 it's the first one, yes, perfect. So I will talk about open data challenge. Maybe some of you knows about open data, maybe not. Open data is to make data available for everybody, okay? And what we are providing is a platform running on OpenShift that can allow any provider to deploy an open data platform. So what is an open data? For example, for the French government, you've got open data, it's a law in France. You need to provide data about financial, your financial account subject, about your subventions, about a number of people living in your city, data about transportation, any kind of data can be open to the public. You've got another one there, which is a French 5G deployment in France. So that's the law. So most of the entity needs to provide data as open data. Open data is also sharing data between people, okay? Meaning you can go, you can take the data, you can offer new service for those data. For example, you can develop a smartphone application, okay, which is called re-usage of the data. So this you can do, you can also access to the data using API. So we need to provide also an API infrastructure so that development companies can use the API to connect not to the database, but to a data request, HTTP request, and get the data. And we have a lot of website example. I put some here so that when you can download letters, PowerPoint, you can have some example. So I will tell you about the first open data application. Maybe you know this, it's the cholera discovery. It's 1854, and this is how John Snow discovered where people get the cholera, simply because he put all the cholera cases on a map and he just discovered there was a pump water around this. So which means open data interest is also crossing data from various database, okay? And putting it on a map, it's very easy and trying to analyze your data and find out something. You have a lot of application now, people put where you have some accident in your city to provide to people some maps about where it's dangerous to run bicycle, where it's dangerous to walk, and so on. So our open data architecture, so for this, I needed Dilbert to help me, and I need to understand what is an open data architecture. We, my platform, we use SICAN, some of you know SICAN, you can raise VN if you know it, nobody, so it's a good day for you. So SICAN is the number one open data infrastructure. The front end we are using Drupal, I guess most of you know Drupal, right? Thank you, I'm safe. And we use our BI platform, Vanilla, ETL, to ingest, to connect to a new database and ingest the data into the SICAN database, and we use Vanilla BI to develop dashboard. We provide also a K-Clock, I guess you know K-Clock as an EAM infrastructure and other subject. So that's our data for citizen open infra. Again, some website example, Paris Grandest, region of, sorry, Grand Paris, you will hear a lot about Grand Paris because in two years we have the Olympic Games. Region Grandest, also some, just very close from Germany. So all those websites are always about data, so data visualization, data on map, data publication. And because we had some interesting success in France, now we needed to improve our deployment strategy, our maintenance strategy, all those things that goes when you start having success with your application that you start on your own computer, and after you have to deploy it on many different instance. So how did we do this? First we move simply from a single base server installation to virtual machine. Okay, and it was four years ago, before for us all those things, we would never use it. And we like it, really Proxmox, some of you maybe know Proxmox, it's very nice. We started with Proxmox 5, now we are in Proxmox 6, it's on top of Debian or something like this. My team is happy with this, it's my team is happy, it's first good things. And my customer also are happy, so I guess my life is better. And after we move two years ago to Image, so we start, and there was some development change, but don't ask me which one, okay? It's our deal together, so my team changed something in the software, so now we are able to deploy as Image Docker, and we had a large project with one French region, and they found us the OpenShift Migration, okay? Otherwise we will not do it to be clear with you, we stay on Docker, we don't know how it works really. But moving to OpenShift, deployment and plaster of that are really easy, okay? All the burden about updating the platform, saving the data, all those things, we have some dozen of instance, all is done very, very easily. So again, moving to Proxmox, even we were happy with Proxmox to OpenShift was a nice pass for us, okay? We took this way and at the end it's okay. What we have now, we have image of Data for Citizen, our open data platform, and we have also an image of Vanilla, which is our business intelligence and ETL platform, okay? It's all deployed on OpenShift, it's running pretty well. You've got here an example of what we do, you connect with Vanilla on any kind of application, you put it, you build visualization, maps, all those things, and after you can communicate with any open data platform, and also provide your data as API so that people can connect on your database, okay? We have interfaces and we have what we call pod. Pod is like a small number of image, some of you may know this, and to start a new instance of Data for Citizen, now it's just running a script and you have it ready in like maybe less than five minutes. Before when there was a virtual machine new instance, you need to copy the virtual machine master, you need to change the IP configuration of the machine, it's little longer, let's say less than one day. And now we can have a new application up in less than 10 minutes, which is really, really easy. And we also add new image because you find it very easily on Docker Hub or other Docker repository. So for example, we added Lime Survey, maybe some of you know Lime Survey, it's a platform to build survey and collect data. So it's very clever in an open data application because using Lime Survey you can build survey and ask the citizen to answer to some information. We can also collect data now using open street map. And so people for example, if they want to declare that somewhere there is some trash in a street or something like this, they can use the survey point to a specific area and said here there is something, please do something or act to clean the street or something like this, okay. And all this is running on container, public or private hybrid cloud, we are now deploying on OVH, it's a French, you saw them somewhere here, they are not here, boom. So okay, so we deploy on OVH, we are happy with OVH except when Strasbourg was burning, maybe you've heard about this, they did not tell you but we lost some server also at that time, but basically we are happy with OVH, it's easy to deploy, it's easy to manage. I have no stock option in OVH so I can talk freely about this, that's it, okay. So we have our inside infrastructure to develop and for any of our customer we deploy on OVH, also because it's a French provider and most of our customers are French government so they ask us to host in France, which is normal, okay. France or Europe, there is always some discussion about this but basically we don't move like this, okay. So some open shift interfaces. So this one I saw it yesterday for OVH was in the low key, okay, it was exactly the same. Here it's me, I just give words, I know nothing, you know, I want container, I want Kubernetes, all those things, I want to have my development team very happy, the infra team very happy and customer very happy, okay. And with this kind of interface, let's say we go, okay and we provide good results. What we can do, we can now deploy large data sets of data, we deploy data using geo data infrastructure, we do real time integration and before when you use some single server, we have some memory problems, we had sometimes some crash, I should not tell it publicly but it happens before and now moving honest, moving now to this kind of infrastructure, it's less. It can happen, I don't know, my team did not tell it to me, my customer don't call me so I feel more happy so we can directly ingest data in real time which is very nice and put the data for example on the map. Here it's something very nice, if you go to this website which is La Rochelle, I encourage all of you to go to La Rochelle in holiday, it's a nice city and if you go to La Rochelle like here in Berlin, you can have a bicycle and this is a real time bicycle availability, okay. It's updated every five minutes. So if you want, if you have the smartphone application because somebody developed a smartphone application with those data, okay, you can see where is the next bicycle available or you can see where is a free spaces to drop your bicycle, all those things, okay. So it's very easy to use. What we do also is we provide API so to manage the success for example of La Rochelle with the API, we deploy also gravity. Don't know if some of you knows gravity, it's an API management tools, okay. If you register with this mega city, they will give you an API key and so you can request many API requests every day. If you don't register, maybe you have a limitation to 10 or 20 requests per day so that we limit with this people who just request, request, request to have the server dropping down, okay. So with gravity, we are really happy also with this and gravity also deploys in OpenShift, I mean it's an image, so all things is very consistent, easy to insert new services. I will not tell you about microservices or something but it's kind of new services and we just have some configuration file adding new plan in gravity and after it's okay. We were able to set new application and new services, for example, it's a real time activity for firemen in the south of France, okay. It can be useful, it is useful, okay. And we can also have some dashboard and the other one, this one is an example of you put two or three different data set on a map. For example, you can put the camping data set with restaurant data set, with the lake data set and to see where is the best camping you can go, okay. If you want to have a glass of wine or something like this and after swimming the lake, okay. That's an image but this is just to image my speech and maybe you have something clever to do, okay. You just take all your data and you put it on a map. Unfortunately last slide, customer is not really happy with OKD because of the price. Nobody from Red Hat here, so sorry. It's running perfectly, believe me, but it's very expensive. So they ask us now to say okay, it's a simple application just to display data. It's not nuclear application. It's not health application. It's something honest that can stop during 10 minutes if you understand my word, okay. So we don't need something 100% available. If it's 99.90, it's okay, okay. So can you evaluate what's the cost human side to go from open shift to Kubernetes? And so now we are in a process to install a Kubernetes infra just to see if it works or not. The script are almost all the same. There is no problem. That's what my team told me. It's just knowledge. We have to, because we like OKD, the interface is okay. So we have to learn a new kind of interface. And again, it's just for cost strategy from my customer side. And also it's better also to deploy in other country and other region because if you can only with OKD, sometimes some people tell you okay for this kind of application, it's maybe too expensive. Second things, now because of this new application, we have a vanilla BI platform available on OKD and Kubernetes. So before also with vanilla, when we started a new project, it's like one or two days to set up a new infrastructure. Now it's 10 minutes. Okay. I go beyond my time. I'm sorry. Do you have any question? No. Thank you very much. I will picture you, you know, because I know some people are there. There was people. Thank you very much.