Gael and Thabang are joined by Dirk Riehle, a professor of open source software at the University of Erlangen-Nürnberg, to talk about business models based on open source strategies, and why businesses might choose to go down this route.
Stay connected with the Entrepreneurial Open Source Podcast by following us on Twitter: https://twitter.com/OSS4Biz
Gael and Thabang are joined by Dirk Riehle, a professor of open source software at the University of Erlangen-Nürnberg, to talk about business models based on open source strategies, and why businesses might choose to go down this route.
Stay connected with the Entrepreneurial Open Source Podcast by following us on Twitter: https://twitter.com/OSS4Biz
Thabang Mashologu:
Hello and welcome to the Entrepreneurial Open Source podcast. If you're joining us for the first time, this is a show where we discuss the power and practical value of open source in business and entrepreneurship with experts, leaders and innovators from around the world. I'm Thabang Mashologu.
Gael Blondelle:
And I'm Gael Blondelle.
Thabang Mashologu:
We've got a really great show for you today. Dirk Riehle is joining us. Dirk is a researcher and a professor of open source at the University of Berlin in Nuremberg. He has a particular focus on the role of open source in business and industry at large and had spent a decade working in industry for SAP in addition to his esteemed career in academia. Welcome to the show Dirk.
Dirk Riehle:
Thank you for Thabang, Gael, I'm very happy to be here on your fabulous show, thanks for having me.
Thabang Mashologu:
You're very welcome. Can you start by telling us a bit about yourself and your research?
Dirk Riehle:
I happily will do so. So you said the key parts already. I am a researcher and a professor at a large public university in Germany. For the last 10 years I have been, before that I actually worked in industry for more than 10 years as well, so I combine industry and university. The research we do is indeed on open source, my official title, I don't know why, but in Germany, the professors always need these titles, it's professor of open source software. So in the name is my specialization, as a software engineering researcher, as a business researcher, to focus on open source.
Dirk Riehle:
We have a couple of engineering projects which are open source projects where the research question is usually the engineering questions. So an example is how to make open data useful for the lot that is more an engineering question, the way they go about it, of interest here perhaps is open source business models or strategies, we'll come to that, I assume. And in that space we research commercial opensource, something we call user lab consortium, which is very close to what the Eclipse Foundation calls industry working groups. So these are structures of how to make open source projects work, how to make them satisfied user and developer needs and so forth.
Thabang Mashologu:
Very good, off the top of the show here. Hopefully you can help us resolve a long-standing debate once and for all. Is there such a thing as an open source business model? Is it a business model or a tool?
Dirk Riehle:
There's such a thing called complicated human language. So I would argue or say, like most will, in a narrow sense, no, there's no open source business model itself, but the assumption is you make money because you're selling open source. That obviously doesn't work because by way of the license the open source is free. There's always something complimentary on yourself. So what open source is and we will discuss this in more detail, I assume, it's strategies, it's innovation strategies, it's product management strategies, it's super helpful. It's disruptive to existing incumbents and all of that, but it's a strategy, not really the business model, except that in human language we like to shorten things. So if someone says, "Oh, this is an open source business model." Don't just kill them yet, they're just being efficient in the use of language. What they really want to say usually if they know what they're doing is, "Oh, it's a business model based on open source strategies." Which is not as short and nice and sexy as open source business model, even if at sometimes it ruffles people's feathers.
Thabang Mashologu:
Yeah, no doubt. So you mentioned that one of your research areas is open source industry consortia. Why would companies get involved in such groupings?
Dirk Riehle:
So there are many different reasons. If we go in the direction of the perspective of foundation and consortia, then let's assume that we are talking about something that I call or most people will call community opensource. That's an important distinction from something we may also talk about later, commercial open source. Community open source is really something where people want to collaborate to really have a shared resource, to have that open source software that's useful for everyone and that is not dominated by any one party. So companies or individual people come to that because they find it on the web and it's just high quality software for free. If you're a vendor, just a software developer, you see this fabulous component for free, with a thriving community, why would you go and purchase a closed source alternative instead? It makes no sense.
Dirk Riehle:
So using open source because it gives you high quality open source software for free, makes a lot of sense for vendors or anyone who builds products. Well it also makes a lot of sense, that's another important distinction, for anyone who's a user of software and then you're not using components and your product that you're selling, you really want to use the software itself. So the original users, not coincidentally, well, what's the IT industry now? So the GCC Compiler is an application on some tool of sorts, it's not so much a component. But LibreOffice, OpenOffice are applications that people use on Firefox, et cetera, applications that people use. So open source is not only important for vendors who need components, it's also super important for users of software who want on fully blown applications. So that's how people are using software. Once you're using software, you get onto that highway off also contributing and eventually possibly creating and leading open source projects in such fabulous places like the Eclipse Foundation.
Gael Blondelle:
Of course. If you can come back to the notion of what you call the product management strategy, because I have the impression that sometimes companies that engage in open source and I'm not so clear about their strategy, they miss this point of the product management strategy. I think that's a very good point, so if you could elaborate on that, that would be helpful.
Dirk Riehle:
So now let's assume that somehow the majority of what you're selling is based on open source. So the key inside is it's free. You cannot sell it sustainably, you have to sell something that is complimentary. And so for many years, we know the typical compliments, the first one is or used to be services. So someone uses an open source application, but they feel unsure as something goes wrong, so they will put you... You can be on a retainer or you can provide services, create peace of mind by servicing and successful open source software. I think that's the original business model based on open source, I think Cygnus Solutions did that with the GCC compiler a long time ago, servicing the IT industry and so forth. So consulting and servicing existing community open source software, it's probably the most common business model, it's not necessarily the most attractive business model if you're looking towards a return on investment or making it big, Silicon Valley style big.
Dirk Riehle:
So the next step is, after consulting and services, which is arguably lifestyle in the sense of it pays a good salary and you will enjoy your life and you're doing good, but the next step would be oriented towards really possibly striking it big, meaning creating companies that could possibly be venture capital funded plus all the benefits and disadvantages and where we have, in my book, two dominant models or patterns basically, one, I call single vendor open-source firms and the other ones are the open source distributor firms. So the distributors are pretty well-known, that would be Red Hat and SUSE, what they do for you is give you peace of mind obviously.
Dirk Riehle:
They make a huge set of components that work together well as one large application or infrastructure like an operating system. So the Linux distributors give you that complete set of components, ideally, nicely working together, and so you can sleep well at night and know that your data center is doing well or if something's wrong, you get the back fix and what have you. Do not underestimate the complexity of putting things together in such a way that they work well. So there's significant intellectual property in the configuration. I think people always underestimate that. So why am I saying that? Distributors can be a successful and lucrative business model, and so the question for anyone, investors always, well, what's your moat? What is there that not everyone else can have right away? And with open source distributors, as complicated as it may seem, it's in between the cracks, it's in between the open source components.
Dirk Riehle:
Linux distributor does not own any of the open source components of a Linux distribution, but they have a lot of knowledge, a lot of test speeds, a lot of compatibility databases about how to make things fit together so that they can give you that installer and updater that lets you create a working system and doesn't break it right away. So there's a lot of IP in there, that in my book is or in my knowledge is not necessarily always open when you subscribe to Linux distributors service. And as we Know, move from the Red Hat access to IBM can be a very lucrative business, so that's a very hard business to get into.
Thabang Mashologu:
Right. Now, It's not the most popular of business models in the sense that it's been somewhat controversial over the last few years, but can you touch on the practice of cloud providers offering open source projects as service offerings?
Dirk Riehle:
Yes, happily. So the third category, which we didn't talk about yet, is what I call single vendor open-source firms. So these are companies which develop some software that they then make available as open source software, but they really have a commercial interest. And so they want to be the single, the sole vendor who makes money off it. So they would dislike it if someone else does it and so they can play various tricks or things around intellectual property to make sure that if someone buys center commercial service, they come to them now. So you pointed at one particular commercialization of a successful open source project, which was hosting it in the cloud. So let's go back to my argument. You cannot sell open source, you always sell something complimentary. What that complimentary thing is, requires creativity. So the typical solution today is you've got this fabulous database now, make it available for free, drive adoption in the market by way of a free open source version and then operate it for customers, that's to pay for solution then in the cloud at a higher quality than people could do it themselves.
Dirk Riehle:
My point here is there are other compliments. So consulting is the other early one, but even hardware can be an interesting compliment. So Google made TensorFlow available, but Google is the only vendor who has TensorFlow processing units, specific hardware that makes it cost efficient to run TensorFlow programs. They're then intertwined with Google's cloud offering, but it's really a superior offering for running TensorFlow based software because they have this dedicated hardware, so hardware can also be a compliment to open source software. So you really need to think if you want to build a business around some software that you make available as open source, what that complimentary thing is that your users, the freeloading users, will want to pay for because it gets so much more value out of the open source. And so cloud hosting is the big topic, that's the exciting thing because of, well, all what's going on with licenses, people love to get excited about licenses.
Thabang Mashologu:
Yeah, for sure. Now you mentioned the venture community earlier. Do you have a sense, based on your research, which business models are most exciting and attractive for venture capitalists?
Dirk Riehle:
Anything that makes them recognizably a lot of money. The way they look at it is, in my book and I've done a fair bit of research. I may live in Germany, but I actually lived in Silicon Valley for a long time and I travel regularly back there to interview VCs and what have you and our entrepreneurs to understand where it's currently at. So they will look, still, at key things like team and market size and market opportunity and that size, what's the addressable market, et cetera and of course the larger, the better, so they use traditional venture capital thinking about it. Open source then becomes a disruption strategy if that market is already taken, for example. On or off could you try to compete with Oracle? For a long time, completely impossible to imagine. When new technology came along and databases and the disruptive commercial consult strategy of going to market with a free open source database and getting driving adoption with it and so fast and then now cloud onboarding people into your cloud.
Dirk Riehle:
So I would argue that a startup, a software startup, which has a good product for a large market, if they can show that they are able to have an open source based strategy where they drive adoption of their product with users, not customers yet, but drive adoption just as users, that this will lead to more efficient sales and marketing than if you take a traditional approach. There's a lot of movable parts here. So not everyone wants to go an open source route, you need to have that capability, but I would argue that that's an additional lever, a very powerful lever, to get that disruption on the ground and up and running.
Thabang Mashologu:
We've been almost exclusively focused on the upside. In your research have you come upon some common pitfalls or challenges that open-source based strategies have to have to contend with?
Dirk Riehle:
Yeah. So that disruptive decision to go to market was an open-source version free thrust means you are delaying revenues. Now and then you give people something for free and that makes sense if you can recoup that later in a sense. So you're pushing, deferring that revenue, which makes sense if the market is big enough. So there needs to be an understanding of what you make available for free, for how long, obviously forever, but when do you slow the rate, possibly, of what you make available for free, which is something that can be very upsetting.
Dirk Riehle:
So the commercial intent has to be increased over time as you have a growing in your market, as you are probably not sharing your market, but gain size of a share to increase the pressure or the push to turn those who are potentially customers also to become customers and that can be super upsetting and that's one of the biggest challenges was this it's a product management question. What is it? So in order to sell something, you need to withhold it. Let me take something really, really bad. You have some enterprise software and some new authorization mechanism comes along and everyone wants it and you just require that people pay for it.
Thabang Mashologu:
Do you have any examples that come to mind?
Dirk Riehle:
Mm-hmm (affirmative). Here's a real world example. In enterprise software open-source ERP system, which had this absolutely fabulous idea, that if you upgraded from version two to three, from three to four, you needed to migrate your databases and the scripts for migrating the database were not free. So the pressure on users was you will be stuck on an old and decaying version unless you pay us. So that is very close to blackmailing users and it's obviously a losing proposition if you relate to your community and your users that way. A much better way is to find something that you can recognize as a fair compliment that people understand, "Well, we really have to pay them for that."Which would be the cloud. People understand intuitively if you operate service, that costs money. So if users want to use your servers, they somehow will have to become paying customers. Yes, Gael, you're excited.
Gael Blondelle:
Yeah, because I love this topic. You use words like upsetting, like blackmailing, et cetera and I get the point and also, to some extent, I have the impression that user companies consent to using the open source version and sometime think that, "Oh, well, we may be able to live forever or if maybe we don't want to be a customer, we may fault ourselves." What do you think about that? Do you have some examples of faults by users of this kind of stuff? I think that's really a core topic in our world today.
Dirk Riehle:
Yeah. So this is where people's opinion diverge. So some open source enthusiasts will say, "Open source forever, nothing ever gets withheld." But on the other hand, there's a company which has been developing all of the software and it pays developers, it pays sales and marketing, so they need to earn a paycheck somehow. So they will argue, "It's obvious that sometime, at some point of time in the future, we need to make money. So it is obvious that at some point of time, we will have to close something down or withhold something." So my answer to that is both sides need to look to each other and be a little bit less naive, perhaps. And from the vendor's perspective, of a single vendor open source firms perspective, it becomes an issue of expectation management and being fair and square about your intentions. Which is really hard because you don't want to say understandably, on day one, "We are looking at everyone as potential customers, so show us the goods eventually." So it's really hard to build a community and make that value proposition clear in such a way that people don't feel goaded in any way.
Dirk Riehle:
So one way of doing it is holding your feet to the fire and say, "This is how we will go about future commercialization, this is what we would never do. We will, for example, never relicense to an non open source version or we will do that at some point of time." I mean, you can look at the extreme examples. So in my book, these companies, they are three generations now. The first generation is through, so my SQL was kind of the last one I would argue, sort of. Of the second generation, we still have SugarCRM, which is a fabulous CRM software, the SugarCRM folks, they coined the term commercial opensource.
Dirk Riehle:
So we owe it to them that they came up with the idea, "Oh, we need to give companies, as potential customers, peace of mind." And so they came up with the term "commercial open source." So they did a lot of good things for us, but today the product is not really open source any longer. So founded in 2004, 16 years later, the lifecycle of the product has matured, no point in being open source any longer. I think that's realistic and I think any user needs to understand that there will be changes along the way. The question is, and I think the company will gain the most trust with potential thought leaders and the open source space, who can be the most open and also the most fair in their promises and expectation management and also stick to it? So don't reverse it.
Dirk Riehle:
So when you look at the troubles of elastic or confluent, all of that, who decided that they need to, relicense away from open source, people who are upset about it look at all the forums, looking for any thing that someone, over the internet, said, "We will always be open source." Which is like people, I don't know, I'm not being very realistic, but it is a challenge for the company and it should address it openly and head-on, I propose some sort of open source pledge or some sort of pledge how the commercial tension on need to earn your investors a return on investment, will play out with the open source community and those who use the software and depend on it in the free version.
Gael Blondelle:
Yeah, yeah. Because I think that's another option, because you present it like, "Okay, that will always happen." That the company will withhold some components, et cetera. And from our experience, of course, both Thabang and I, we work for an open source foundation, so I have the impression that's... Oh, I think that's part of our role to also ensure the sustainability. So maybe we are lower in the product stack, but I like to think that what ends up in an open source foundation, whether it's Kubernetes or it's Eclipse platform or Jakarta or whatever, is there forever and in a sustainable way.
Dirk Riehle:
So open source foundations in the overall ecosystem are super important. The role of an open source foundation is to provide that fair playing field to companies who want to engage. It manages the trademarks, it has a good governance that helps people understand how to go about open source, manages intellectual property, there are a lot of centralized services that a foundation can provide that's hard to recreate, a lot of competence that a foundation also brings to the table. So as a consequence, because it brings different companies to the same table or the defined rules, the software that will be developed is not something that will be competitively differentiating for any one company involved. It is clearly community open-source software that you can build products on, but there must be a fairly long way from that one component or even 15 components to a commercial product.
Dirk Riehle:
If it gets too close, you will actually possibly face the infighting of the vendors of the companies involved, trying to keep functionality out of the software. Because if you add too much functionality and it gets too close to a vendor's products that build on it, the vendor will be unhappy. Now they have a very direct need to keep that functionality out. Now you don't like that, because you want good community open source software, so I think good vendors understand that and they understand that the components that you can collaboratively develop this, your competitors and the Eclipse Foundation and other foundations, really are high quality components, good for your product but shared with others and are not competitively differentiating. So they are not going to help you sell more, what they do is they save your costs, they make you conform to standards, one of the open source solution or the component becomes the reference implementation, all of that give you lots of innovation that lifts your boat, but that's not why your customers buy.
Thabang Mashologu:
Yeah. I think you've hit the nail on the head there in terms of community-driven collaboration on non-differentiating components. I mean, I think we see this with infrastructure, I mean, as Gael mentioned, Kubernetes and also platforms such as the Eclipse platform, where people are able to collaborate on the non-differentiating elements and they're able to invest in and monetize those differentiating features on top of that common base of components and infrastructure
Dirk Riehle:
Absolutely, with one caveat, which is where as the industry, I love our industry, it gets complicated.
Thabang Mashologu:
Of course.
Dirk Riehle:
As soon as we joined forces with other companies like Intel and Oracle and IBM joining forces to drive Linux forward or IBM starting the Eclipse Foundation, you get great components and great infrastructure software for free eventually. Users get it for free, you invest if you're sharing the development costs, but on this large level, on this big level, you do it to kick someone's knee or chin, one particular company's chin, which is the people who will dominate exactly that layer that you are just turning into open source. So I think it's no secret if I say companies joined forces to make Linux viable so that they wouldn't have to suffer Microsoft's licensing fees for windows. And IBM started the Eclipse Foundation so that not all developers would be on Visual Studio and C and C# and the Microsoft developer technology. IBM saved the Java ecosystem, if you will.
Dirk Riehle:
And today the Cloud Native Foundation is squarely set against making it even possible to compete with Amazon Web Services, right? So in this last case, there's also the capital investment and the infrastructure and the computers, but you just need software to operate clouds as well. So if there's one company which is willing, which is possibly dominating a key piece, like databases or networking or what have you, then now the software industry found solution to keep that monopolist at bay, which is join forces to develop an open source alternative. That isn't that fabulous, so the software industry learned how to keep monopolists in check this way. Well it was all special purpose situations, but it's just amazing how versatile open sources and how all the costs of paying developers to contribute to something that's freely available, that's public good, makes economic sense, total economic sense, high return on investment for Intel and Oracle and SAP and whoever invested in the various infrastructure layers, where they are not making money, but someone else was threatening basically to eat up all the margins.
Thabang Mashologu:
Right, indeed. So switching gears slightly, Dirk, as you mentioned, you live in Germany, you've spent a lot of time, you lived in the Bay Area and you go back fairly often. What, if any, differences in approaches or fundamentally in the ecosystems, have you identified between the European and North American open source ecosystems?
Dirk Riehle:
So Europe is clearly still behind in some aspects, but it's the internet, people travel. So learning from the playbooks of others, people intently study, not just the competitors, but also anyone from... I was doing similar things. I'm helping a fabulous commercial open source firm right now, which is providing identity services, [RE.SH 00:31:13]. And they Munich based, but they are super in tune how Silicon Valley works and they have to be in order to find their own path and learn from others who have gone some steps before them. So really there's very little delay, I would argue, in terms of what you can do in Germany as you learn it from the Silicon Valley.
Dirk Riehle:
The one big difference still is that the latency or the delay is sometimes tied to people and not everything is written up in blog posts and not everyone can be chatted up at a conference. In Silicon Valley, a huge amount of knowledge still can walk in the door on two feet and you don't quite have that yet in Berlin. So I would argue there's a super intense connection on the one hand, but there's still some generational gap in competencies of people sometimes, but people are learning and so I'm hopeful that all of this really is... I know all of it is a global phenomenon that's not tied to [inaudible 00:32:25].
Thabang Mashologu:
Right. What words of advice would you have to a founder, an entrepreneur who's considering an open source based strategy?
Dirk Riehle:
The piece of advice I have is the one you would give any entrepreneur because open source is just a strategy, just a tool in the tool chest. So the first piece of advice always has to be, get a team, get complimentary skills and find a good market opportunity, iterate a trade, go through the search process and all of that. Within that, since I guess you're asking about open source, yes, I think. Watch out for whether driving adoption by way of an open source version is the right lever for your market and many times it can be, so the playbook of providing free open source and then providing a compliment in the form of a cloud service, that is a standard pattern by now that can work really well, so watch out for whether that pattern fits your type of software, your market and if so, go and run with it.
Thabang Mashologu:
Great advice. Now, as we've disclosed on other episodes, Gael and I are company men, so to speak, as employees of the Eclipse Foundation. What role do foundations play in the ecosystem as it pertains specifically to enabling the success of commercial ventures based on open source?
Dirk Riehle:
Oh, okay. So open source foundations are super important in many ways. So right now you see we've got some of the APR enablers, we provide that level playing field. The companies need us, without us, without a foundation, they would just be in each other's hair. There's some example of foundations for how does Mike Milinkovich, the idea of the Eclipse Foundation, call it shenanigans, that some foundations have played, meaning lots of complicated rules in the bylaws. That's not how to go about it and that's a problem, but if you do it right, like the Eclipse Foundation, yes, you're a fabulous enabler and there's a bright future for a foundation. We talk about commercial open source, but then we talk about community open source, because the foundation really only supports community open source and it used to be vendors who were the companies supporting the foundations. And so the projects are in an open source foundation would be permissively licensed components to be used in the vendor's products. So that makes sense, share the cost of some components that are not competitively differentiating, make it permissive licensing so it doesn't interrupt your business model and so forth.
Dirk Riehle:
Now the big change, what we call user led consortium, and that's the notion that you have the users of software join forces to pay for the development of the applications they need for the business. So the users want applications to operate their business, they're not selling them as products. So this is how then, for example, it's my prime example of course, there's always the German energy distributors who role with the Eclipse Foundation. They joined forces for operating the German energy grid or parts there of, that always complex software. Or you have the automotive manufacturers and the open MDM project, I believe, to develop applications, to manage the test data or the metadata, the measured data and so far.
Dirk Riehle:
So these are applications, not components and those who are paying for its development are the users. And so here's a business opportunity for the foundations and for companies who want to service these applications, because how are the energy distributors in the end, they still want to buy from a company to have peace of mind. So let's look at these two constituencies. So the foundations, you haven't exhausted the software industry, I mean there's still a huge growth, but all of the other industries that need software, that's so much larger than the software industry. Companies that have affinities engineering make the automotive folks, so they are working with you already. And the energy distributors also sounds a lot like engineering, but where are the chemical manufacturing folks? Oh, okay, still engineering. Rather, logistics people. Where are the, I don't know, beauty folks? They all need software.
Dirk Riehle:
Do they like being at the mercy of closed source vendors? No. They have a business to run, they want a software for their needs, ideally it's open source software, so they can switch their supplier if they are unhappy with how the current supplier behaves. So for the foundations, you have huge potential. So as the Eclipse Foundation, it has the construct of industry working group, that's a very nice way of realizing what we call, usually, a lack of consortia, it's the same. So you can do it, can provide the governance, you can help those non-software companies to get the applications they want. At the same time, they still need servicing, so there's a business opportunity for companies to service those applications as well.
Thabang Mashologu:
Yes, certainly.
Gael Blondelle:
That resonate very much with one of my topics of interest at the moment, like I see that we are now talking to some companies that were out very far from being software vendors, but they are all interested in software and now they are also interested in or they understand, at least, that they need to do something about open source and then comes all the topic about... First they may be afraid of open source, sometimes, so I think that maybe collectively, you, me, Thabang and our community at large, we need to help them understand open source and then at some point, in this maturity scare, at some point they may come to the conclusion that, they need an open source program office or a more structured approach to doing open source. What do you think about that? About education aspect and not only for students, but really for people in companies?
Dirk Riehle:
So we did sampling, so we research as we looked around what objects are there to study. We found more than a hundred open source foundations and we have actually substantially more by now. They're just one every day, a random one, so completely random. You would think and you would ask, why are they not joining the Eclipse Foundation? Why are they doing everything from the ground up? And I think the answer, to move on to that, is that education that they need and that they may be not half, so they underestimate the handling of intellectual property, they underestimate the coordination of a community, they underestimate nicely templatized boiler plate, but really well done governance rules and so forth. And so they want to do it themselves and thereby learn it the hard way, how not to do it.
Thabang Mashologu:
So Dirk, what are some of the most interesting topics you and your team are researching right now? What's got you excited?
Dirk Riehle:
I have different ways of excitement in my life, one is tonic our research into startups, so we have, right now, a tool in the making for qualitative data analysis, that's going on, that's curiosity. So it's exciting to me because yet again I'm guiding a fantastic, soon to be PhDs from my team, to have a successful startup, I really love that. In other dimensional things that are to me is of course the actual intrinsic value of the work being done, not creating a startup, which also has its value, but that would be around open data, we really want to make open data useful for the world. Open data is like 20 years behind open source, still it's amazing and we all know how important data is. Our J value project works on making open data easy to access, deal with all the protocols and what have you, safe and reliable and so even legally reliable.
Dirk Riehle:
And then we also have purely empirical topics like the user led consortia/ industry working groups topics, and this is going to revolutionize the world, it is. You're leading the charge and I wonder about your marketing budget, because what I wanted to say to the previous question was, "No, these folks are not just taking from you and doing it themselves, they simply don't know about about you." I would argue. So they just have some person, sometimes a big shot person who says like, "I want to organize my own open source foundation and we figure it out." So I think that you could up your marketing game and reach those and eventually they will come to you anyway when they realize that the mundane tasks, back office, all of that is much better taken care of at a foundation which does it efficiently and at scale.
Thabang Mashologu:
There you go. You heard it here from a world-class expert, we need a bigger marketing budget at the Eclipse Foundation. Yeah. Well, in all seriousness, I think there is an opportunity for us to better communicate the value proposition of industry working groups around vendor neutral governance, ensuring a level playing field, intellectual property management, antitrust compliance, open specifications, not to mention the growth and sustainment of community development. I totally think there's a great and important story for us to tell there.
Dirk Riehle:
Yeah. Well, it's a great place to organize community artifacts that doesn't just have to be source code, it can be a fair access to the trademarks, to specifications, to test suites, all of that, it's an important role.
Thabang Mashologu:
We've really enjoyed having you on the show Dirk and we're almost at the end of our time together. Is there anything else that we haven't discussed, that you'd like to tell our listeners about?
Dirk Riehle:
I want to point out that open source is not just changing industry dynamics, it's also really relevant for engineers inside companies where it's called inner source, where you work in large companies in such a style, as you know it from open source. And one of our research projects looks at how this really is a better way for developing software at scale and how it's an alternative to the overly complex product line engineering that you can see in some companies. So I would very much be interested in meeting large companies who use inner source to organize the internal development shared reusable components, A, within a product line engineering and B, that's a different topic, but related, across company boundaries in a holding company, because at that moment, you have the issue of profit shifting as you ship intellectual property over text boundaries.
Dirk Riehle:
So we are working actually also on finding good models to deal with taxation issues in large holding companies, where you collaborate and you want to collaborate like that, except that the financial compliance department will stop your dead in your tracks because you make yourself and make the company vulnerable to lawsuits or complaints by the tax authorities. So product line engineering versus inner source and the issue of taxation and financial compliance in large scale inner a source in large multinational companies.
Thabang Mashologu:
Very good. That's a whole podcast [Ralph 00:45:55].
Ralph:
Yes, next time.
Dirk Riehle:
Oh, no.
Thabang Mashologu:
Oh, yes. Yeah, we'll be sure to have you back again. That's about it for this episode of the podcast. Actually, before we go, Dirk, can you let our listeners know where they can connect with you online?
Dirk Riehle:
Sure. I'm on Twitter, @dirkriehle, so that's my full name, first name, last name in one word, no space or punctuation in between and I generally use that also, it's my website, dirkriehle.com. You can subscribe to a stream of sometimes better, sometimes worse commentary on the open source world and the industry strategy. Very often from a business perspective, sometimes from an engineering perspective.
Thabang Mashologu:
Great. Well, thanks again for coming on this show, it was a lot of fun and very informative.
Dirk Riehle:
Thanks for having me. That was a lot of fun.
Thabang Mashologu:
Well, that's a wrap on this episode of the Entrepreneurial Open Source podcast. You can find us online at entrepreneurialopensource.com and @OSS4Biz on Twitter. We'll catch you next time on Spotify, Apple Podcasts or any other place you listen to great podcasts. Thanks for listening. Entrepreneurial Open Source podcast is sponsored by the Eclipse Foundation. The Eclipse Foundation provides a global community of software developers and organizations of all sizes across industries with a business friendly environment to collaborate on open source software innovation. Visit eclipse.org to learn more.