 So to introduce ourselves. My name is Liam. This is Valentin. We are from Europeana in this hat You probably know some of us either of us or both of us with different hats They we are representing Europeana the metaglam of Europe Elects the records of cultural heritage institutions of all types across Europe to make them visible searchable discoverable in interesting ways that they would not necessarily be Just on the collection of their own institution. It's a European Commission funded project. So public European cultural heritage for the win I will be handing most of the presentation across to Valentin who is the data coordinator for Europeana and You'll be presenting the kinds of things that we currently do and wish to do with Europeana and Questions about how we should move forward in that direction in the future Valentin Hi So yeah, so the idea of this session is we wanted to do a small presentation to put things into context But then we wanted to give you the floor for more questions so because we really want to have a discussion with you and in this discussion in fact, we would like you to You could sync with different point of view So this presentation is the Europeana point of view But we would like also to have feedback more from the glam's part But also if you are we key data again What do you think also about the the different thing we are going to present and the question we are going to propose to you So yeah, so quickly. What is your penalty? I'm already the short introduction. So You open eyes the platform for Europe Digital cultural heritage. So we aggregate metadata from all the EU countries so at the moment it's around 3500 galleries libraries audio visual archive archives across Europe We have around 53 millions objects and we have to cover Around 50 languages. So it's either the languages of the objects we collect Metadata for all the language of the metadata themselves. So this is an extra challenge For us. So you can imagine when we collect all those data We are also collecting huge amount of reference to Places people concept time, which is also what you have in wiki data. So The idea is So coming back to 2015 so there was a Europeana wiki media task force looking at how wiki media could work With your payana and and this report gave several recommendations and one of those recommendations back in 2015 was make wiki data a central element of the platform at a portal to platform strategy for your payana So moving your payana into not just being a portal, but also becoming a platform. So Yeah, we are not saying anything anything new. So it was already a recommendation that was was down a long time ago and now the idea is to Slowly but hopefully surely to realize this this vision we had in 2015 So what we do at your payana at the moment is trying to Move towards a link data strategy and trying to use a bit more all these data we we collect About yeah references to places people etc. So what we are building currently is what we call an entity collection. So Beside all the metadata we collect about our object the descriptions for the book the painting We are also collecting data about specific entities and the idea is to build a service that acts as a centralized point of reference to access the data about those contextual entities So we want to use them as entities that can provide context to the object We are describing on the objects people can access through the your payana portal And this would become a sort of your payana knowledge graph So again an idea that is very similar to what wiki data is in some ways and this is not new So you have the Google knowledge graph So decide the idea is for your payana to have a kind of knowledge graphs that then we could use Within your payana within the your payana services to build different kind of cool stuff so from for example using the entity collection as the base to have Entity base facet in the your payana portal Using it for semantic and metadata notation creating entity pages. So the example is the the first The first relief of the first release of the entity page we have in your payana So they are very simple, but it's it's a starting point or using this entity collection also to feed some Services such as auto completion. So when users search in your payana, they can have this Suggestion for entity and then move More and more from a free text search to an entity based search so What data we have so far so previously to this entity collection your payana was already Was already starting to do some data enrichment and to do data enrichment we We started by linking our data to some quite Big corpus quite generic because they were giving a nice coverage of the entities your payana has and also because your payana has To cover many languages. So when looking for Potential potential source for linking we were looking at those big corpus that provides a lot of languages. So For places we are linking to geonames For the agents to dbpedia and for the concept to dbpedia. So this was the first Base for the entity collection So what was done is all these data that we were using for data enrichment Well, this data was indexed in your payana and then we have separate We have used this base to build the entity collection and now the idea is why not making wiki data? The base for this entity collection. So it's at the moment. You see we use a lot of dbpedia and then the idea is to Get into the entity collection more links to to wiki data and So we are ready We have already starting to do it So this is few screenshots to show you how it is at the moment in your payana. So again, there is not so much wiki data This entity collection is based on the data We had at the beginning so all the language variants for example are mostly coming from dbpedia at the moment So you can see the data representation. We have the Creator pages of for example this one for Coco Chanel. So it's very few information But we hope that those pages will gather more and more information and then you also see the Suggestion the auto suggestion when you search in your payana We are also using wiki data in other ways so beside the entity collection At the moment we are using wiki data more and more to support Annotations so either to tag your payana items So we have this mini service called your payana radio where basically we wanted to offer the possibility to people to tag Sound item with sound genre and we have used those from wiki data We are also linking your payana items to external resources. So either wiki data resources Hello to wiki media comments and of course Even so at the moment we are linking your payana to wiki data But we know that some people have helped us to do also the linking the other way around so because this is also Really valuable to us, but unfortunately This is something we might not have time to we might not have time to spend on this So it's also a question maybe for later how we can collaborate in a way that maybe if glams makes an effort to Link to wiki data how we can also have the linking back And I don't know maybe the wiki data community can help into that because The glams might not have the resources to to do the linking in a in both ways And so this is an example so the your payana radio so you have a sound So a sound recording and basically we are offering the possibility to tag the item directly with sound genre from from wiki data and then we are showing the user annotation in the portal and Also linking to wiki media comments for images So So far so we have done this linking We are trying to link more and more to wiki data But the problem for us is how the collaboration should happen and how we should Try to to use how we should try to use wiki data So even though your payna is doing this so we are doing these strategies to link to wiki data because for us It's it's easier and wiki data is is the resource we need it's big covers a lot of languages The problem is we work a lot with glams that might have smaller vocabularies smaller Teaser eyes that are very relevant to us, but we can't link to those because it takes a lot of effort So another part of the discussion is how Your payna can also benefit from this work from glams through wiki data So whether we should encourage glams to also use wiki data for their own Vocabularies so that services like your payna can also benefit from those through wiki data So maybe yeah, it's maybe for later for the discussion to sink around this kind of Virtual circle how wiki data can how your payna can benefit from wiki data and also benefit from Resources from smaller glams and how wiki data can also reuse those resources because I guess having links to more Maybe control us authorities in wiki data. I can also make wiki data Can also improve the credibility of wiki data as a source of data. So so yeah, this is a discussion and Looking at this example of linking to wiki media comments also bring us to another topic another project Which is the structured data on comments because this is also something that is going to be become really relevant for your payna And maybe to the glam so how to link to digital resources From the your payna perspective We are looking very closely at how the project is evolving and we can already say that one point that is Going to be really crucial for from our point of view is if we want to reuse the work Of structured data on comments interoperability will be crucial and especially in term of right statements because basically If structured data on comments doesn't provide clear rights information in a structured way of representing those rights We won't use it. So it's something also to to to include in the discussion What should be the requirements for wiki data for wiki media for structured data on comments so that's all the in Projects such as your payna or the glimes can also reuse it because it's not you just create something and you say well use it We have also to think about the the requirements. So maybe Yeah, the Difference from for your piano in strategy that are in commons Compared to previous work with wiki media comments directly is it? It's the first time that we start to talk about things that are often work or in copyright Previously, your piano obviously with wiki media commons can only talk about and utilize Out of copyright or freely licensed Works under creative commons now we can start to talk about wiki data items and entities and things that are often works Which is a whole new layer of copyright shenanigans that we've never in the wiki media community Had to deal with before because it's just out of scope But wiki data means we can talk about Contemporary artworks as well and use the use metadata for contemporary artworks. Just not the images of them So it's a new field of copyright wiki media is not experienced in yet Yeah, so the question is whether initiatives like rise statement dot org that provides structured way to represent a rise statement can be something useful that Structured data on commons could reuse for example So your Pena has also participated to to other activities such as your Pena 280 that Allowed the creation and improvement of some wiki data item related to art and we are also trying to Try to start the discussion and for example, we have written this blog which was explained to glance partner how they can align their Authority authorities using mix the mix and match tool so So yeah, this is a kind of summary of where we are and what we do So of course you can already try the entity So the entity API which allows you to access to the entity collection data So you can find all those links in the ether pad that is attached to our session And we will we are there with with Liam. So if you have any questions or feedback, please let us know so The discussion we wanted to to ask you today And this is going along the points that I've tried to raise during the the presentation So how should your Pena and its partner glance better integrate? With wiki data how wiki data can benefit from your Pena's work or also other glass work in this area and what should be the concrete step for participation because quite often We we want to have some initiatives such as project or use wiki data into some campaigns But it's not always clear. What are the steps we need for participation and what are the different requirements such as for example If you want to link your data to wiki data The first step is to create this ID that will help you to this wiki data ID property that helps you to do the connection So if you don't know that You are a bit stuck. So this kind of requirements and then also starting the discussion on structured data on comments how what about the integration with this project because I guess the experience with wiki data can also help us in Trying to have a better approach towards integration with this new project and try to accelerate maybe a bit more the adoption Of this so yeah, those are questions for you Our job is done. I would say from now and and yeah, so I don't know if anyone would like to Just answer to one of those questions or share remarks or anything that can help us to to have this Discussion together and I think the presentation after also coming back on this topic So maybe we will have time in during the session to to improve on that So if anyone would like to respond to these questions or and ask a question relating to the presentation Stefan while the microphone comes to you remember that your piano is not a glam in the traditional sense of a cultural heritage Institution with its own collection we aggregate through national aggregation authorities of Europe The collections of other kinds of institutions. So the diversity of content and input metadata is very high Some people think that that's the diversity of that content therefore the inconsistency of that in content is our fault not really what we try and increase the quality and Detail and openness of that input data But it is in our interest as Europeana to encourage more Openly accessible high quality standardized metadata from the institutions as well Stefan yeah, I just wanted to break the ice to be the first. Thank you end of the discussion. I think I Would really like to see efforts in two Areas once is is to use Europeana and the whole network behind it the aggregators you have mentioned which we are one in Germany to use the the experts the metadata experts like Valentin who are natural allies are drawn to to wiki data to use them as multipliers, I think Daniel meet you talk in a lightning talk before about these these engagement projects how how we organize outrage for wiki data and I think one could switch a bit from the Wikimedia and residents to to train or to to to to Make the people working in the glam aggregation institutions Advocates for wiki data, and I think the willingness is really really there Lots of people are quite keen. It's not so easy. I myself not being a metadata expert, but Quite worse in the wiki wasm still find it There's a steep learning curve and if you have have the the luck to to be guided into the world of wiki data It gets quite easy easier if you know a person next to you working in the same museum That's something that helps to to break the ice. I think and the other thing is tooling, but we will talk about that later so just in term of collaboration so Because I'm not sure so I think your ideas are really good Then I'm not sure at what level so when we say glam's collaborating with wiki data for you. It's would Should glam's directly interact with wiki data or do we think work towards? Linking the authority files that already exist in the libraries museum to wiki data would be more relevant because I think From what I've heard this morning There is different strategy either doing collaboration directly by editing wiki data or Linking so I don't know in the glam world Where the forces should be and and how glam see that and maybe from a wiki data perspective What do you think is the most useful for the for wiki data as a as a platform and as a Database I think it's a there's still a difference of cultures and if you don't experience The wiki version and wiki data and the specific culture And I think it helps if you if you build your queries and and try to edit stuff and use the tools available Even if you just want to link to it. I think to have a have a Force of people that are well-versed in the wiki way and and can communicate I think it's really a bit of a cultural difference because And that's why I think it's good to speak the language and I think you need to have to be a bit of wiki data fluency Even if you don't want to want to go that far and especially the multipliers should be it should be we should be We should be well-versed and be proper wiki data people In order to to to have the movement. Can I summarize that as your piano should be using its position in Relationship to national aggregating authorities to talk to them Not spending our time so much trying to advocate wiki data necessarily to the individual institutions But we should be trying to increase the training and awareness and technical capability with the national aggregating authorities That's okay. Do we have any other we have a couple of minutes left. We have a another question. Yeah Mm-hmm What kind of tracking tools do you have to see the penetration of wiki data in the Glam's and in your a piano the different databases there? Is there like some do you have something that says over the top hundred Glam's in Europe? Such and such percentage of the databases are actually integrated with wiki data, so we know where to focus I Don't think we have done the this work of Of really looking at the numbers. I know already that there are a lot of big Vocabularies that are already linked in wiki data. We will hear about GND. We have 80 But as far as I know in the inside the data we receive and we aggregate I don't think I Would say less than 10 percent really has wiki data links in there So in the data we receive we don't see them much unless there was some campaign And then we know why people have added the specific links So so yeah, even and and they don't and also there is not so much discussion on The institution letting us know that this has been done so we don't necessarily know so but on the tracking and statistic itself we haven't done that work yet Would like to respond to the first question So how should your piano? That's partner glam better integrate wiki data I think you should your piano is mainly a very network of Glam's and there's a lot of context you should encourage good things within the network like Have everyone develop an open access policy make it clear how we can reuse their data Have them use stable identifiers would also be nice if your piano started providing stable identifiers We have a lot of broken links right now those kind of things. I think that scales much better than Then trying to do an end-to-end problem But for the broken links, it's not always our fault because basically the link to the objects are the links to the resource from the Institution so if they change the link and they don't let us know That's why the link seems is broken. So that it's also part of another collaboration point. I guess Yeah, I just want to say I missed the first minutes of your presentation But I'm really looking forward to the entity collection coming coming together because that's going to be like an aggregate of the terminology of European culture, so I think we should really Try to work together and you being able based on the work that we do on wiki data to clean it up and to link it together because that's really Powerful link that we can make so I see lots of potential there Which events? And just a quick point really about the the first point there The National Library of Wales has got several collections With detailed metadata and they've been shared with your piano But there's there's nothing on wiki data or in your piano to connect the two things together So we definitely I think I don't know whether it's a matter of creating Identifiers in wiki data to show that something's part of European or as well as a glam collection But I think it would definitely be good to develop some way of connecting connecting these things together There is already the European ID property in wiki data so And obviously an image of a painting is not necessarily this and a wiki data item Are not necessarily the same thing there might be 25 images of that one painting and Europeana might have 25 other images of that painting, but there already is the wiki data property Europeana ID so that can be used as the The linchpin for that that triangle Yeah, it was just a general point about the The some that you mentioned about getting you know facilitating the links from their side into wiki data like or you know if they want to handle on one side Because I feel like it's a really low hanging fruit for all these places that are worried about copyright They have to give anything away Just focusing on that means that they end up with this massive mapping and the minute They're ready we can just you know click the button and pull in the identifies on our end But I really feel like it's a it's a powerful thing to encourage So if your piano was encouraging that it would be your real benefit We do require that all metadata is cc0 coming in from the institutions as well. So it's all cleaned But but it's the kind of thing that I think should be Promoted in a way, you know, because if a museum has a bunch of links to wiki data ready And they might not have the time to do the the the relation in the other way on wiki data Who should we contact because I know there might be some of you that could do that after their work in the evening, you know And then in a way doing it in a much faster way than Me at your partner trying out for two days, you know, so it's just I think I mean as you said That's why the presentation was called low hanging fruits because I think there is a lot of potential there But sometimes might be this tiny thing That we might not know so so yeah if I don't know it could be also on the wiki data side Whether there would be a place where we can ping someone and said well, I have all these link and you do a burn back a batch upload to wiki data to have those those relation back to your partner that that sort of thing so I think that could be The thing the community can could work a bit better in terms of communication Yeah, but just yet absolutely but I think that's the other crucial thing is hearing the communication of what you need And I think the the person having a point of contact is really critical and that will be the last question So I don't know about the Your licensing of course for for the images and all you could have different license What about the metadata? Is it CC0 all the time all the time metadata requirement of of sending into Europeana metadata is CC0 the license of the content and the license of the original object we have a lot of Political statements or policies in Europeana to encourage open licensing of the material as much as possible We have a position statement saying that you may you should not claim new rights Okay, just because you own the object and you digitized it. Yeah, but Europeana does not have the political power or to enforce that Okay, no, we encourage that but the input information and the copyright information coming in from those institutions about what kind of reuse they are comfortable with We encourage more open, but we can't enforce that and we have tracking levels of what proportion of our collection Has an image has a high resolution image has a reusable high resolution image, etc. Etc. And we are always encouraging More of that the metadata Okay, so it's easy to if you want we can import it to the big data Yes, okay, and obviously the quality of the metadata depends on the input information. We don't Say that's wrong. That's their responsibility for their collection. Okay. Thank you. Yeah, thank you Next one