 So, from one Michael to the next, this is Michael Chroma, Vice President of Technology and Architecture at Copano, and he's going to talk about how we can convert our open source project to profitable business, not only in the financial way. Give an applause for Michael. Thank you very much. I'd like to start with a big, big thank you for all the event organizers. It's really amazing. It's, I don't know, my fourth or fifth time at FOSDEM and it's every year really amazing. So, big applause for FOSDEM. Thank you. So, let's start off. This whole talk is not going to be about political correctness, not whatsoever. That's the fun thing when you're also in a business world, you don't have to be. But you will see that not only business guys have to be politically correct, also in communities they often don't need to. So, this talk is really, can be offending to some people. I know many of these people, so they won't have a big issue with it. Who am I? My name is Michael. I live now a long time in Germany. I'm an open source guy, enthusiast. My name is Jast. I've committed code to many projects, Asterisk, Linux, kernel, VirtualBox, many others. I really truly believe in the values of open source software or generally the free and open source movement. So, don't get me wrong. You can also check it up in the big internet. So, but many of the statements that you will see today will be a little bit contrary and that's absolutely intentional. Because it's a little bit also what I want to try to make is a wake up call. Because in general, most free and open source fundamentalists believe that they are mutually exclusive to business. And there are many counter examples. I'm sure you know many of them. But my beforehand, Michael, was bringing out quite some nice things in terms of funding and how you can actually deal with it. And yeah, I just try to show you my view of the things. As you realize, I just skipped in for Saida, which unfortunately got ill. There's going to be a reference later to her, so I think she'll like it. First of all, we want to start off with definitions. These are the Oxford dictionary definitions of what a community is and what a business is. So the community is traditionally just a group of people living either in the same place or having a common characteristic pattern. And businesses, well, businesses just have the intention of earning money, right? Regular occupation, profession or trade, as it's in the definition. It's not just a person, it must just not be a physical person, but it could also be a natural person like an organization, corporation, whatsoever. So when you take two quotes from two hopefully not unknown men to you. One being Richard Stallman and one being Steve Jobs. You could, in my opinion, interchange them. Not saying that Richard Stallman would say the same than Steve Jobs. But I think also Richard in a meanwhile realizes that things can really be done by a single man and these are two men which really did a lot of things. One more in the open source world, one more in the communication and electronics world. But both had an impact, both made something real big. And yeah, Richard is one of the guys why we're basically also here today. But this general mass perception of community and business to each other, especially from the view of fundamentalists, is that a community is always the good ones and business is always the bad ones. Just think of all the battles that we have with large corporations such as Microsoft which were yeah, they're just evil, they want to make money and who cares about them and we just want to be free, we want to be open, we want to share data, that's all true. But the thing is, look at a company like Microsoft now. They've open sourced key core technologies of them, not all of them. Maybe they'll get there, I don't know. But for certainty, a corporation is also nothing else but a community. Maybe with slight different visions, maybe with slight different tools or ways to do it. But just saying that a business is money driven and they're the evil guys, that just doesn't work. And most people think it's either a dictatorship in the worst case or maybe a monarchy because you have your traditional king and that's it. But even in the community world, not every community is the same as the other. Some have really moral intentions. They want to do something for the good. They want to protect your data. They want to make sure that you can use things in a free way. You have the transparency obviously. You want to know what the computer is doing essentially. You don't want to get ads for something that you don't ask for. So every community has its own goals essentially. And as in any case, communities as corporations, they just also do it in awesome products, they want to try a solution to a problem. So GPG is a good example. GPG tries to protect your data from prying eyes. So you can encrypt, you can decrypt, you can share, you can, you know, just deal with your data in a way where you think this is the data that I want to see protected. This is nothing of anyone else's business. Some communities unfortunately do lack a vision. They have some ideas. They think of, yeah, we want to go there. We want to go there. But in the end, they just always hunt the next iteration. Yeah, they fix 100 bugs. They move on to the next one. This happens especially to projects. And I want to give a good example. Spacewalk, for example. Raise your hands if you know what Spacewalk is. Not too many. So Spacewalk is actually the core backend upstream component from Red Hat, Red Hat Satellite 5, which is used as a conflict management and update and patch management solution. Spacewalk is really a cool product, no matter what. I personally at least love it. But they don't have any vision anymore. Because they are, I have to correct myself, because Red Hat has decided to say, like, well, Spacewalk is okay, but we just, you know, make four men and pop it and whatsoever. We make our new satellite six. And that's where actually the company that decided, oh, we have a different direction, but Spacewalk is still not dead. But they don't have a vision anymore. So any community needs a vision. Where do I want to be in a certain amount of years? Also, this is also a little bit from a definitions point. This here is the kernel 4.9. Based on the contributions that have been made to the Linux kernel from individuals with their, yeah, employer, basically. So when you look at this, you will see Linaro. Yeah, it's a great organization, especially when you take all the device world into. But also Intel is contributing a lot and many others. Open source organization and closed source organizations. In fact, you, I don't think you see them here even, and I'm also not a Microsoft fanboy. But Microsoft has also contributed a heck of a lot of code into the Linux kernel. Of course, it's helping their own business, but that's the way it is. So don't mispercept a community with a project. That's also what's happening very often. Especially in the Linux world. This is actually a little bit outdated, but this is from Greg Crow Hartman, officially from what they brought up in the Linux foundation, how much single individuals, especially in the large project, or the largest project, open source project, like the Linux kernel has to do. So we can thank all these guys in doing such a great job. Greg Crow Hartman, without him, the kernel wouldn't look as it does now. So, but in the end, the reality about a community and a businesses, they both share the same general idea. If you're part of the delivery office team, you might envision and say like, I want to make sure that everyone has free access to an office suite. And actually, I want to make that product so good that there is no one else, that this is the main project. Because everyone who puts his time, even if it's like, you don't often get paid for what you do, but in that case, extremely want to want to see that this is really moving forward. So you sort of have a vision for what this community moves towards too. And also, that's one of the reasons why a project, you know, there are many projects that are evaluated for its activity, for example. Ah, half a year, something didn't happen. Good example I can give you here is Open Change. Open Change had the best intentions possible. They had as a goal to say, we want to implement the DCRPC protocol from Microsoft in an open source way and get rid of exchange altogether. The goal was nice, but the backing of it was horrible. Indeed, with one guy only leaving literally the project died, unfortunately, because that's not what change is about and open source is about. So in very short words, community and businesses have quite similar ideas of how they want to move forward. And it's nothing bad. I mean, look at the Linux kernel. There is no question as a number of billions of devices that are on the planet, the Linux kernel is dominating the world and it's good. It's really good and we do want that in any case. So looking at those based on a comparison, a community and the businesses, the community or communities in general have ethical purpose. They want to do things in a better way than others do. They want to do it in an open, in a liberal, transparent way, all the things that I just mentioned. But there are also communities that also are profit oriented. So profit oriented, let's take a foundation for example. If a foundation wouldn't work sort of profit oriented, well then they could not afford all the things that they are doing. Maybe also paying once in a while a developer or just making sure he has the travel expenses compensated and whatsoever. So at some point the profit is not really like they get cash out someone, but they use that profit to grow the project. And that's a good thing. Also community. There are many, many, many people who are very politically touched, especially in open source, free and open source movement world. People say and there was a great talk from Richard, Richard from OpenSusia, who really said, well not every community has to be a democracy and that's good. So I will bring up some examples where you can actually see that point even doubled. Also in this comparison, businesses are always very focused on intellectual property, obviously, because they invested into it. There are many companies, you have people that take money, put it into an organization to pay the bills, to pay people, developers, to pay infrastructure, to do all that stuff. And yeah, communities often get this by donors and whatsoever. So the thing is from how they are acting, it's obviously not exactly the same, but they have quite some important similarities. So let's take a stupid example, I admit, but I just love it. At FOSDEM, this event, as I just started my talk off, it is a really community-based project. And it's now, yeah, 20 years, and I hope it will be the next 100 years, because this is really great and the spirit is awesome. And when you want to get a water, you pay one euro, and that's good, because every water you pay actually gets into the donation and it makes sure that the infrastructure here is paid and that we have all the cool stuff that we have here, power and so on. You get my point. Let's take a stupid other example, which is very recent. Red Hat bought for 250 million, a company which essentially had pre-software, open source. So who knows about this purchase? Hands up. See, a lot of people do. That's great because this is exactly also what communities are about. They are worth something because there are people behind it and people that make a difference. This technology is important towards Red Hat, and Red Hat is also an open source company. Everything that they do is really open source. I give you an example when you take Red Hat Enterprise Virtualization, Red Hat Enterprise Virtualization, it's about product for Red Hat at the time. And what they did is they just thrown out the window server and just replaced it by rewriting everything based on over. So it's an example of business and community working really well together. Red Hat has the interest because there are shareholders behind there. They want to make sure that, yeah, they can pay everything. But they do it in a much more ethical way that just fits. So if you translate 250 million into water bottles, just imagine how Fosden would look next year. Okay, so I also want to give you a little bit more practical example of what we've been doing. Copano is something that you earliest could have heard of two and a half years ago because we, as Copano, we were formally a company called Serafa. They had multiple reasons and I really know what I'm talking about because Serafa was a company that had a tagline that said, best open source exchange drop in replacement. And that was, yeah, that was the truth. It was great, but we were an open core company. Open core company says we have about 90% or XX% on open source code and the rest you have to pay for. Our goal was actually delivering a real choice for business customers. So business customers tend to have money, right? So we didn't, to be very honest, didn't really much care about community in general. We had some, we had a lot of partners and they brought in their things, but we reviewed them and everything and so on. You know the game. But in the end, what are we now? We're 100% open source. Not one line of code is closed. The only implementation of open source mafia back in the world does from us and we now strive for higher goals. And these higher goals we want to see working together with a community. These higher goals are not, how can I say, just making more business customers because then we could have stayed as we are. Our goal is to make something better than the product that we originally wanted to replace. That's why we're better than a Microsoft Outlook replacement. The thing is, Surafa had a past of over 10 years. He's 10 years. We had one product that we wanted to challenge and that was Microsoft Outlook. That's email, contacts, tasks, notes. And Kaliner. This product though is just staying as it is. From a feature set it doesn't really change much. But we as human beings we've changed a lot. We do have web meetings now. We do have a lot of requirements in terms of document management and so on. And when we would have stayed with Outlook like other projects also stick to their thing just forever and never till the end of time, well then the project will get irrelevant at some point. And our thing is providing a real choice for business customers and anyone else. So the benefit to the community is if you have a Raspberry Pi, run your Copano, have fun with it. We don't care about it. But when it comes to the business world if you say like, hey, I want to have a supported environment. I need clustering, whatever. And I want the cluster set up by someone who really knows what he's doing. No matter what, then you have a company behind there because there are people, professionals that get salaries paid and that have a great time in this community because nothing else is also a company. It's a community to make things work in a better way. So from a community perspective there's nothing better to achieve than get downstream in a distribution to spread the word to make it available to all of them. This is the example of Copano. It's in OpenSUSE. It's going to be in WN10. It's going to be in Ubuntu. And also one of the things that I want to bring up and that's one strong thing that Michael before me also brought up. Don't look at your community as that's only the best in the world and no one else next to me because that simply doesn't work. There are more liberal persons that say, hey, you know what? We're actually doing a little bit of similarities, like this whole... I mean, variety is good, right? Docker, LXD, LXD, and so on. It's good. But when it comes to common ground, sometimes it would be nice to, you know, reach over the aisle and just ask, hey, guys, how are you doing that? Let's do it together. Let's maybe make it in this project happen. And that's what we did as Copano Core. When we open-sourced, we just said, like, yeah, we're not open-sourcing all the stuff. To be honest, we actually rewrote a lot of our components, but we also took a lot of external dependencies into account. On software that we base on. So when you see a look at this co-varity graphed and you will see that at that point in time, we thought, like, hmm, let's take the external dependencies into account. And our so-called co-varity score, which covers the defects of 100,000 lines of code, or how many defects there are in each 100,000 lines of code, then we had a quite horrible score. Compared to the average mean of how many defects are generally in open-sourced. But then we started to say, like, well, okay, that's the way we go. So we also fixed that software. A good example is G-Soup. Also a good example is MySQL. And that's how also a community can help. You've seen that on a Linux kernel, and you also see it, for example, with us. When you as a company have a goal, or you as a community have a goal, you can also take all these dependencies into account. Or you should, not could. You should. So real-world view, just to have this comparison in this area, a business can be ethical as well. We did everything open-sourced, and we recommend everyone. Even big players already realize that it's the best way to go. We can have morally good intentions, social environment, whatsoever. Yes, we want to be also able to pay salaries and hosting and whatsoever. And generally, if you run a company in a good way, then you're not doing it like a dictatorship or a monarchy. People, especially the fundamentalists, live like the practice of, I don't want to get told what I have to do. A real company doesn't work that way. A real true free and open-source thinking company doesn't do. So my point about going back to free and open-source software, I really recommend that you forget anything about democracy in open-source world. That's simply, well, it works, but it works badly. And I can give you proof of it, and I think you will agree on that. This is one of the political correctness things. Please note this talk replaces love what you do every day, and I bet he loves when he did this. And I have to say it's good that he did that, because it was a wake-up call to NVIDIA, which was actually the recipient of this photo. NVIDIA had the idea of we have to protect our intellectual property. We don't want other competitors to just take our code and make better GPUs and whatsoever. So my thing is this is clearly the unmatched top guy. He's the king of Linux, and no one, I will get back to that also later, no one will challenge that for many reasons. Then we have someone like Greg Craw Hartman, basically the prince of Linux, if you can call it that way. He's the guy who's making sure that we have a stable kernel. He's the maintainer of the stable branches. So, all in all together, to summarize a Linux kernel, yes, it is as ethical as it gets. Free code, free platform to run your applications no matter where you are. It's as transparent as it gets, it costs no money, you can run it anywhere, have fun with it. So the moral of good intentions as well as well, are there as well. For the social part, there's a cool movie called Anger Management. Linux, watch that, you'll have fun. It's non-profit already mentioned, and the most important part about any community and company, in my opinion at least, and I'm doing this now for quite a long time, is really do it merit-based. I know a community like LibreOffice, for example, can say like, you know, anyone is at the same level, because the reality and the truth is not everyone is. When you get started on a project and you want to grow into it, of course you can start at the absolute bottom level and just say, hey, maybe I do a little bit of translation. Every bit helps, it's important, no matter what. But certain projects have had quite some interesting developments by just giving the wrong people the right credit. And that's quite important, because also a community, also a community-based project, should always go by what the merit is of that each person. And from the business side, well, you won't find many free and open source projects that get that much funding, and it's good that way. But there's also a big but. Who has ever attended LinuxCon? Hands up, please. Okay, recommendation, please do. LinuxCon is really recommended, and there is a t-shirt. Well, there has been, I don't know how it's this year, or last year, but many years beforehand there was a fun t-shirt actually, and there's a lot of truth with it. It said, 2013, the year of the Linux desktop. 2014, the year of the Linux desktop. Guess what? 2015, the same shit. So the reality is, Linux is awesome. It runs on every desktop, but it's not used on every desktop. And that is a pity, and there's a great blog post that I want to recommend you from one of the GNOME guys, where why hasn't the year of the Linux desktop happened yet? And he will bring up quite some valid points. And one of them is this fragmentation. GNOME in itself is a great project. KDE is a great project. Like, LXDE, there are great desktops that are available for the Linux world, but they are fragmented. Some work there, some don't work. Then we had some change from Xorg. You know, this is seriously, when you want to have mass adoption, please, we all know that we want to have as a normal user when we want to bring a laptop to someone and say, hey, try Linux. And he says like, yeah, okay, Linux is not equals Linux. Yes, that's true, and choice is good, but give it a good default. And this is where I'm getting very politically incorrect. We have some great standard called LFS. LFS is really, it defines standards for common usage within Linux. It's amazing how many distributions don't give a shit about LFS. And that's where we have fragmentation. We have fragmentation applications. We have fragmentation in desktops. We have fragmentation in usage because everyone thinks he can do it better. And some of us, some of us really take changes. Look at what Ubuntu did with Unity. They realized, hmm, we made maybe a mistake. We should really go back to GNOME and they did. And I think we can learn from that step. We should try to unite more and stop fragmenting everything because this is where the real power of free and open source movement comes from. We can agree to disagree on certain areas and we can make things configurable. We can make them available on different premises, no matter what, but I really, and to be honest, I see this now for 20 years. I am really pissed off about that. I thank you really much for this chance to bring this up because this is not only my opinion. When you talk to many of the leaders in open source areas, I just had a talk today from someone from Postgres. He says like, you know what, I built up my business based on Postgres and the community just doesn't understand what I try to tell them. He's trying it really hard, but it just doesn't work. So every good open source project needs good marketing and needs good structure and it doesn't have to be democracy. Stop requesting that everything has to be a base democracy because you're hurting your own project. So, as good marketer, this is the stack of Copano, that's it. Thank you. Any questions? Oh yeah, it works. Questions, raise your hand. We'll make sure the microphone spawns in front of you. Sorry, didn't see you. First, thank you for this speech. Okay, thank you for this speech. I would like to ask you how big companies affect standards and in turn how they affect the communities which implement these standards. To be honest, I have not understood you well. Can you repeat please? How big companies affect the standards and how these standards affects, let's say, the software we are building. Because for example, when you are talking about open ABI's, you always find the big fish, something like Oracle or Microsoft couldn't provide that stuff. You're more on the standardization part as far as I understand you. Okay, so from the standardization point, I must say that even the big players have learned about it. If you take protocols like DCRPC that was invented like 25 years ago, they made it closed obviously at that time and they've realized that they benefit actually when people can interoperate with their stuff. So actually the big corporations generally from at least what I see have adapted that they realize actually we can make more business if we're open and let just the others integrate. We just have to go merit-based. We have to make our products better than that. And at some point when you look at desktop stack as I used that as a basis, they're like GNOME VFS and GNOME VFS 2 and incompatible changes back and forth. There's a good reason why Linus said we don't make user space incompatible changes and that's a good rule. That's really good. In my opinion, this is one perfectly executed open-source project but there are many others like Rocket Chat and so on. But let's take Microsoft for example. One of their latest developments is the Microsoft Graph API. Completely transparent, completely not the same as it was before because they just say hey, if you want to use our API, just use it. Have fun with it. It's actually the same mentality that it's run. But the reality is that especially in an open-source world this is not happening often enough. There are enough examples of that. There's like in chat ops, there's 10 solutions now for chat ops. The one is better, the one is worse. It's all started from Slack which is a good product. But then all kinds of new products showed up and they all say hey, I have a bigger, sorry, not politically correct, but the reality in the project. Sorry, we have to stop the Q&A right here. The next speaker is in the line. I'll be around at the Copana booth. Thank you. Thank you for the talk and thank you for the insights.