Evolution of Open Source in Industrial Automation

In this episode, Eclipse Foundation Executive Director Mike Milinkovich joins us for a discussion all about open source development. We discuss mastering the art of open source to complete digital transformation, the requirements to foster successful software collaboration, providing the infrastructure to enable development in the community, as well as some common misconceptions surrounding open source. Mike also shares his predictions for 2021, and discusses the trends that are having the biggest impact on developers including IoT solutions, edge computing, privacy, and AI. Plus, learn more about the Sparkplug working group, open hardware, and how the community can get involved.

“There are no companies on the planet that can build a complete end-to-end solution and innovate fast enough to keep up with what is happening through the open and collaborative innovations in open source.” – Mike

Bio:

Mike Milinkovich has been involved in the software industry for over thirty years, doing everything from software engineering, to product management to IP licensing. He has been the Executive Director of the Eclipse Foundation since 2004. In that role he is responsible for supporting both the Eclipse open-source community and its commercial ecosystem.  Prior to joining Eclipse, Mike was a vice president in Oracle’s development group. Other stops along the way include WebGain, The Object People, IBM, Object Technology International (OTI) and Nortel. Mike is a past member of the Executive Committee of the Java Community Process (JCP), and the Boards of the Open Source Initiative (OSI) and the OpenJDK community.  

Episode Transcript:

00:39
Don: Well, hi everyone. This is Don Pearson with Inductive Automation, and really pleased to have Mike with us today. Mike, thanks a lot for taking the time to join us, and I'm gonna basically start by just asking you to give a little introduction to the Eclipse Foundation and a little introduction to yourself, and how it got started and you got started with it.

00:56
Mike: Thanks so much, Don, it's great to be here. Thanks for the opportunity. Yeah, so a little bit on the history of the Eclipse Foundation. The original Eclipse project was started by IBM way back in 2001, and the Eclipse Foundation was created as an independent not-for-profit in 2004. So we're coming up on 17 years of life as an open source foundation, and really what's happened over those years is we've changed and morphed quite substantially from very much focused on a single project, that being the original Eclipse IDE project, to a broad community that has over 375 projects in a huge variety of areas. I'm thinking, IoT and edge computing, cloud-based tools, Jakarta EE and Cloud Native Java, MicroProfile, Automotive. So just a very, very wide range of technologies that have found a home within the Eclipse community.

01:53
Don: Just a question off the side of that about you personally. I know that you mentioned IBM in the process of the beginning and involvement there, and I happen to know a little bit about your background. But for the people listening, what's your background and how the heck did you get involved and take on the role of executive director? And just fill us in on you a little bit here, Mike.

02:12
Mike: Yeah, sure. So the first thing to know is I've been the executive director for the entire history of the Eclipse Foundation, and one thing about that is when I first started this, I certainly never intended to be sticking around for 17 years. But the job keeps changing, and so it's not that I've been doing the same thing for 17 years. It's been a fascinating journey. So I come from a background of business and technology. So I was a hardcore software engineer for a bunch of years. I also happened to be an accountant for a bunch of years, and I had worked... I'd been employee #10 at a startup that was eventually acquired by IBM, called Object Technology International, and I was one of the executives that helped sell it to IBM way back in 1996. And that team within IBM was the team that created the Eclipse IDE.

02:58
Mike: So when IBM and its partners at the time, HP, SAP, Intel, decided to recruit an executive director, I joked that I pattern-matched my way into the job. They wanted somebody who knew how to run a business, they wanted somebody that understood the technology, ideally somebody who had relationships with the development team behind the Eclipse projects, and somebody who didn't work at IBM. And at that time I was a vice-president at Oracle. And so like I said, I sort of pattern-matched my way into the job, sort of a right person at the right time for the right opportunity. And it's just like I said, it's just been an absolutely fascinating journey to be in open source since 2004.

03:40
Don: Yeah, that's a big 20-year history in terms of the evolution of open source. And yeah, it does sound you kind of pattern-matched. You got the accounting background, the technology background. Because you're putting together a foundation like this, tremendous amount of business challenge also, because you're dealing with a lot of the organizations, individuals who just come in involved in a project, involved overall, but you gotta keep the whole machine going. It's evolved quite a bit to that level of projects from your meager roots, if you could say that.

0:04:09
Mike: You know, I think if I was to say one characteristic of me that does well in a role like this, is I am very much a technologist at heart; I love technology. But at the same time because of the business background I've got, I feel like I can talk to a senior executive in a company, but I can also talk with a developer about the challenges that they have and have an honest conversation about how fascinating the stuff that they're building really is. I sometimes think of myself as sort of a translator, because I can listen to a lot of different stakeholders and help communicate with other parts of the community about what their interests are and why they, as a stakeholder, have a role to play in the community.

04:52
Don: Sure. I mean, absolutely. It's a pretty important role to be that sort of bridge, if you will, between all these different viewpoints. This is just a side personal note. Where the heck do you live? Where do you live? Where are you talking from? Where are you now?

05:03
Mike: I'm in snowy Ottawa, Ontario, Canada. And by the way, that's where the OTI, the Object Technology International headquarters were, and that's where the sort of... The original core team for the Eclipse IDE was based here in Ottawa. So that's a little bit more of the history. Yeah, but I've always lived in Ottawa. I got a brother in California and a sister in Texas, but I never strayed too far from home.

05:25
Don: I actually loved Ottawa in the times I went there. On a previous job, I was involved in public sector work as a publisher, and we did events. And of course, Ottawa being the capital town. And when I was a jogger, I remember jogging along Rideau Canal and seeing people ice skating to work in the morning on the canal. And it just struck me that you have a beautiful capital building there. You see these folks just skating to work, put their skates off and go to work, and then skate home at the end of the day. You have a really nice beautiful environment up there to live.

05:53
Mike: Yup. It's true.

05:54
Don: Now, about the Eclipse Foundation. I was speaking of where to live. I heard rumors in that their headquarters moved to Belgium. Did that happen in 2020? Was this move and why this move?

06:07
Mike: So we haven't completed the move yet, but we're still in transition, but we have incorporated Eclipse Foundation AISPL, which is a French acronym for a Belgian international non-profit association. And so, yes, the rumors are true. We are moving the legal domicile of the Eclipse Foundation from being a Delaware US 501c6 not-for-profit to being a Belgian international non-profit. And the reason why is pretty simple, is we just saw a real business opportunity for our organization and our community. If you think of the other prominent open source foundations that are out there, Linux, Apache, Mozilla, they're all U.S. based, in fact, they're all California based, and there is a lot of interest in Europe around open source, and when we sort of started, this all came out of an analysis that we did where one of our board members said, "Hey, you guys are doing quite well in Europe. I challenge you to come up with a strategy to do even better."

07:05
Mike: And we did an analysis, we found out that 70% of our paying members and 70% of the committers who work on our projects were based in Europe. So over time, we had basically become a European organization. And at the same time, there's a number of different strategies from the European Commission and national governments in Europe, thinking of digital sovereignty, citizen privacy, ethical AI and the likes, where we felt that it was inevitable that Europe was going to create a prominent open source foundation of international reach and reputation. And so, if we didn't sort of seize that opportunity, we were gonna be inevitably competing with an incumbent. And so it seemed to be an opportunity for us to jump in and fill that requirement.

07:54
Don: As you describe that, Mike, it makes total sense, and European area has had provided a lot of leadership in the world of open source, and so actually just that analysis, 70% of paying members, 70% of committers, it makes total sense for you to have a little center of gravity leaning towards Europe a little bit, and a nice place in Belgium. How about another side question. EclipseCon 2020, how did it go? Give us a little update.

08:18
Mike: It went really well. Yeah, it was a... Because of the pandemic, we, like a lot of other organizations, had to switch our event strategy from in-person events to virtual events. And so it was our first foray into doing a large-scale virtual event, and we're super happy with the results. We had over 2500 registrations, over 1300 attendees, and folks gave us a lot of positive feedback on the quality of the talks, and the platform that we used for the event stood up pretty well. And so we learned a lot and we're planning on doing another virtual event again this year, because we just made the decision that it was gonna be less risk for our community and our organization to plan for a virtual event for 2021. So that'll be coming up in October of this year. I suspect that one of the things that's going to change for quite a while as a result of the pandemic is that, I think, events are gonna come back in more of a hybrid model and sort of the assembling thousands of thousands of people into a single place is... It's going to be a rare thing for a few years, I think.

09:27
Don: I think you're right. We've had conversations, we did a similar thing with the Ignition Community Conference this year. I know you'd been there and spoke and participated in the past, and we went totally virtual and the jury's still out with our planning team that I'm involved with, because we're looking at the hybrid that you're talking about. Is it too soon to hybrid this year? We may end up being, as currently planned, totally virtual. And I think you're right, I think the gathering group will probably be smaller, and the extended family, if you will, of attendees will be larger as was the case with our conference. Let's switch gears just a little bit, talk about industry trends, because you recently did a blog about sort of the biggest trends in IoT, Edge, AI development, those things that are impacting developers. So what were some of the 2020 IoT developer survey results regarding the IoT-based concerns and solutions? Kinda give us a little highlight of that survey.

10:21
Mike: There was a couple of key takeaways... Actually, let's start off with sort of the things that didn't change too much. And the kinds of things I'm thinking there is that security, communications, encryption and so on, still remain sort of one of the top of mind concerns for developers in that area. But then I think a couple of things that I think were more interesting findings, 'cause that's been sort of consistent since almost as long as we've been doing the survey, is that if we're really starting to see AI as being the edge computing workloads that's, use case that's driving demand and driving deployments on the edge. I think, when we first started talking about edge computing a couple of years back, there was some debate over what would be the workloads that were gonna really drive demand for edge computing, and it really seems that AI is emerging as the workload that's gonna be driving edge computing forward. So that was certainly one thing. The other thing that came out was, and this was honestly a little bit of a surprise to me personally, was that we asked what are the sort of vertical domains that you're primarily focusing on, and industrial automation has been sort of perennially in the top three or four. For a number of years, smart buildings was #1.

11:42
Mike: But in 2020, agriculture emerged as the #1 industry focus area for the respondents in the survey, which I personally found very interesting. That was one thing that came out of the survey. And then I guess privacy is a growing concern, and I think is another one, and perhaps related, is that we're seeing more interest in distributed ledgers as a way to secure IoT solutions. So those are a number of the key findings from the survey and I just wanted then to thank you for all the folks that responded. And it's been a while since we actually did a data gathering for this, and it's only gonna be... I think, four or five months from now, we're gonna be starting this whole annual cycle all over again for the 2021 survey.

12:25
Don: That's a big endeavor every year, I know that for sure. So that's interesting that AI use case is driving deployment at the edge. Anything else, any other comments on the edge computing technologies that you wanted to comment on? Or does that cover it?

12:41
Mike: No, for me, edge computing is a fascinating area. Maybe just talk a little bit about how I differentiate IoT from edge, because the two terms almost get used synonymously sometimes, and in many cases, the architecture diagrams and the use cases look very familiar, but when people talk about IoT, I find that generally speaking, they're folks that are coming from an embedded software background or an industrial automation background, and they think of IoT as a way to move forward into a more open and more multi-vendor or vendor-neutral kind of scenario for, first of all, remodeling issues and scaling what had previously been the very siloed and proprietary solutions that they've had for years. The folks that talk about edge computing are coming from a cloud computing background, so really folks that are working on edge computing are thinking about, "Okay, we've built all of this great distributed computing technology for cloud, and how can we push it to the edge?"

13:44
Mike: And then the main difference between what's an edge compute problem versus a cloud compute problem is really, do you know or do you care where it is? If it's cloud computing, by definition, you don't care where the server is, but if it's an edge compute, then almost certainly you do, because you're designing for reducing latency, you're designing for having a camera in a very specific place, so if location matters to you, it's not so much about how big the server is, what the software stack is, it's do you know where it is, or do you care where it is. Anyway, that's how I differentiate IoT from edge, and then I think... I do predict that there's gonna be... Edge computing ultimately is gonna unfold to a certain degree, like cloud computing has, where there needs to be a very broadly accepted, broadly adopted edge compute platform that fills the role that Kubernetes took for cloud compute, sort of the major infrastructure substrate that everybody rallies around and builds on top of, and I think the Eclipse Foundation and its Edge Native Working Group with projects like Eclipse ioFog, and Eclipse FogOS are particularly well suited for, for meeting that need for a high-profile, open source platform for edge compute.

15:04
Don: A couple of years ago, you were at our conference, and we appreciate you coming and sharing your thoughts, and I sat in the session, you were on that session, Arlen Nipper was also in that particular panel, I know you've known each other for years, you made a comment, maybe my quote isn't exactly right, but I'll give it to you and correct me if the quote is wrong and comment on it, Mike. You said something like, "Software is eating the world and open source is eating software." I have actually heard Arlen quote you on that at a previous podcast I did with him, and so when you talked at our conference, you talked about that too. What does that mean? Where do you see it going? Share us a little of that perspective.

15:44
Mike: Yeah, so first of all, the quote is accurate, I'm guilty as charged. So what do I mean by that? So, “software is eating the world” comes from a blog post by Mark Andreessen quite a few years ago now, where he was talking about basically all of the investment is floating into software and software is disrupting traditional industries, and this to a certain degree, this whole digitalization or digital transformation that's happening across virtually every vertical industry you can think of is a reflection of “software is eating the world.” But if you look in the software industry, so the people that are furthest along in this software journey or the organizations that are furthest along, all software, all incumbents in the software world have mastered the art of using and engaging with open source and the open source communities. And the reason is really simple, is there's not even Microsoft, there are no companies on the planet that can build a complete end-to-end solution and innovate fast enough to keep up with what is happening through the open and collaborative innovations in open source.

16:50
Mike: Really what open source does is it allows permissionless innovation, so developers can try an experiment with different combinations of components and software at a far more rapid pace than it is possible in a proprietary scenario. So software is eating the world, open source is eating software. Where does this go? So what's happening, I believe, is that there were so many companies out there, and I'm speaking virtually of every enterprise and every industrial out there is talking about, they need to have a digital transformation, they need to... They need a digitalization strategy. What I think most of them haven't yet figured out is, you have not completed your digital transformation until you have mastered the art of open source, and because you can't claim to have digitally transformed your business until you have embraced the open innovation that comes with open source. So it's almost like open source is a means to an end, digitalization basically means become a software-centric organization, and my belief is, is that you cannot become a software-centric organization if your organization is incapable of innovating openly and collaborating openly, which is really driven by open source.

18:11
Don: No, I totally see that, and I stole one of your quotes, innovation without permission, in a keynote presentation I did, when Arlen actually first introduced me to you and the Eclipse Foundation, when he was talking about the evolution of work he was doing and moving it over to Eclipse. I think that innovation without permission is so critical, I know it's... And Inductive Automation has certainly driven a lot of the decisions around the evolution of our software platform, just because we wanted to unleash the capabilities of people across the community to be able to innovate, particularly in our case in the industrial sector. When you talk about open source and open technology, you had said it's kind of a fundamental trend repeating in every sector of human activity and all areas of technology, just as you just described it.

19:00
Don: And if you maybe could go a little more into smart buildings or agriculture, why do you believe some of those things are coming up on the survey as areas where there's going to be a focus? I was fascinated when you say agriculture came out of #1. I went, "Well, I can see some implications for where applications could occur there." Can you just comment a little more about that? I think that's a fascinating area. All the areas of industry where open source and open technologies are having to play.

19:26
Mike: Actually, I'm sort of a mile wide and an inch deep in all of the various industrial sectors that we see open source happening, but let me just maybe talk a little bit about why I think IoT in particular, and for that matter, edge compute will have to be open source. And I think the answer to that question is really simple, it's because the internet is open source. So if you think about the modern world that we live in is an accident in the sense that there were two key pieces of technology, the TCP IP protocol, and HDP HTML protocols and content definition, both of those became completely open technologies, quite independent of one another. One because it was Vint Cerf at RAND built the TCP IP protocol with his colleagues. And that was made open by the DoD to increase its proliferation. And that created the internet, which was a walled garden for defense and research organizations for many, many years until it was opened in the mid-90s.

20:35
Mike: And then in parallel or independently at CERN in Europe. Tim Berners-Lee created HTP and HTML, and that really created the World Wide Web. And CERN had a strict policy that everything they did had to be open. And before there was the internet, there was CompuServe and AOL and things that did similar things, but they could not compete with the openness. And the reason why they can't compete with the openness is because once those technologies are freely adoptable by any organization, the amount of innovation and the rate and scale at which they can be deployed becomes virtually infinite. The business models of Google, Facebook, Twitter, all of the internet giants that we rely on would be impossible without open source. And if you think about the scale of the Internet of Things, about the many billions of devices that people are talking about deploying as part of IoT, there's not a chance that that will ever occur at scale if the main components going into those software solutions are not open source.

21:42
Mike: Now, I wanna make it clear, I don't mean any of this to say that fortunes will not be made, this is not about a lack of business, and there's going to be some wildly successful companies that are gonna build on top of these open source solutions and be incredibly successful. So it's not that when I say that open source is taking over the world, it's not in any way that I'm not against the commercialization or against business, but what I'm saying is the only way you can scale software to the degree that we expect in modern systems and deployments, the only way you can do that is with open source.

22:18
Don: Sure, and I think that brings us to another question though, in terms of the other side, let me just be a devil's advocate for a second, that all makes sense, but why are some industrial automation individuals, companies ... Why are some in this field so skeptical or so negative towards open source? What are some common misconceptions out there that might be barriers or challenges to the open source community and its evolution?

22:43
Mike: So first of all, it's because inertia is easy and change is hard. So that's top of mind in the response, secondly, that they have procurement departments and legal departments that have had certain policies and procedures for many years about how they deal with software, and it's difficult to change those organizations. They are typically some of the most conservative branches of any large enterprise, and so there's a lot of change in process that has to happen there. But I think that as I sort of loop back to what I said before, ultimately, they're all going to have to fit... Or actually, before I get there, one other thing, which is, there is this common misconception that open source is a bunch of folks living in their parents' basement doing this for fun. That's been untrue for at least a decade and a half, and I'm not even sure if it really ever was true. Maybe at the very, very beginning. But if you look at the companies that are participating at the Eclipse Foundation, who's on our board? Well, we have companies like Bosch and IBM and SAP and Oracle. And these are very serious companies that are worldwide in scale, that have come to realize that they need to embrace open source.

24:00
Mike: Ultimately, this misconception is going to sort itself out because what I said before, every one of these companies has to go through a digital transformation, and the ones that fail to digitalize and the ones that fail to adopt open source are ultimately going to die or be acquired by others, because there's just... They're just not going to be able to be successful in the modern economy if they don't make these transitions.

24:26
Don: Yeah, no, absolutely, I think that's critical. You know, I... My first experience at all with getting involved in any kind of anything with Eclipse Foundation came through Arlen Nipper, co-inventor of MQTT, who basically brought it to my attention and said the work that he was doing on the Sparkplug spec with MQTT, was going to be given to or brought over into the Eclipse Foundation for stewardship and development. And then all of a sudden I became part... Our company, Inductive Automation, got involved in the Sparkplug Working Group. We're both big believers in collaboration, and I'm beginning to understand how that works from being involved in the working group, but what would you say are broadly the requirements for... Or building blocks in your experience over the last 17 years to have software collaboration in the community really effective? Just some comments on the success factors would be good.

25:18
Mike: Yeah, so the reason why an organization like The Eclipse Foundation exists is ultimately to enable collaboration, that is what we do. So what goes into that? So the first and foremost is what we aspire to is to provide governance frameworks that will allow everybody from the largest companies, or the people who work for the largest companies, down to individual contributors who are just there for their own personal enjoyment or fun to participate in our projects. And to make that possible, the kinds of things that you have to do is you have to provide, first and foremost, a vendor-neutral governance framework. You have to make sure that no one company, no one person can hold sway over the technical direction of an Eclipse Foundation project. And so that's really job #1.

26:07
Mike: Then there's providing the infrastructure to enable development, so originally we had our own forge, which we still support, but now more recently projects are using GitHub in our instance of GitLab to host their infrastructure, managing intellectual property. So how do you bring in contributions, how do you make sure that the software that's produced by our projects is safe for commercialization? So, intellectual property management. And then, finally, community development, that goes from everything from events to social media, to raising awareness of our projects however we can and do. Those are all aspects of what an organization like the Eclipse Foundation does.

26:48
Don: Since I have a little bit of a bias, being involved in the Sparkplug Working Group, I wanna ask you to take that project and its evolution... Which it's young, you've got... It's young as far as projects inside Eclipse go, but it can use it as an example of how they evolve, how people could get involved in the Eclipse projects and maybe just use that so that our listeners could see how they might participate more, but also an understanding of where the working group for Sparkplug be it actually is and where it's going and what the aspirations are.

27:21
Mike: So Sparkplug is a... First thing we noticed, Sparkplug is actually a specification. So one of the things that we recently got into at Eclipse Foundation by recently, like two, three years ago, is in addition to doing open source projects, is doing open specification development. Sparkplug has a project, is a specification, there's also Eclipse Tahu, which is an implementation of that specification at the Eclipse Foundation. And those two projects really work hand-in-glove together to help support the broader goals of Sparkplug. Maybe just a moment on what Sparkplug is... So in the world of IoT, the MQTT protocol is huge, in many metrics. It's certainly in our IoT developer survey by far the #1used IoT-specific protocol. But MQTT is... One of the good things about it is it doesn't really say anything about the payloads, the bad news about it is it doesn't really say anything about the payloads. What Sparkplug does is provide a definition for the topic structure, the payload structure, sort of an application lifecycle model session management to allow out-of-the-box interoperability for industrial devices based on MQTT.

28:28
Mike: So one analogy that I know you, Don, have used yourself is Sparkplug is the HTML for MQTT's protocol, which is similar to HTTP. So the way HTTP is the transport layer and HTML defines the content and MQTT is the transport layer, and Sparkplug is the content. So the project's been around for a while, and I think it really shows... By the way, one of the things that we aspire to in terms of attracting small companies and large companies. So if you look at the membership of the Sparkplug Working Group, we have companies like Chevron and Intel, Inductive Automation of course as well, and Cirrus Link, which is quite a small company, HiveMQ over in Germany... So we have a nice range of companies and participants that are interested in this going forward and that's... We see the opportunity for Sparkplug is absolutely enormous. We think that its ability to provide out-of-the-box MQTT integration for machinery from multiple vendors in a SCADA environment. That is a problem... That is a multi-billion dollar problem in terms of deploying new smart factories and the like, and we think Sparkplug is an elegant solution to that, and we're looking forward to helping that working group grow over 2021.

29:50
Don: Sure, well, I think we're really aligned on that because we also see the huge multi-billion-dollar problem of making digital transformation a reality, actually making these architectures capable of being deployed, that Sparkplug and its evolution and development are gonna play a key role in that empowerment and that acceleration of the process, because the brownfield world of the industrial sector has and will continue to be a major challenge. We don't live in a greenfield world at all, so you're all restarting with how do we take what we have and migrate it to where we wanna be, and certainly Sparkplug's going to play a significant role in that process. Just as we sort of kinda wrap up here, I wanted to ask a side question. We're talking a lot about open software, but I've also heard you talk about open hardware. I know that Eclipse is a partner in the Open Hardware Group. So for those of us who don't know much about open hardware, can you just maybe take a minute or two and talk about what's happening in that space?

30:51
Mike: Yeah, sure. So the Open Hardware Group is a Canadian-based non-profit that we have a partnership with where we're providing... Their projects are Eclipse projects, and so we're providing them with our development process and all of the services that we were talking about earlier. And what they're working on is the definition of processor designs based on the RISC-V instruction set architecture. So RISC-V is a risk architecture that has taken the world by storm over the last couple of years, because it's a completely open way to build, to do processor designs. But what we're seeing... What's happening in the RISC-V ecosystem is that there is sort of an embarrassment of riches. Everybody is taking this and building their own processors, and there's really no center of gravity around processor designs in the RISC-V ecosystem.

31:45
Mike: And so Open Hardware Group is an attempt to provide some really high-quality designs based in SystemVerilog, that companies can pick up and use for building their own processors. This has the opportunity to make major inroads into lowering the cost and the barriers to entry to processor designs that are currently driven primarily by Arm and Intel down the road, particularly relevant to industrial automation and embedded scenarios. Because this is low cost, because these are risk processors. There's a real opportunity for folks that are doing SoC or MCU design to really pick up and use these processors in their architectures.

32:27
Don: Thank you. I just wanted a little bit on that. Thank you for that introduction. You know, I just kinda wanna wrap up our discussion, maybe with an open-ended question for you. Anything, what's next for open source? Anything you think that's on your mind or that you see in the near horizon that you wanted to share as we wrap it up?

32:46
Mike: Yeah, sure. So 2021 is gonna be continue to be a big year in open source, and because of the pandemic, and because of all the other things that are going on in the world, and you just get on an automotive sector there's sort of carnage in there in terms of the lowering demand and so on. And so that's really forcing people to accelerate plans, so things that were originally so they thought they would have a decade to do, now they're gonna have to do in a couple of years. And I think open source is going to be a very, very big part of that. And so, 2021, put it to a large degree, I think the challenge that we're gonna have at the Eclipse Foundation is keeping up with the demand for our insights and our services, and bringing on new projects and new working groups.

33:33
Mike: And I think that's gonna be across all kinds of industries and sectors that are touched by software and open hardware technologies. So I think that's really what's gonna happen in open source and I think for us at the Eclipse Foundation, continuing, completing the pivot to Europe, is gonna be something that's gonna be consuming a lot of my personal time, because as the Executive Director, there's a lot of moving parts to complete a corporate restructuring like that. But we also have, we think that we've got some exciting new projects coming our ways in areas like cloud computing and AI that I think are gonna be real game-changers for the Eclipse Foundation.

34:14
Don: That's great Mike. It sounds like it's not gonna be slow, you're, your short tenure that started 17 years ago, sounds like it's got a very active future, as the joking saying goes, "Moss don't grow on a rolling stone", so you're gonna keep rolling forward. I just wanna say from those of us at Inductive Automation, we appreciate the opportunity to be involved in the project we're involved in. And also I would say going forward, the work that's being done there, it's so critical to our own strategies and the role that we wanna play with the ignition platform out in the industrial community, that to have the work of the work of the Eclipse Foundation continue to expand and grow like it has been. So with that, I really appreciate you taking the time out today to talk to us. And thanks so much me. Let's have a great year and let's as we talked before we started, let's have 2021 on a really nice up trend and go forward nicely throughout the year, Mike.

35:06
Mike: Yeah, thank you, Don, for the opportunity. It's been a pleasure chatting.

35:10
Don: Great, thanks.