 Hallo zusammen. Danke für die Einleitung. Thank you for the introduction. It's possible that I sound a bit bad, but I have a bit of a cold. The portal went well, it's not because of that. I'll tell you a little bit about the transparency law and which content should be in it and how we implemented it during the last year. At the top you see a long list of names and we will show you later who worked on this. Im Informatikum der Universität Hamburg ist ein Verein der Forschungsprojekte. We are a group at the University of Hamburg, which builds a colloquium of research organizations and commercial funds. And this time we built it with the finance administration. First, I will go over telling you what the transparency law is because I'm not sure everybody is on the same page. Then about delivery systems and basic decisions. For example, the multiple sources of documents which are fed into the delivery system and how the portal will be realized with those. I have some screenshots where we can see our looks and you can right now use it. The metadata model will be explained and the architecture and API will be shown, which allow a periodic download and what we have planned for the next two years. The transparency law had the goal of delivering administrative documents and data to the public. And the target was fulfilling this by October 6, 2014. We decided in a web portal was built with a lot of trying from the authorities and they wanted to get it done until the deadline and it had to work until then. The transparency law that's the first article the purpose of the law is to provide comprehensive right to information and thus making accessible and spread all the information helped by the authorities listed in the second architectural paragraph directly to the public. Personalized data excluded. And the authorities which I meant are basically all authorities. A central point in this law is information, articles of information, information items so to speak. And I'm going to just read them to you. It's the preamble and the subject of choosing those items was to discover how to present them and it was interesting to see what the messages of the city the senate of the city parliament were and this law, this list grew through the participation of citizens through a wiki where they asked which types of information would be interesting. So for example 0.9 geodata is something which would be more interested in the public than the subjects above in some cases. And the goal is there's a lot of different types of data and how do we structure them and the law said that these 15 types of information every type of information should then later on in the portal be categorized as shown here. From the law itself there is a list of delivery points which I need to be fulfilled. And on the other hand there were generally geodata rather generic things such as geodata and then we had to decide what is meant by this and immediately insofar interesting when contracts are concerned from the city of Hamburg with interested parties with firms, externals within 30 days one should be guaranteed to have the option of counteracting a contract so that it can later on be published and have the ability to react. And this is most relevant with geodata. The searchability is interesting in terms of metadata and full text and metadata such as title or publication date and the goal here is to make things easily findable. And it's important that all the states of information have to be supported via revisioning so that we can show historical data from a year ago and we can see and we need to show 10 years of data so that the database does support this. So, starting from 2014 on for the next 10 years the machine readability default formats vendor independence are all basis for new apps or application die Daten quasi verwenden sollen das heißt das eine Zielgruppe should be the basis in case their use on the data provided. Die Anwendung schreiben und die anderen sind einfach Menschen und der Rest, das heißt, der Data und der Privacy und der Protection der Personal Data should be given to protect the person. In some way the authorities didn't do that before and so they had to develop the process how to do it. On the right side you can see in 3-4 Monaten die Datenbanken oder Liefersysteme digitaler Grünpark digitaler Grünpark digitaler Grünpark Fluglerndaten all those things that are listed they are all connected to all those things that are listed. Das heißt, es hat sich ich glaube, wir hatten teilweise so 90 Liefersysteme 90 different systems of delivery und now we have 16 Alres ist das dritte Alres, for example, da gibt es quasi ein Alres-System wo man Bezirke veröffentlicht werden und wo man über das Portal publishen kann was man in der Gruppe decides. So Alres ist interessant, Ratruten-Systeme. So, das ist interessant, das sind so die Datenbanken. Das sind separate Databases. Das nächste interessante ist der Hamburger Metadata-Catalog. Der ist da unter der Schülerstatistik. Es ist unter der Statistik der Pupils. Und mit diesem war es möglich für eine Autorität ein Set of Data abzulegen und das ist das, was der primary usw. war. Das heißt, wir haben diese verschiedenen Liefer-Systeme und wir haben diese verschiedene Types der Systeme, die Deliver Data oder einen Begriff kreiert und wir haben einen Termin die den Export vorsehen aber nicht selbst speichern können und die Autorität kann Export-Data, aber nicht support- und save-it selbst. So, wir müssen das Data speichern. Und manche Autorität und manche Systeme, die das Data persistieren können und wir müssen diese Autorität und das das Data speichern können. Und das ist das, was die Systeme, die die Autorität speichern können. Und das war das, was die Systeme, die die Autorität speichern können. Das ist das, was die Systeme, die die Autorität speichern können. Und das war das, was die Systeme, die die Autorität speichern können. Ich habe auch das, was die Autorität speichert. Ich habe auch das, was die Autorität speichert. endlich auch für mehrere steht, der übrigens jetzt MetaFair.de heißt Meta-Datenverbund. And there is now association of Meta-Data-Modelling, where there is a rename, which will be possible for other different Bundesländer to support their indexes and upload them, but that's top in the future. Wir haben ein Meta-Daten-Modell angelehnt an das Open-Data-Meta-Daten-Modell verwendet. Das heißt, wir haben nicht nur die Informationsgegenstände, die natürlich bei dem Open-Data-Portal nicht vorhanden sind. Da sind Open-Data-Categories, which we kept, and they are all structured, for example, traffic or a school. The main idea is to have one dataset and different resources. Wir wollten ein Web-Portal, where you could search through the datasets and as well an RP for the app, and an architecture, in which I will go into further now. The portal itself is seen right there, the search point transparent, the pages transparent point, the search page, which supports the different delivery systems. Ich habe hier sozusagen nach Transparenz-Portal, and I have searched here for the transparency portal, and we had 16 results there. So the priority portal itself has also information about itself and how it was built, for example the process of contract decision making. You can decide whether you like the look and feel or not, but it's a search page and on the right here you have the filters, as they are known, and here the Open-Data-Categories, which can be seen. And there is a detail view, which contains the primary mode of actually downloading the resources. In this case a PDF on the right. You also see a time-based information, an interessierenden Zeiten suchen, zum Beispiel. Advanced searches, which you can look for. And you can look at this, I will go further into the depth of, I said that a metadata model is based on Open-Data, on the right you have the RP field names, if you ask something through the RP, and on the left the German names for it, and the author is the authority that gave the information. The Informationsgegenstand is the groups, the Open-Data categories. The others belong to resources, which you can download. The offline date is the date when it's going to be republished, 10 years, a data set is supposed to stay in, and there are as well data sets, which are put in on free will, which can be deleted earlier as well. And that is the architecture, it looks really dramatical, really big bubbles, I'm going to go to the next picture, we have implemented the upper green thing, search, searching, harvesting, mapping, the delivery systems, don't belong to the metadata scheme, in the middle there's a full text analyze, we have done it with the tool, you might know it from the open data group, the green part take it from us, and especially the RP, the portal can be checked through the CCAN RP. CCAN is a comprehensive knowledge archive network for the Open-Data group with the Open-Python here, in the Solar Index, in governmental parts of British authorities, and CCAN delivers harvest possibilities, where we can download the metadata, to get the geodata on the tables, and reanalyse them with Apache TK tools, and the solar search itself delivers these filters, which we used to advance search. The website itself, Hamburg, the A, is the design, the CCAN as well has a UI, in which we had especially to fit to our metadata. We have built advanced things for CCAN, especially to be harvested at 1900 hours, and then we present at the right time and place, and the distributed running of these harvesters is efficient, because otherwise we had always had to wait for the slow system. Highlighting is not yet possible, so for example if I search something and the match might be coming from something in the abstract, or so, it's not visible yet why and when that is matched, but in the next version it will be supported by CCAN, that display of highlighting. And what we of course have is the harvester implementation, the API allows programmatic access to the portal, and can be downloaded into this web and URL, and reviewed. And the search, primary search is the package search, which allows search directly via the API. So it's also, for example, possible to set feeds and get notified about new items. So, if you want to use the API, I have listed it here as metadata package API, so this was CCAN at the bottom right, you can see the idea of provider data pod, which means what we call the central storage. And the literary export of the authorities is exported to the processing system. So when you are actually getting a record, you are retrieving it from central storage. So for a CCAN, the central storage is like another delivery system. And this is contained in this principle, and at any time we can rebuild the entire index as long as they are in the central storage, even if it's not in a persistent delivery system. Und hier unten ist das Rote. And at the bottom here, the red part is the new workflow for the authorities, the employees. Unter now in the authorities, you had a folder, which went into the archive, and now it's eine Checkbox, das in die Transparenz-Portale bewegt und dann eine weitere Metadata, die zu enthalten ist, und dann haben die Deletions einen Platz, und das ist ein Werkflug für jeder Employee in der Autorität, und dann ist das Werkflug verabschiedet. Für dieses Portal, ein wesentlicher Punkt ist Qualitätssicherung und das Main-Point ist Qualitätsassurance. Für das Beispiel, dass alles jetzt ein guter Titel ist und das ist der Job aller Employee und die Authors, und es braucht zu mir sicher, dass das Harvesting-Prozess korrekt für alle Sätze von Daten und Auserhalt durchführt und von den Autoritäten von internen und externen Daten. Wir haben ein bisschen das Problem, dass die Daten für Struktur, und ich habe gesagt, dass die Kontrakte archiviert sind als PDF, und sie kommen bei uns als PDF auch, und wenn du es in den Open-Doc switchst, dann kannst du es bauen, wenn du es möchtest. Axel ist von der Firma, die wir aus jedem Axel bauen können, ein Zip-File, ein Zip-File. Also wenn du nicht Axel hast, könntest du es weiter erweitern. Wir wollen es mehr öffnen, dass es möglich ist, Feedback von Daten zu bekommen. Also wenn du das und das missstattest, solltest du das fragen. Automatische Metadata, das haben wir noch nicht, das wird in der nächsten Version. Es gibt noch etwas, das wir noch nicht gebaut haben, es gibt noch ein paar Synonyms, Open-Data, Gafdata, Community, We've got, we will advance those, there are other Terms, which are needed for mining data, if I want to have a summary of things, if you have 50 pages, you want to have a summary, so I want to have a certain information about a data set. And this topic advanced delivery systems, we will build it into the next data systems. So thank you for your attention, that was my part, thank you Lothar, wer Fragen hat, es gibt hier Mikrofonen. We have time for a Q&A session right now. Let's go to the microphones. Fangen wir gleich mit dem Nummer 2 an, ja, Mikrofon 2. Ich habe eine technische Frage, ich komme aus Baden-Württemberg, ich bin von Baden-Württemberg, red-green coalition, now works on an information freedom act, und nothing has happened and the interest is low, and in other federal counties, there is much happening. And the question is, are you in Hamburg just building this, working on this, or is this being looked at by other parties, other counties, and is this just proprietary to Hamburg and everywhere else, there is blocked. And the second question is, in the transparency law, you have the exceptions like trade secrets and external data. And in all these federal countries, the authorities try to block anything they can, whoever is in charge, how is it, so the question is, how is it in Hamburg, are the authorities different here, or are they the same blockades that are everywhere. So my first question is proprietary, please only do one question, I have added them together into one. So the first question already is posed in a way that, I mean, there are other countries that have interest in it, federal countries, so for example NRW is starting a new portal right now. So I see interest in our work proprietary, maybe especially because it's only about Hamburg, but from the reductional side, I mean, I don't think it could be used by other federal countries. And the other question was, some authorities do it, some authorities don't, but during our project, there was an extra teaching and multiplicators to try to change the behavior and change the thought, the idea how they are. Some stay how they are, and some have, also dieses Thema veröffentlichen ist ja auch was Neues und verteilweise auch Leute. So the topic publishing is a special one. We only have time for two more questions, and I am going to number three. So vielen Dank. Thank you very much. The parliamentary database really needs to be cleaned up. If somebody ever looked at the UI, it's death. I have some concerns regarding the decrease, which are confidential, and this will change in January, and we'll all run around with toilet brushes again through Hamburg. And the question here is if that would give problems within authorities in terms of publishing information in the reserve guard. And what in the last ten years I've read in the Hamburger Abendblatt is that references to the newspaper in 1980, and the problem is often that the resources there are only in scanned images instead of index data. Well, the ten years are only now for starting now, at some point this will then better, and it can now be harvested, the portal, and it's being done now. If you also harvest the resources, you could then keep those information. I can only say that something is missing, something is bad. Let us know. Okay, und dann kommen wir zur letzten Frage. So, the last question. Okay, everybody now. We are still at the last question. Let's bitte daher leise in den Raum kommen. Please be quiet. Die letzte Frage. The last question. Ja. In Letz wo die Frage gestellt. Auf welcher Entscheidungsebene in der Verwaltung solltet? The level of decision in the body is the one where decisions are making. Auf welcher Entscheidungsebene? Administrative level? Kann ich nichts so sagen? I can't say. We don't know. Das ist nicht für jede Behörde gleich. It's not the same for every authority. That's decided by every authority, if they have to publish data. Vielen, vielen Dank. Thank you very much. The possibility to give further questions after the talk, with the loader directly. Outside would be best, because the next talk is coming in. Thank you very much. And a warm applause for Lothar. Thank you.