 The program you didn't even have a proper title of the presentation just mentioned you from data side. This is because I'm a sort of last minute invited to this session. Normally, there should be a more technology related speaker here, which what I am not, but the technology guys weren't available, obviously. So this is a slightly more general presentation about data side, but I come to technology at the end. I start talking a little bit about do is and also about publishers. So this is here you see a table of contents of a journal, which is, by the way, high impact factor open access access journal, nuclear acid research, and you have in all these articles, you have the do is this is well known. This is used by almost all publishers. Now these do is are provided from crossref. And if you go into the detail of one of these articles, and you go, the do is mentioned again, when then you go into the references and the bibliography list, then you have all these references where it's mentioned crossref and if you're clicking on crossref here means you click you go to you link to another article via do I. This is not standard standard presentation in some cases presented differently here it is by clicking on crossref. But the the principle behind is that you this is what is called crosslinking and then you get go to the article the next article and which has another do I and so on. And we think that in this chain of the resources linking data set data sets should have the same should be should be linked to in the same way as articles and this is something that that is starting to happen. Of course you can happen to you that here in this particular case probably this this will this will be the case that you haven't access to the to the to the full text for data. This should be slightly different because, for instance, we had something that we often remind within data side this publishers Brussels declaration and STM publishing where they say, for instance that raw research data should be made freely available to all researchers. And in particular sets of subsets of data that are submitted with the with the paper to a journal should wherever possible be made freely accessible to others. So this is something that we normally should always remind the publishers this declaration staying staying with publishers. You also know that many publishers require as a condition for publication to make the underlying data available for pure you in particular and not only this is from from nature. Not only they want to make it to make them they require to make them freely available but they also indicate precisely what kind of data in which discipline should be put in what kind of data repository. And then we have something different. This is an Elsevier journal and there you have the guide for authors guideline for authors and there is something called data at Pangea Pangea is a very important big data center in earth sciences in Germany. And they have a good agreement with with Elsevier in particular particular about bidirectional linking of day of articles and underlying data sets. So if you have a look here this is this is made this is explained in the in the in the guidelines. As soon as it will be startable you might want to refer them to them in your article and in any case data supplements in the article will be automatically linked. As in the following example we will see the example and then they also also also say that the eyes are guaranteed never to change. In fact what is guaranteed never to change is what what is behind the do I think this is the idea of persistent identifier. So and then this is still from this the same journal. This is how it how it works in practice. You have the the article and you have a button to click on to go to the data set and here's even a nice additional application with Google Maps where you can see what what this where this happens because it we're in the earth sciences here. This is the do I home page the international do I foundation home page in fact and do I's are registered by registration agencies if you want to have a do I today you have to contact a registration. Typically your local what I would say registration agency and data side is such an agency. And what we could what can also be mentioned is that do I's are now since last year I think in an ISO ISO standard and you may also know that do I's are strongly linked to handles and this is the home page of the handle system which is developed by the CN CNRI which is cooperation for national research initiatives yes cooperation national research initiatives in the US and and in fact technically it's the same thing on the do I web home page. They you can read this they say at the infrastructure level do I names are handles because sometimes there is some confusion about it. Do I's are opposed to handles but it's they're not exactly at the same level. Some two or three slides from from a workshop I participated in in 2011 knowledge exchange workshop on persistent identifiers and handles were presented. Do I's were presented others as well but I focus here on handles and do I's and you can hear the list of you of users of clients of handles and IDF is here. Listed international do I foundation as such a user of handles and in particular. Below the IDF you see the different today existing agencies there are not so many crossref and data sites being being the two main today and then in this workshop there was a particular question put to. Yes though this to highlight that IDF is user of handle it was a particular question on quality of persistent identifiers put to all those who were present and hand this is the was the answer of handle. They say use is not actively monitored and they also say that prefix holders must agree to be good citizens something like that. And do I goes much further than that and at this to the same question was do I answer the question do you have a policy on quality of persistent identifiers. The answer was yes this is the result of our organization that's what we are there for essentially we want to build this trustworthy infrastructure needed behind behind this identifier system. Okay oh yes this is slide shouldn't be there it's a French one but nevertheless this is the French Wikipedia page for for for DOI where they you very usefully point out that first of all do I say in fact special cases of handles that do I has this social what they call social infrastructure. And we also should also be mentioned that digital object identifier is to be interpreted in the sense of digital identifier of objects not digital. Not identify our digital objects because do I names maybe of any form digital physical abstract but this becomes a bit philosophical but something that should sometimes the question was put to me do you do we have also physical physical entities can we can we assign the eyes and the answer is yes. Now I come to data side data side global consortium carried out by local institutions focused on improving this the scholarly infrastructure around data sets promotion of data sets in the information services in particular. Initially but not exclusively based on the UI system we are we have members who register handles of course but also things like arcs. It started all in 2009 with with just six members six research libraries. Well yesterday there was some discussion or comments on the future of research libraries. This is an initiative in which research libraries engaged in three years ago. We were six and today we are twenty two seventeen regular members around the world and five affiliated members members that just support data side but we don't assign the eyes themselves. The different roles of the data side registration agency on the one hand and the what we call publishing agents a generic term for data centers research institute data publishers. In particular it is important to point out that the publishing agents data centers remain responsible for quality assurance storage. They also create the identifiers and they have to create and update the metadata we the data side members sign agreements formal agreements with with the publishing agents where all these. Responsibility responsibilities are mentioned. What type of what type of data are we talking about anything we have a very broad definition of of of data here you have some you have some some examples. We have for instance a member who essentially assigns do is to great literature. But of course the the the idea is that the the the large majority of objects should be and will be and are already data sets. But in fact it's anything data is anything that is a foundation of further research. Data is evidence in fact. This is a structure we are a member of the International Doi Foundation. And as such data site pays a membership fee and the member institutions in that turn that don't pay a membership fee to the data site. And as I said the data centers have to contact up one of these member institutions to to to get the eyes. Bridging the gap here it is the case that is also your your your goal. But here it is particular the case for bridging the gaps between publishers and data centers. And there is a formal sort of agreement with Crossref who assigned the do is for for article journals for journal articles. And they as it is mentioned here data site and Crossref have committed to the interoperability of the do is. And there should be this what we call bi-directional linking between data articles and data. This was formalized or officialized last year by a formal statement between STM the publishers association and data site on best practices on. On making data available on how to site data. And this statement was then also joined by Crossref later. As I said we will see an example that we have here a data set which is a supplement to an article. This is a data set within this Pangea data center. Here's the the record of of this data set data description on the Pangea platform. And then you see that this is a supplement to an article the article itself as a do I you click on the do I you go to the article. And on the bottom on the right there have this supplementary data button button and you can go back by clicking to the data set. Which in the gap to sum up we data site provides services for for for the different. Players involved the researchers themselves the data centers and also the publishers and then you see a long list of services. In forms of specific URLs related to to to data site. And I just mentioned some of them also what the data site has has a lot of a lot of some working groups on specific topics. Like for instance the metadata working group who has worked worked for the release of the metadata schema that was mentioned yesterday. Metadata store this is this is the the DIYs have can only be created minted as we say together with metadata. We have a limited number of mandatory fields in our scheme. And as we heard yesterday for the for the need for the purposes of open air plus this would have would have to be changed. Although in principle you you would as far as I understood use this this metadata scheme. This is more formally an interface for data centers. In fact, a well organized data center does not is rather autonomous in creating their their their DOIs they just link to this to this platform. We are just giving them an access code login and then they can create the DOIs and they they upload the metadata. And this can then be these metadata can be searched for this is the French interface. In fact, this what we call MDS metadata store is kind of central portal allowing access to the metadata from all registered objects. In particular, it's can be harvested via or I or I or I PMH. So this is the you find this the schema on the on the website. We are at version two to but the version two three will come out in the in the coming months I think you have also the XML. This is the document you can you can visualize on the data site on the website. This is the XML version that you can see as well. And this is the search interface. We have a simple search here. We have also an advanced search and I as an example I looked for for the from the advanced search. I looked for I looked for Minio and I wanted to see a data set. Not sure because the general resource source type can be things like text of video. Yes. Databases and we have general resource type and we also have resource type which a bit more specific. You can have thesis and reports and things like that. And I saw just one result. And then if you click on the do I hear you go in fact to the landing page of this data set. And this was a data set hosted by the UK data archive. And they received the do I from the British library and then you can download if you have the the right to download. Of course talking about open data we always encourage all our data centers all our customers to make the data as open as possible. But this is not always the case. And at least there should be some information on how to get the data. This is something that is made precise in in in the contracts that we signed with with data centers. This is some information about harvesting that you find on our website. And then something that this is becoming slightly a technology more technical data site content resolver. It's a service for displaying data site metadata in different formats. And if you make a search you you see all the formats that in principle are available. And this can be done. This is also done by Crossref and even in cooperation with Crossref. This can be done either by something that is called content negotiation. It becomes a bit really a bit technical documentation that you have on the bottom is about that. Or by using the DUI proxy the DUI proxy is the DUI formulation as in URL. The form you see here is the sort of gateway between the handle server system and the HTTP. And you can put in a URL of this form and then specify the type the mime type type which is the the the internet format of the of the resource. It can go directly to the data to the metadata in the requested format. So I have two slides about explaining more more detail how content negotiation works. But I think I have to skip them. These are the slides. So I go don't go into the details here. And just to mention that we have also a list of repositories of research that are within our services. And you may know that Thompson Reuters has created launches like data citation index and they started they all want to work with data site. They started to to to all these repositories that were listed in on our website to to get in touch with as a starting point. Orchid and data site work together in a European project called Odin. It's about connect information across multiple servers and infrastructures for scholarly communication. Orchid is the author or contributor identifier initiative. And we are also involved data site is involved in a group in a co-data group on data citations standards and practices. Co-data was mentioned yesterday by Jeffrey Bolton together with Ixu as one of the international bodies who can could possibly do much advocacy work for open data. And I think this was almost the last one. Yes, and we had just a colorful slide with all the logos of our current members. Thank you.