole repository of the project, which complement each other very well? So there's a space, there's a couple of angles I could take here. The first one being that when I joined the mainframe community, I joined from kind of distributed world. I worked in cloud. I'd worked in open source previously, but it was all in the cloud forum. And so when I first came into Glassa, I thought, great, there's a way for me to test and access the mainframe without knowing the mainframe. And that was my major hook-in into the project, really. And that's what I love so much about it. And what I love is that Zoe takes very similar valleys and kind of vision for the mainframe. And I think Glassa and Zoe are definitely a potential roadmap item. I think because it holds the same type of mainframe modernization story, looking at how we can get... So for instance, if you were in the cloud team or you're a cloud tester and you wanted to be able to access the mainframe in order to run your integrated tests in your automated pipeline, then you don't have to know intimate details about how to use a 3270 in order to execute those tests. So I think there's a really close alignment with kind of the valleys of the open mainframe project through modernization and accessing from a space of little knowledge of the mainframe. The other project I mentioned would definitely be CableCheck. That's around specific testing of... I think it's unit testing of Cable. So Glassa can definitely look to extend that into CableCheck. So those managers that I talked about previously, that would be a good extension. So you would have a CableCheck manager that would be able to handle specific CableCheck tests. So really good integration points into the other projects in the open mainframe project, which we're really excited and it's hopefully on the roadmap for future implementation. What kind of community that is already here or what kind of community you are looking at building around this project? Of course, because we sort of had an open source variation of this before officially announcing this as part of the open mainframe project. We have garnered some interest already, which is great. We've had some users, some customers looking at POCs with Glassa. So as a baseline, we're in a really good position to kind of start building the rest of the community. At the moment, we have started, obviously I'm chair of the Technical Steering Committee, so we started that as a monthly meeting that we get together. There's currently four different companies that are involved in that, looking at the roadmap, looking at the blockers, kind of what a development doing day to day. We already have a development team running as part of IBM, and we've now open sourced that. So the scrum meetings three times a week, and then we have an iteration planning that we do every two weeks to look at storyboarding, look at what we're trying to aim for. And the, well, what's really exciting is that now that we're being announced, now that it's officially out there, we can publicize those links more effectively, hopefully through the open mainframe project website. So it's not just a closed room that only the IBMers can access. It is now a full official open source project. So I'm hoping to gain more of a conversation around testing. I think that's number one. I think the second is around, well, how can other companies, how can other users extend it? So I talked about Azure. We've got some users that have built an Azure manager. So how can we get that out into the public domain? And yeah, how do we get access to those people that don't necessarily want to play with a mainframe at its core at the nuts and bolts level? How can we get access to that from a kind of a higher level in the testing that they're doing? I think that's the main idea for us growing this community. It's around extending that platform with lots of tentacles. And earlier, you were also mentioning that there are other companies which are also involved. Can you talk about who are the other contributors who are contributing to this project? We run our public technical steering committee meetings at the moment and involved, we've got Broadcom. Macro4 have given some really great talks on Galassa through GSE Nordics and we're hoping that Michelle will come back and do another talk at GSE UK about their use of Galassa. And finally, Rabobank are on the technical steering committee. The project has been around for a while, but it's kind of new for the open management project and the committee. Can you also talk about what kind of release plan do you have for this project? We're currently, I believe, on version 0.29. So the team have been fairly kind of baby steps around making sure that when we do release our first version, official version one, that it's the right time. And I think we've been waiting for adoption by the open mainframe project to really get that going. So I said before that we have weekly scrums and we also have bi-weekly iteration meetings. That's probably fortnightly iteration meetings. And so that forms the basis of our road map. So in that, we've got, at the moment, our biggest work is around the CLI. So Galassa at its inception was accessed via Eclipse and that worked quite well for the first time being because the internal kicks team used Eclipse. However, now we're moving on to the CLI terminal-based move and that's partly based on some of the conversation we had earlier about mainframe modernization accessing the mainframe without knowing it. So being able to access Galassa through the CLI is one of the really big development points we're making at the moment. I think long-term we have some moves in terms of how we can interact with other managers. So making sure that that ecosystem of applications you can interact with are growing. But we do have a full road map between now and I think the end of 2024, we at least have items in that we're looking at doing. Authentication being one and looking at the Cable check in Zoe is something kind of for future that we're looking at. We do go into more detail in the technical steering committee meetings of course but at the moment they there are quite granular levels of roadmap items that I don't have to hand. Can you also talk about who would you like to see to get involved with this project and how they can get involved? So I would love to see some more testers involved. I'd love to see how we can really embrace the product, how we can, or project sorry, and how we can really include more usable features within the CLI. So for example at the moment we screen capture all the way through that test plan. So from the web interface all the way through to the 37270 we're able to make the screenshots of every time we make a change. So I think there's a really great test bed for all of the users of testing platforms, people that are currently using manual spreadsheets. I'd love to see them get involved and see how they can improve Galassa, see if they can find anything that would improve it from a usability perspective. I think the other great opportunity we have is around the cloud providers. There are lots and lots of different clouds out there. Digital Ocean, AWS, Azure, there's lots you can mention. And I'd like to see how other companies, other users are extending Galassa into those other applications. So they're the kind of main few that we could start with, but there are many more cases. For example, if you're a company that are doing internal testing, it's other ways that we could help with Galassa. So I think they're the kind of main things to start with. And then if you want to get involved, there are a couple of places you could do that. You could speak to one of the team directly on the Open Mainframe Project Slack channel. There's the Galassa website. So the website itself allows you to, it's got documentation to sign up to look at the CLI. You can download that. And then, of course, the GitHub page, Galassa-dev. And that's where you have all the code base to browse through, to have a look at, and contribute back if you feel like that's something you want to do. Louisa, thank you so much for taking time out today and talk about this project. And I would love to chat with you again whenever there is an update to the project. Thank you. Thank you so much.