 Yeah, Mikheil Ilbeck from the Techhun University of Denmark. I'm also, as already said, I'm the regional coordinator of Region North and task leader of the task of the OpenAir Guidelines for Data Archive Managers. So my presentation is called OpenAir Guidelines for Data Archive Managers, but it probably should have been called for Data Archive Managers, Repository Managers, Managers of Crysis. But first, a little story about OpenAir and about the OpenAir Guidelines. OpenAir is about linking. It's about linking publications to projects. But not just any publication or projects. At least that was the beginning of OpenAir because OpenAir had a very specific task. OpenAir was set out to support and monitor the implementation of the open access pilot in FP7, also called the Special Cloth 39. So our strategy was to build on the existing Repository Infrastructure Bureau. Basically, many of the participants of the DRIVER project was in the project. And there was a reason to reuse the work already done in DRIVER. To harvest the Repository and then link those articles to EC-funded projects. So that resulted in the OpenAir Guidelines, which were based on the DRIVER Guidelines. So that were guidelines with very specific requirements to supporting the goal of OpenAir and the EC Open Access Pilot. So getting peer-reviewed articles from the Repositories and link those articles to projects by getting a grand agreement number from the Repositories. And then display the Open Access mode of the articles and possible the embargo period. If such existed. So that was a very simple approach we had and that was very intentional. We needed to get resolved very quickly. But the uptake was rather slow. So new situation. OpenAir OpenAir Plus, which is the continuation of OpenAir and we just call the OpenAir. And as Natalia already presented to you. OpenAir got publication Repositories and data Repositories trying to link those to provide services. This means OpenAir needs to be more inclusive. So OpenAir is a learning experience. It's a learning whatever it's called. But we also have to unlearn some things. So first of all it's not just linking publication to projects that are funded by EC. For instance the country like Denmark. I mean EC funded research is less than 10% of the total public spending or total spending on research. So if you just look and say to Repository managers and universities in Denmark you have to do this because it supports the Open Access policy of the EC. It might not be such a big leverage for making people change specifications for the Repositories. Also I mean looking at Repositories and other databases at universities they might have a lot of different contents than just you know research articles in them. And especially data is moving. And we wanted to link data to the publications. Another thing we discovered by trying to get the data and get the results to try to monitor how much is open access of the output coming from EC funded projects. We found out of course that the content is is key. So even though we might not get the formal information from the repository through text mining we can go a lot a long way to identify which publication was actually coming from an EC funded project. So for OpenAir it is still very much in scope that we built on existing standards. We're not trying to invent something new. So that's why we're talking guidelines. It's not standard. So anything like that. So the new approach for the OpenAir Guidelines and I got the S underscored because it's several guidelines. It's three guidelines. It's guideline for literature Repositories coming in a version three at some point which means that we also will include links to citations to data for instance. It's guidelines for data archive managers which I will talk more about in a minute. And it's guidelines for crisis. It's not because I'm Tweedy. I'm just checking the time. And so literature Repositories another important step forward is that we are talking about different levels of compatibility. So before there was just one level of compatibility that was you had to be compatible with the OpenAir Guidelines. That means you had to expose a grant agreement number you had to expose open access mode you had to expose in barcode periods. We're still going to include driver as Natalia said. So that means we get a lot of records from from the driver said. So driver will be the basic entrance level to become part of the OpenAir Information Space. But then there are a lot of you could say non incremental levels of being compatible with OpenAir. Not going to mention all of them but repositories might include different types of these information that is relevant to expose in an infrastructure like OpenAir. For data archives we decided to go with the data site meter data scheme because it's a proven meter data standard for heterogeneous data sources. And cross disciplinary archives and that was coherent with what we're doing in OpenAir. It's proven to be useful in for instance the ANTS data portal to Australian National Data Services data portal. It has a trusted and sustainable organization behind it. And we approached them because it just had one problem for us was that it only accepted data that has had DUIs which was natural because that's what data site promotes. So they were accepting to include other types of persistent identifiers. We still think it's a very good idea if there are DUIs but there are other types of persistent identifiers out there. So Chris you're a Chris this is still very experimental and I hope we can have a good discussion tomorrow about that. But for many universities Chris is up becoming not only just the database you use for sort of monitoring the research it's also becoming maybe a substitute for repository or very integrated with the repository at the university. It might be the place you go to to harvest data to integrate in a service like OpenAir. So for that we want to go on and work with Sheriff XML. Again you're a Chris is an established and sustainable organization that creates Sheriff XML and therefore a good choice for including data from Chris systems. So a little more about data archive managers because this is the one of the guidelines that we are working on which is the most mature one. So built on data side which have currently the version 2 2 of the meter data data schema. So some minor adjustments. So first of all of course that we accept other types of identifiers that means that we include a vocabulary for different kinds of identifiers that you can include in the meter data. Also we have some other requirements because if you look at the data side meter data schema they have very limited requirements. Well mandatory fields that you need to have and for a discovery service like OpenAir it is relevant to have exposed more information that you need in the data side and the data side meter states meter data schema. So it provides many different properties but they all a lot of them are not mandatory. So we recommend that you can link to publications to related publications and data sets. So that's one of the key issues for OpenAir and we also recommend that you can relate data set to funding information and that you can well sorry that you can make a we have we need to to have information about the rights of the data that is behind the data. Sorry this is a bit confusing. I'm not going to that more now. Sorry. So first of all we invite you to influence the development of our guidelines. So tomorrow at the technical workshop I hope you'll join us many of you to discuss obstacles possibilities for the guideline approach that we presented here. So that's it.