 Good morning, everyone, and thanks for the opportunity to talk to you about Fujitsu's journey in open source now. Even though you're all here, you may be wondering why you should be listening to such a story, except that it's a nice seat to do your email. I want to give you at least two reasons. The one is, I'm introducing a project which you will be able to see at the breakout session and also at our booth, which I think is sort of cool and some of you might be interested or tempted to even contribute or to engage with this project. The second thing is, of course, it's about Fujitsu's journey in open source, meaning some of you might be about or might think about doing something similar in the future. Maybe there's the one or the other lesson in what I'm going to talk about. There's a third reason which is a little bit more selfish, of course, is if someone can give us a help of how to do things better in this game, we're really very interested to learn about that. It's also why we're talking, we're thinking of ourselves as apprentice contributors. We're still in the process of learning. Open source. Arthur C. Clarke, the author of 2001, A Space Odyssey, once said, every sufficiently advanced technology is indistinguishable from magic. Now, as for open source, I think no one will doubt that open source is such a sufficiently advanced technology. Think of all the things that we heard from Jim yesterday. And looking at all the technical challenges it can solve. But from my point of view, even more importantly, if you look at all the social problems that open source can solve, it becomes even more obvious that open source is really magic. Here in Berlin in July this year, I had the privilege to listen to a presentation by Emerson Tan, who talked about his experience with building the national Ebola response system in Sierra Leone. And he did it with three basically young programmers using only open source technology. And besides the fact that they actually managed to get the system going, his apparent or his take on what he actually achieved was much more far-reaching from my point of view. He said open source actually in this country restored the faith and the ability of a government to be able to run its systems honestly. Now, if you're looking for magic somewhere in the world, here you have it. A bit later this year in August there was the DARPA Cyber Grant Challenge. I haven't been there, unfortunately. Must have been a really cool event. What happened there? Computers trying to fix computers, right? So software running against software and trying to correct software. And actually an open source project won the game, won the prize. So you could think of it as open source actually being able to fight a skynet in the future. So quite impressive also social feat, I would say. So if all of that is magic, of course it's also, you have to be careful about how to deal with it. Some of you may know the Ballet of the Sorcerer's Apprentice by Johann Wolfgang von Goethe, where the Young Sorcerer's Apprentice tries to do magic without the master being around and things just go awfully wrong. And so you really have to know what it takes to tame the open source beast. And we're still on our way as Fujitsu. And of course, as I already said, we're also looking for help on this. Just very briefly, of course, as everyone for many years we've been using, obviously, open source, we're running many, many Linux servers. Now recently we're also moving our infrastructure as a service offering, it's called K5, to OpenStack. We've joined many projects, Linux Foundation from the very beginning, up to the OpenStack Foundation most recently. And we're contributing actively. We're number nine in the Linux core development. And for OpenStack Newton, we're also number nine in the recent edition. Next step for us was, of course, to lead projects. And so far this was only for sub-projects in the OpenStack Monastka project, the cloud native monitoring of OpenStack. We took over the log management function and leading the log management function. We're working together with HPE on the whole project as an overall. And about one year ago for the first time, and this was really new for Fujitsu, we created an own open source project, which is the open service catalog manager. So that's basically Fujitsu's first time full software project contribution to the open source space. And you can learn more about this in the breakout session at 11 in Schoenberg or come to our booth if you're interested. So I want to talk a little bit about what experiences we had along the way of starting this project and where we are today. So it's been a pretty winding road, actually. We started out having to cross something that I'd call consensus rich. Fujitsu is a big company, right? Many people having their say about things like, well, should we really do this? Should we do open source as Fujitsu? Next thing, there was something, and I'll say more about the details of those steps. The confidence river, which is basically about can we actually do that? Do we have what it takes to do an open source project? Or should I say stop? Because every one of these steps could have or was at some point in time almost about to stop the whole project. So the next step was mechanics maze, which meant how the hell do we do that? What do we have to do in order to make it work? And then we had to pass a very long stretch of what I call the governance desert because this was, okay, what are the rules? In this room, I think about the technical part, most people are totally clear about what the rules in the open source are, but there were also legal and license issues that we had to solve and contribution contracts and all that. And so before arriving at something that feels like a community house where we would really like to invite the world to join us, this governance desert was a tough place to be in. So now just a few experiences about those individual stops or steps on our journey. About the consensus ridge, of course, the question of whether we should do that or not was easily answered because the industry and the customers are more and more demanding open source these days. We all know that, right? In a way, open source is more future proof than many other technologies. Is there a feasible business model? Well, that's not necessarily true for every type of software, but our open service catalog manager had always been of the kind that, let's say a lot of our business was really about integration and service around the tool and not so much selling the license. So in that sense, quite suitable for open source business. What will existing customers think about that? That, of course, in a big company like Fujitsu is a tough question to answer, but what helped us is that, especially in the European area, we have lots of customers in the public sector and I think many of you know that especially in the public sector, customers are more and more actively asking for open source as preferred technology. So then we came to the Confidence River and I called this river because to a certain point, this step or stop requires a certain leap of faith. So there's things where you can actually do sort of checking. Is our software good enough? So that had to do with questions of patent rights. We had to check whether all the modules that we have in there definitely don't infringe any patent rights. Of course, you should always do that with the software, but if it's not open source, it's less of a hassle, because it's not so easily to see. We had Hibernate or we have Hibernate in there and there was a long discussion about whether the Red Hat agreement that Red Hat reached with Firestar a few years ago would keep us safe. Finally, we said, OK, yeah, it should be OK. We needed to check whether security things are OK in our software. We had done evaluation by Fraunhofer Institute here in Germany, ISEC. So that was good then also. Is our product architecture modern enough? Good. So our OSM was built for the cloud from day one. We have SOAP APIs. We don't yet have REST APIs fully, so that's work in process. It makes it a bit more difficult, but still at the end of the day, we thought, OK, we can go ahead. Will we be able to find high quality contributors? OK, we're still on the way. We're here at this conference in order also to entice people to join, but the software is such that it can easily accept modular contributions. So we might finally made this leap of faith and moved on halfway along the road. We come to the mechanics maze. So the first question, which online resources we need, that's easy to answer, right? We've been using open source for quite a while and knowing what you need to provide to the online community or to the open source community. That's sort of a no-brainer. Next thing is already more interesting. How do we make our documentation easy to use? Because we don't make money with our documentation. So here what helped us is that for quite a few years we had built something that we call a rapid onboarding system for a rapid onboarding center, which is sort of a virtual experience center combined with a sandbox for our product and our software. So everyone can go to our webpage and actually play with the software as it feels live, right? In the meantime, also things like a Docker container for download is available. And the final thing is at some point in time we want to have subscription customers. So how do we differentiate the resources that we offer to those from the resources that we offer to everyone? Basically, fortunately I have to say we made the decision open source should be open source because there is no difference in what we offer as resources. The only thing is obviously we support companies who use our software in the integration process in the learning process of how to integrate and that is something where we can add value but we don't differentiate on the documentation side. Governance Desert was a desert because actually the first time we met with something we had almost no experience inside Fujitsu. Choosing the license was still relatively easy. We chose Apache 2.0 because it's used in OpenStack Foundation. It's relatively flexible for contributors and also commercial partners alike. So that was okay. Are all software components compatible with this chosen license? Took quite a few weeks to go through that and be sure that that was the case. How do we describe the process for contributors? Okay, we finally chose the CCLAs of OpenStack Foundation but seemed straightforward the choice but in a big company many legal guys who want to have their say on that. And the final thing is if we make all these decisions can we become part of an existing open source project? We still aren't. We have thought at the beginning of joining OpenStack Foundation but that would have limited the scope of our project a little bit too much because it's not just about OpenStack. We're working very actively with the CNCF on some interesting things like service broker API so at some point in time maybe but our software needs to be a bit more modular to fit the strict rules of CNCF so still in the process of finding a suitable open source project to combine with so we're still with one foot a little bit in the government's desert and still in the process of building our community house. Of course we started about a year ago telling the world we launched a project at OpenStack Summit in Tokyo last year, last October and we're present at many conferences. Now in the first phase we've been talking a little bit more with potential users and integrators and only at this conference really we're starting to talk also to potential contributors so if you feel interested to do something the OSM is a project that allows you to build an enterprise store for managing virtual resources in your company or a software as a service store it has been used in many such ways you can learn more about that at 11 in the room, Schoenberg or come to our booth. For the community management I'm sure it's not state of the art yet our community is nice but small so any one of you who knows how to do this we are still in the apprentice phase and so I would like to invite you let's do some magic together and I'm sure this is a nice project where we could do it. Thank you for your attention.