 Right. Thank you for the introduction. Artificial intelligence, cloud and the conversational web. How does this work together? I will show you some examples. So sometimes you find a chat window in the bottom right of a web page and you can chat with a customer support representative and maybe get information about the product and at a certain point this person says can you give me your contact details and it would be very strange then if you have to change to a web form like this. So you can put in your information right in this chat box and so it's much easier to get a contact information from a customer. But mostly the person you are talking with here is not actually a person, it's actually many times already a bot. And so therefore there's a kind of technology which is approaching here. Another example is if you're talking to a web page, to a search, to Google and you have a full text search phrase, then you get a direct answer. So it's not that Google is a search engine for web pages anymore, it's giving answers to any kind of questions directly. Here's another example. And when I gave a talk two years ago here about search engines of the future I said maybe search engines of the future will simply answer any kind of questions. They will answer anything. So this was the timeline I've seen there are different kind of search engines and then social media came up and the question was will it all go and the answer maybe is that search engines in the future will simply answer to our questions and not to web search anymore. And in the social media area where we created LockLock, LockLock is a search harvesting engine for tweets. We found out that we can ask questions like who tweeted most about FossAsia yesterday. And this is also a kind of question that you wouldn't put into a web page with search facets and so on. So it's better to just ask an engine. So the next step is that social media is also a kind of a contact point where you can roll out a conversational web. So this is the idea that everything goes into this kind of conversational interfaces and these are the products that come out. The big companies are providing all these personal assisted gate-based in form of a speaker and they actually call it a smart speaker. But it's also a device which is able to answer any kind of questions but also gives you entertainment, plays music and so on. And the big question is why isn't it possible to make this as a free and open source software application? So is it hard or is it easy and what should we do? So what would be necessary to do this in a kind of open source because if this is the future of the web then we should participate with this. So let's see. If you want to start a project like this and you want to create an appliance with a speaker and maybe other applications, you have to take care that you not only make the speaker, you also need maybe an Android app, an iOS app, maybe a desktop application and a web application. And also you need all these kind of many skills so you can answer questions about the better and sports news and do entertainment and so on. This all together I call the digital assistant ecosphere and it looks like it's a lot of work. So the first few and open source digital person assistant would be a lot of work. So the question is how can we do this? If you look at what companies have to do to make this, so you have to create an appliance in a closed source form, you have to create software, you have to provide some data services because that's the content which is delivered at these kind of devices and you must have an application platform where you can do a hosting of skills. It's a cloud hosting of customer skills and there's currently no standard way. Amazon Alexa is showing us in a very good form how this is done but it's still a proprietary way. So what would we do as an open source community? So these are the alternatives at the open source community and at this point I would like to take the chance to name this with a phrase which I heard from Daimler last year which was very impressive. They put on a slide saying open source wins over proprietary every time and this is kind of a proof that we may be able to do this as well. So how can we win over proprietary? So one basic principle of open source is standing on the shoulders of giants because all these products which have been done by open source communities are giant products and we can stand on these shoulders and do better and this applies also to hardware. There's open source hardware and free open source software which we create instead of closed software also has the opportunity to use a collaboration of people. Many people can collaborate here. Cloud data services can be replaced by any kind of data but we think of not only data which we have to buy but also data which we can take from somewhere else. I will give you examples and customer skills that's a crucial point without very good skills. We cannot do this project so creating skills must be very easy. It must be in a collaborative way. It must be in a way that's proven to work and here's an example that the Wiki like collaboration would be nice. So creating skills in a Wiki like environment is a good example of what we can do. So here are the examples. Raspberry Pi is a good basis for hardware. Debian is only one example of an operation system we can use and also enclosures can be made in an open source way. There are platforms where we can share enclosures, descriptions. The open source community of FOSS Asia had a good collaboration with Google Summer of Code in the last years and we created some large big good working projects. This is a really good example that this is a good starting point to create a project like this. Data, there's so much open data in the Wikipedia. There's the linked open data cloud which contains a lot of information. The creation of more data collection services or search engine like technology has been shown with the Yasi peer-to-peer search engine and log log harvesting. And there's of course the web in the internet which is now driven by dynamic web pages. Dynamic web pages take their content from JSON services and we can just take out these JSON services and run our applications with the web. These things. So the web is now our data cloud, the data cloud which provides almost everything we want because it's already there in the web. The Wiki like skill collaboration is something I will show you in a minute. So here are the components for the ecosphere of our application. And I brought you an example of Susie application here. I was afraid it stops working right now at this place. So I will try to show it afterwards. We have of course made an Android application, an iOS application, we have made a desktop application and there's a web application as well. You can do this going to our web page on Susie AI and then click on this small burger icon then on chat and then you can instantly try out the Susie chat. And there's also in this ecosphere these skills and the skill language is actually really easy and you access it from this web page as well. Susie AI, click on a burger then you need to lock in because editing a skill needs an account for this. Then you click on skills and you see a lot of skills which are already there which you can then edit and enhance. So you click on just anything and then there's the edit button, click on edit and here is an example skill. Skills are very easy, this is only meter data here and then the only access to the content is made with just two lines to have an external JSON API attached. I will explain this in another talk tomorrow so we can learn how this is connected to actual AI algorithms. And this is the landscape of our ecosphere where we can create such kind of application. So thank you very much. I hope everybody joins in and takes part in this project. Thank you. Michael, I want you to be here. So a few days back I was actually playing with Susie and I asked her a question. Why did the chicken cross the road? You want to answer because you trained it. Let's try it again. What we see here is the status here because of the noise it recognizes always the hot word and therefore it doesn't work like in a silent room. Can you please turn down the music? Susie? Susie, introduce yourself. I am Susie, a personal assistant made by the FOSSATER community. All my parts are made of open and free software. I don't want to risk any other demonstration. To attend FOSSATIA, that's what it said. So, thank you Damini. Thank you Michael for the wonderful talk. How many of you know Daimler?