 today. We're really happy to have you here. What we are going to show you today is first of all the housekeeping rules, let's say. So this event will be recorded. Your microphones are off by default. But of course, if you want to participate, you can use the chat, you can write a question, you can interact with others. Of course, you can use the raise hand functionality in Zoom, and also all the presentations and recording. Don't worry, they will be shared afterwards on the event page. Also, please remember that we are very social and we like social interaction, so you can use the Open Air Nexus hashtag. My name is Andronikio, of course. And now about the program. We start in a few minutes with Paolo, with a welcome on the presentation of the Open Air Nexus project. Hope you enjoy it. And let's start. Paolo, the floor is yours. Thank you, Andronikio. Let me share my screen. So thank you all. I'm Paolo Manghi. I'm the CTO of the Open Air Infrastructure and I'm coordinating this project, Open Air Nexus. Open Air Nexus is one of the Infra EOS Go7 projects funded together with others to deliver services in support of the use and to contribute to the overall construction of the use. As the other projects has a duration of 30 months and a budget of around four millions, the coordinator is the Open Air Anke, no profit. And we have 11 partners. So many of them were already part of previous adventures of Open Air. We've been building services and digital services for Open Science together for more than 10 years. While others were new to this context and joined us to offer their own services in the package way in support of Open Science in general. So as you can see the project is highly tailored to virtual access and so service provision. There's not much of investment into extra functionalities or extra and extensions. The only portion of extension and enhancement that has been supported is the one for the integration to the EOS. To the EOS can its interoperability framework when this will be delivered. So just to give you a view from the moon, a panorama of where we are in this context, how Open Air Nexus and Open Air as a whole, in fact, is linked to the overall panorama of science. We take a look at this picture. So on the left side you see the general high level modeling of a life cycle. So where you have experimentation, you have analysis of the results, you tend to publish them when possible in early stage or in later stage. The idea is that you want to share with the whole public or maybe subset of it like a group of researchers you love, your organizations, but in general you tend to share. Also at early stage, the normally is the final publishing. Then you want these results to be discovered. So that's why you're sharing the metadata and all the possible things that enables the discovery. And then still on the left side, when somebody will find what they need to perform science, they will need to reassemble and experiment. So in this case, we have clusters, research infrastructures, storage and computer infrastructures, living on the left side of the picture. And AI, of course, embracing the whole picture. Because what is really important after publishing, or in general as a consequence of publishing, is also tracking usage, is also ensuring a quality assessment of the given results. And this is done in several ways through peer review citations or feedback and also monitor, for example, the results of this quality assessment as well as the publishing. So to keep track of all the aspects of science in general, the life cycle. So the principle is that whenever I publish something, I should be able to track it, link it to the rest and do this in a uniform manner, following standards. This is where Open Air Nexus leaves and stands and offers the services for especially in this project, we focus on the aspects of publishing, discovering science and monitoring science. So the project objectives are five, as described in the description of activities. The first one, as I mentioned before, is to support publishing services in support of the research life cycle as a whole. So the idea here is again to fill the gap, because publishing, of course, takes place also within research infrastructures already in data repositories, etc. But when this is not possible for several reasons that depend, of course, on the research infrastructures or the long fail of science, we offer services for that. Zenodo is an example. But we also offer publishing for other kinds of products like D&P's data management plans or data minimization, again, with the intention of compensating where the gap stands or supporting when this is needed. The objective, too, instead is about tracking and monitoring the evolution of science. These are two key aspects which we believe are separate. On the one hand, we want to track, so a key track of everything that is an event that happens from publishing, citations, interlinking, processing, usage. And on the second, in the second hand, instead, you need to monitor. So to come up with indicators that may result from all these basic facts, these information that we're tracking. These two activities require definitional standards, common intentions, agreements where possible. And of course, ability to monitor, so offer services that are able to cope with large amounts of data, stand with the requirements of those who need monitoring, like institutions, funders, arise, etc. Objective three is instead support cross-discipline discovery. So we are also aiming at providing discovery services across the different disciplines within the different disciplines, and especially across the different kinds of research products that are today available out there. We're not talking, of course, anymore about publications only, but also to a very large and broader spectrum of products such as data, software, protocols, virtual machines where this is necessary. And with introduction of the use, these will include and span over services, computing, resources, etc. The objective four and five we believe are, of course, critical for the use and for science as a whole. So defining commons, defining best practices, common intentions, and common ideas and agreements on how we should publish in order to make science open, fair, and reproducible. In some aspects, this is very specific to the disciplines. Of course, reproducibility, especially, is a very disciplined flavor in a way. In other cases, this is not a case, and where possible, we should collaborate. And here, Nexus, a lot to the use. Finally, the revenue models for open science. Open science is free at a point of usage, but of course has a cost as a price. And so we are also trying to identify the best ways for this kind of services, this category, specific category of services, to become sustained, to support the right and proper business models. So this picture that represents, again, from the moon, what we are offering as a project. We have three, let's say, sub-portfolios, which will come up into one catalog, the one we're offering with Open Air. On the publish side, we have epic sciences for publishing journals, overlay journals, Zenodo, which we all know about for publishing any kind of content. It's called the catch-all repository for that. Amnesia for the anonymization of sensitive data, and Argos for the creation, maintenance, and publishing of data management plans, which are machine-action. For monitor, we have a core service, which we call the research graph, which basically builds one of the largest collection of metadata on the planet regarding funders, regarding projects, publications, data, software, organizations, authors, and all the possible relationships around them that interconnect them. And through this graph, we offer a number of monitoring services to identify statistics about initiatives, research initiatives like research infrastructures, for example, the research impact, or the funders, so how many publications are to be have been funded by a given founder or data or software, for example, and the same goes for institution. To build the graph, we have a number of services around it, which are building sub-collections, which are strategic for the aim of monitoring. For example, Skoll Explorer, which brings in the largest collection of links between data and articles. Open citations, which brings one of the largest collection of open citations between publications and publications. Open APC, that grows a collection of a database of open APC charges for publications coming from the libraries, which helps us a lot in tracking how open science is taking up and which are the costs behind it. And the users count, which is a service that builds on standards for user statistics, which we are today using to track the user statistics for publications, but we are standing for data. Skoll Explorer, of course, exploits the graph that I just explained to offer discipline-specific or regional-specific, because we can, let's say, drill down to the scope of the country or continent, discovery function like this. We offer these also through APIs, as you will see, which are open and usable. And we also offer finally, and that's part of the discoverer, because we believe it's a key element, provide. Provide is a service that allows us, basically, to validate the content that we collect from the data sources that are part of the open air, the scholarly communication data sources from which we collect this data. We validate their metadata in terms of guidelines that have been defined by the community to align on common standards. And this is an approach that we'd like to extend to the experts as well. Now, where is open air and open air nexus placed with respect to the EOS Core and how we defined it? We offer platform services, as we said. Some of these are part of the EOS Core, and will become part of it. The open air research graph, namely, is the one that will offer the EOS Resource Catalog and support that provide, as I mentioned, will extend the sort of validation to the communities that do not offer these functionalities to include metadata information through the catalog of the EOS. We'll offer the open science observatory, which is a way to inspect the trends of open science in terms of openness, fairness, publications, links for the EOS. And also open air, as used statistics, which will extend to data repositories. The rest of the services, of course, will be part of the exchange, and these are already part of the exchange, where they're all registered. As a whole, as I mentioned before, most of the project funding and budget aims at, is aimed at operating the services, so virtual access and provision of virtual access, and part of it to the integration with the EOS. So the EOS is only partly establishing rules of participation for the moment. There are strong indications of what will happen, and of course, we are trying to anticipate some of the choices, and we strongly believe that picking standards, widely used standards, is the way to go in this case. And so, for example, on the left side, we are, of course, onboarding all of our services. We are making sure they comply with the AAIs, like EduGames and so on, as offered by the EOS today. We are also going in the direction of accounting and monitoring by making sure that all services are bringing indicators of usage, and KPIs as defined in the description of activities, of course, but through standards, which will allow us to then centrally bring to the EOS these results when the new indications and rules will be delivered, because we believe standards are interoperable in most cases. And semantically speaking, we're trying to adopt all the data models that we have to the indication of the EOS, which we are going towards. The service description templates, we're going towards the data profile templates being defined in EOS key lands, and we are building together the resource products, data models, starting from the ones of open air and converging towards new EOS can doors data models. The same happens for user statistics and other concepts, like wide and broad adoption of persistent identifiers, trying to embed fairness as a property of our data sources, etc. Finally, this is the last slide, which brings us all here, and we also have a session dedicated to this. But we also believe that Nexus will be a chance to effectively bring something to the EOS by collaborating with the rest of the Infra07 projects. This is key for us. We have a few ideas, a couple of ideas, and we'd like to propose them and see what we can do together, especially in the direction of enabling seamless publishing of research outcomes. So the idea is basically to make sure that scientists, while they are performing sciences in their own infrastructures, buy simple actions, delegate machines to publish on their behalf, and to do it properly, to do it in a way that is fair, driven, to do it in a way that is at least trackable at all times, so tracking the links between the results and in a way that would allow users at the final end, at the end of the stream, not only to discover these results, but also to combine them in proper ways, where possible to be brought back to the original research infrastructures where the experiments can be reproduced as a whole. We already have some experience in that, and so we'd like this chain that you see from the left side, where research infrastructures are and infrastructures are, to bring the uses of these scientific services integrated with the repositories, where the objects are deposited, and finally be published in the scholarly communication services, which include OpenAir, that reaches out to Scopus, to Orkid, to Software Heritage, etc. You're all integrated. So I am done with my presentation, so I hope you understood the flavor, what I wanted to say. You will now have a series of presentation going into the detail of the individual services that I mentioned as part of the architecture. And again, as Andreniki said, please write questions or ask questions when you think this is the case. Thank you very much. Fantastic. Thank you very much. Paulo, you have a question here. Where do preprints hit, and is it open to independent preprint repositories other than Zenodo? Yes, of course. So OpenAir is collecting from 14,000 data sources today. It includes not only Zenodo, it's just one of the 14,000. So it includes all possible journals out there, preprint servers, data repositories, software repositories. Whenever these are, let's say, sources that we identify as trusted by users, so used every day, then we want to include them. We want to include them, especially if they rely on persistent identifiers, that helps a lot, because whenever we bring in a data source to metadata persistent identifiers, this data source becomes interlinked with all the rest in the ground. So the data source gains invisibility, but also in content. For example, if we have the links that are pointing to your objects, you will get all these links. And we can also return them to you. We have services that do that. We call them brokers. And we can send back all the metadata that you need, for example, the open access version of an article you have, or the links to the software or the projects of your publications or data in your repository. So, yes, we are very open to that. Okay. The second question is about sharing the presentations. Yes, everything will be shared. We have one more question. Do you have one minute? Yes. When do you say data repositories, do you mean the ability to ingest actually, okay. So there is a difference here. So in OpenAir, we are not trying to offer deposition facilities. Okay. So we're not trying to replace what the research infrastructures are there for. We strongly believe that the thematic approach is the one to go. So if you have repositories in your community, this is where you have to deposit your data, your software, your publication, these are the right places. In all those cases, where these repositories are not available, and this may be well the case for some communities which we know already, then we offer them all. Okay. Then on top of that, we have, we are building this research graph. And the research graph is the way to bring the information contained into these data sources together in order to build the overall map of research, right? And how it evolves over time, how the objects are linked, how the authors are related, and so on. So these are two distinct actions. So by data repository, I mean fixed share, I mean pangea, I mean name one, PDB, PROD, all this will be part of the graph because they must be there to be interconnected. But we are asking the researchers to go there on deposit. So we're building the bridges. Okay. Thank you very much, Paolo. Let's please proceed. Okay. Let's maybe move on to the program and as we see the question in the next three, five, six. If you want to write the questions, I can reply in the chat. Okay. So let's continue now with the, thank you very much Paolo, with the public's portfolio where you will have a brief presentation of the five of the four services and not the epic sciences, Argos and Amnesia. So Jose or Alex, you can start. I will be. Yes. So hello, everyone. My name is Alex Ioannidis. I'm currently located at CERN in Switzerland and I will be talking about briefly kind of like I mean overview about Zenodo, called the catch-all repository for the long-term research. So Zenodo is a digital multi-disciplinary repository. So we don't serve only one specific community or domain. We have content from diversity, from humanities, from computer science, from health sciences. And Zenodo is hosted at CERN's data center, which has a long history of serving some of the biggest experiments in particle physics in the past years. And Zenodo is also a place for all types of research objects. So we don't only support if you're talking about the text, let's say some papers and preprints and things like this. We also support data sets, software, posters, presentations, anything that's part of the research process. And by default, we allow users to upload up to 50 gigabytes per record. But of course, users can have multiple records. And of course, we're also flexible when there's special use cases where larger files and larger data sets need to be shared. We try to have a very rich metadata schema so that data itself can be described very thoroughly and to be easily, let's say, shareable and searchable and trackable for other systems. And we also try to integrate very well with, for example, the funding agencies and grants and projects. So there is the possibility to have this holistic view of what the project's outputs are. And Zenodo, of course, is accessible via a web interface, but we also have REST APIs that make it possible, for example, for users to set up automatic workflows and automated workflows that upload or fetch or how let's say more complex use cases. And to date, basically what we see as a service is that we have around 50 million visitors per year. To date, we have almost two million records. And these records amount to about half a bit of the files stored in the Zen data center, which is just a fraction, of course, compared to all the rest of the data that the Zen produces. So why use Zenodo and who should be using Zenodo? Basically, Zenodo provides a reliable infrastructure for all researchers and especially for those who don't possibly have a dedicated domain or institutional depository. So it's a way to make it easy to share this data when it's not, let's say, it's not straightforward where this data should go. And also the idea is that it lowers the barriers to share in general data or software or basically any kind of research efforts. And the idea is that by sharing on Zenodo, these objects are automatically compatible with the fair principles. And it's also an easy way. We try to make it very easy and as hassle-free as possible. Zenodo also exposes user statistics. So there's a way to track the impact of what the results have. And we also try to integrate citations to be able to expose this for research to be able to interpret basically what's the impact based on this. It is part of FIOSC. And basically the value that is that it makes the process of sharing research as easy as possible. So it's a no-brainer to be able to use Zenodo when in doubt. You can always use Zenodo and always post your outputs there. And it can be used by researchers directly. It can be used by communities of researchers for specific domains or topics or thematic, let's say, cases. But it's also useful for project coordinators and PIs that want to organize all the outputs of a project, for example, in a specific place. To use Zenodo, of course, I mentioned there's a web interface and we try to make it as simple as possible and understandable to everyone. Of course, there's also a REST API for these advanced use cases I mentioned. And also we try to integrate with other platforms to make, for example, the process of researchers that develop the research software on GitHub, for example, they can easily archive it and so on. All this process for them to be kind of like easy and not something that they have to consciously think about and track. And now to kind of give you a picture of what exactly is the position of Zenodo in this ecosystem of open air and EOSC. Basically, as I mentioned, researchers can directly deposit things in Zenodo, but also there's many cases where they come through other services. So, for example, they use GitHub or they use, for example, Amnesia or Argus, which are services you will hear about later. And the activity they do in these services ends up in Zenodo eventually. And then we have also automated platforms that have, like in the biodiversity community, we have a very complex, we have a use case where basically there's an automated pipeline where the figures and other type of information in the data is extracted automatically from papers. And of course, everything that gets ends up in Zenodo, also ends up in open air explorer and, of course, the user statistics of all those and to the open air user service. So the key takeaway I would say is that Zenodo is a place where it's best free for everyone that's for sharing the data. It's made, we try to make it very easy to do that. And we also try to follow the best practices in doing that so that researchers automatically are compatible with this and we don't have to think too much about what these best practices are. And of course, it's more so more about just papers, trying to make it the place where every part of the research pipeline and every part of the research workflow is something that is archivable and something that can be put, can be tracked and shared basically in the open. And this is also part of our mission to promote the reproducibility of these results. And that's all. Thank you very much, Alex. And now we can move on. Alex, you have two couple of questions. So if you maybe you can respond one live and Okay, because we have a 10 minutes session for Q&A later. Okay, okay, that's one. Let's collect all the questions. Yeah, so type in your answers. I will type in the Thank you. So Rafael, please start. Can you hear me? Yes, clear. Yes. Okay. So good morning. I'm Rafael Torna. I will try to share my screen. Okay, so I'm Rafael Torna. I work for the peace sciences platform. It's an overlay journal platform. And what is an overlay journal? It's a journal whose content is hosted on OpenArchive. So for the peace sciences platform, we operate on top of open access repositories such as archive, CWR and very soon Zenodo. The peace sciences platform enables the management of the entire scientific publication cycle. So it means submission, certification, copy, editing, dissemination and preservation. The platform is a layer of services for the scientific communities and enables them to operate high quality open access journals with content hosted on the open access repository of a choice. By design, it is both gold open access because journal content is available in open access, of course, and green open access because content is already self archived in an open repository. We can also call it diamond open access because there is no reader fees, no paywall. There is also no auto fees, no IPCs, and even no fees for journals to use the platform. So why to use the peace sciences? Peace sciences is designed for the scientific communities. It is designed to save time for the readers because preprints are already available and all of the subsequent versions are already accessible during the peer review process. It saves time also for the researchers because documents are submitted once and are available in open access at the same time on the repository and on the journal website. Also it saves time for editorial teams because we provide an all-in-one submission management and publishing system. By design, it complies with third principles with both for the repositories aspects and the journal aspects. The scientific communities can control the anti-publication process. So this means the submission of articles, management of peer review, follow-up of reviewers, automatic renewing days, for instance also, and copying and finally publication. We provide the tools and organize the content at the peer review as they wish. Long-term access to the articles is guaranteed by the submission in an open archive. So whatever the evolution of the journal, if a journal were to disappear, for instance, the content will remain online thanks to the repositories. That was the content. The operating costs are reduced because we share an IT infrastructure with other public services such as the repository for instance and the hosting costs of preprints and articles are in fact spread over all the different open access repositories. The platform itself is hosted in Europe. So how to use IP sciences? In fact, in the journal has its own domain name. So you have to select the journal on which one you want to publish. For instance, let's try to read LMCS, logistical methods in computer sciences. It's hosted on IP sciences. This is the homepage on the right. So the first step is to submit your preprint on repository, say for instance, we will try with archive for this article. You see on the right the article, this is the first version submitted on archive. It has received an archive ID. So the next step is to copy and paste the archive ID on the submission page of the journal. You just have to paste the archive ID and the version number. So here, example is to submit the version one. The next step is the job of the journal, in fact, because there will be multiple rounds of peer review and multiple versions may be submitted online and on archive and also in the journal. If you look carefully on the right, now you will see the submission history of this paper and there is in fact 10 versions that have been submitted on archive. It means that could be also 10 versions submitted on the journal and reviewed. The final step is when the paper is ready to be published by the journal. Once it has been accepted and reviewed by pay viewers and the copy editing is done, it is published by the journal and you see on the right, you have the paper that is published on the LMCS journal page. It has received a DOI. It is in volume 16 for extent issue two. And these information are also automatically put in archive. So you see on the left, in the green box, you have the journal name, reference and the DOI. All these things are automatically added to archive by the platform. So what are the key points here? The key point is that science is an easy and cost efficient way to operate high quality open access journals. It is designed for the scientific communities to operate journals free to read, free to submit and free to publish from the point of view of the journal. We open to new journals or already existing journals. Please submit new journals or submit to our existing journals. And we are, of course, open to every scientific fields. Thank you. Thank you very much. Thank you. There are some questions already. We will collect more also in the chat. So we have a specific session. Let's move on to Argos. Hi, everyone. Good morning. Good morning, everyone. Let me serve my screen and make a full view. All right. Perfect. So good morning, everyone. I am Elpa Badapolu. I work for the Research Innovation Center in Greece. And to give you a brief overview of Argos, which to give a brief overview of Argos is really hard for me. So bear with me, please. First of all, what is Argos all about? Argos is a tool for planning a research data management activities according to policies set by funders, by institutions around open access and third data. So essentially, it's a tool that helps researchers create their data management plans. In short, DNPs, I will call them from now on. It's open source. It can be configured to different to different domains and needs. And it can be also extended through the software, the open source software repository. It's online. You can use it online by going to the website Argos at up in your data. You or a deployed at all institutions on instances. It's available also in EOSC and you can freely use it for research purposes as in your projects. Our user database is currently around 1000 users, 1000 users, but it's growing. It's fairly a new service so this is growing every day that we speak. Let me see. All right. So you might be thinking that this is yet another DNP tool, but I would urge you to think otherwise. And these are my arguments for doing that. So here, I've selected a few key points that show the value added of Argos and some extra functionalities that it provides to the open-sense community. So first of all, why to use Argos is because it not only helps create data management plans, but also extends and supports a full DNP publication life cycle because it integrates with Sonoto and it supports the publication process that is needed to then expose those DNPs once they are generated from Argos to expose them in their repositories in appropriate ways, according to, of course, open and fair principles and practices that are tied to this process. Second of all, the outputs of Argos are machine actionable in that they are not just plain text. They are more rich and they can be exchanged between different compliant, machine actionable compliant, let's say, tools. And one of the things that we are doing in Argos is we try to, through this platform, through this tool, we try to normalize the descriptions of data. So Argos pushes actually towards that not only, this means that it would not only support domain protocols and standards to be configured in the templates, but also we allow multiple data sets to be described in single DNPs. And this is important because the approach that we follow is data-centric. So different types of described data sets are handled appropriately within Argos because we see that there are different criteria that have to be met, and not only by researchers but also by funders. When we talk about produce data, different criteria when we talk about reuse data, sensitive data and so on. And this is very important because we try to give control over the individual dataset rather than having all data sets mixed in a single DMP record. Also Argos is interlinked and connected with reference services and data sources. We'll see in a minute how it is, where it stands in the open air ecosystem, and we also try to, we actually love collaborating with others. We thrive in collaborating with others and in communicating and exchanging ideas and knowledge with the open science community. So we try to get involved in this evolving, let's say, the environment of machine actionability in DNPs. And one of the things that we are doing, for example, we collaborate with Provide, the other service for from open air to try and influence repositories to apply the correct resource type in the metadata for DMPs to be properly exposed. Our main users are researchers, research projects funders, research communities and institutions, and here you see how our machine actionability DMP output looks like. Briefly, what you can do as an end user in Argos, here you see a DMP record, you see the basic of the data, the basic, here's the title of the DMP, here is the version of the DMP, we support versioning, the last time that it was edited, what is the status of this DMP, it could be draft, here is finalized, you can clone the DMP, you can delete it, you can see what grant is, it is associated with the researchers that have worked in this, in the data management activities, what is the description of a DMP, and how many data sets are described in this particular DMP. You can deposit it immediately by clicking this button, you create a record in Zenoro, you can undo the finalization and work on, finish work before finalizing it and deposit it, you can export in PDF text and XML and JSON, you can start new version, make public, so revert the private settings and make it public in the Argos environment for everyone to see, and you can also share it with colleagues so that you manage workload in this process of the DMP writing. Once you click the deposit, sorry, you will immediately get the reference, the DOI here and you can always view it if you go to your DMP record in Argos and click and get redirected to the, it's another record. Here is the same screen, but for data sets, so this is a data set record and you can see the difference because we are using different color schemes to note that this is data sets, data sets are in yellow and the DMPs are in green, again you can finalize it, export it all the same functionality supply, and what you see if you are working on the data set level is the template that is provided by your funders or institutions and you see on here, on the left hand side, you can see the different steps, the different sections, including this template that you can view the screen where you have all the information on how to do properly, let's say, complete and answer your questions and provide your input in the fields. Argos is not only provided in English but we have, thanks to our NOADS, the National Open Access in Opener, we have different localizations, you can see some of them here, so we make it easier for native researchers to use it and we're very excited about this, it's coming in March this month, the administrators interface, we have tried to simplify how admin users can create the DMPs template, the DMP templates in Argos, how they can configure their APIs and to control the RDA compliance, but for the latter we have to, this is done in consultation with us, with the Argos team because we have to test that it actually works very quickly how these are all linked together, how the different services of the Opener ecosystem are linked together, you see from Argos we have integrated Zenodo, so we publish DMPs as machine actionable DMPs in Zenodo, and sorry I have also these screens and you can see that before being published we integrate, we have configured some Opener and the EOSC API so that we make it easier for researchers to complete their DMPs by pre-selected lists and then make links with the research graph at the later stage, but once the machine action DMP is there we can then, and this is progress, this is working progress, we are working with Provide so that once a DMP mentions a dataset is going to be deposited in X repository, Provide will send a notification to the repository monitor so that they organize themselves and prepare for this dataset to be deposited in the repository, then we also send notifications, we work with the monitor to create indicators, basic indicators for understanding how different statistics get different statistics for DMPs and we work with funders also to get tailored to understand their needs about this and once Argus is published it enters the research graph, it creates a new entity in the research graph properly described and creates also links with different outputs, here we create links with datasets and also with projects that are associated with the DMP and all that are searchable in Explorer plus we are working with Explorer to create a tab dedicated for DMPs, the project coordinators will view all the different versions of DMPs in one tab and can report back to the EC, the takeaway is that, but I want you to remember is that Argus prepares all stakeholders for the next rising group of DMPs requirements and for the CHISTER, CHISTER a new call requirements, CHISTER is also another funder that we are working on, funder's consortium, Argus simplifies administrative processes and connects with university institutional workflows, so if you have own instances you can create your own let's say links with the local ecosystems and Argus also enables the implementation of the data domain protocols so that research communities create templates tailored to domain standards and practices, I'm very happy to get more get and answer your questions, thank you very much for now, thank you Ellie, thank you, we need to move in to move into the next one, Amnesia, so Manolvis, okay, I'm Manolvis Theravitis and I'm going to give you a brief presentation of the Amnesia tool which is a data anonymization tool, it's been developed in Athena Research Center where I work, okay, as the previous presenters I also have this problem of presenting all this in a very concise and short presentation, so basically what an anonymization does is data anonymization, now anonymization is a word that we have been using forever in everyday life and even in science, meaning several different things, what I mean here is a bit more specific is anonymization as defined in the GDPR and as technically approached in several research papers which is an irreversible transformation of the data where the guarantees that the anonymized data cannot reveal properties of the original personal data and this is a distinction from what GDPR calls pseudonymization which is what we very commonly mean as anonymization which is just the removal of direct identifiers but without any type of guarantee that we cannot return to the original data with linking to a third database which can provide us the links, this can be done explicitly but this also happens when the secondary identifiers like the date of birth or the zip code of where a person lives that can be used to re-identify the data, so the data anonymization performed by Amnesia is a transformation that makes, that transforms personal data to statistical data and I will also articulate this is a very important distinction because statistical data are no longer restricted by GDPR, now Amnesia is available the best way to use it is a standalone application, it is made in a modular way, there's a very clear distinction of the backend of the anonymization engine that works with the front-end through a REST API, the backend is made in Java, the front-end is in JavaScript so we have deployed it and it is available as online in the site but this is mostly for training and demo purposes, the idea behind this is that usually both technical and legal restrictions do not allow you to take the personal data out of your own premises so you have to bring Amnesia to your premises and not the data to Amnesia servers okay except this little type about year 2020, we really had 34k unique visitors in Amnesia site into 2020, I hope we have a lot more in 2020 and there's a bit less than 5,000 users of the online service, I guess that's people training, that's people using it in demos so why should one use Amnesia, as I said before by transforming personal data to statistical data the data are no longer restricted by the GDPR, so basically it can be shared an institution that produces data can give it to its partners at the university, can give the anonymized data to its partners at the university and they can do research on them so anyone who has data and wants to give them to partners or to a greater or smaller audience will be interested to use Amnesia, so it does irreversible transformation the data and the basic idea the key technical challenge is how to remove all identifying information but preserve all the useful information for research and there are many solutions to these different trade-offs and in the technical level this is where research and technical challenges arise, it's a value-added services for EOSC, we expect that data owners, data curators and researchers will use it and benefit for it, researchers are also of course it can be the same personal entities acting in these roles at the same time but the basic idea dream of anonymizing the data is that instead of having a limited quantity of personal data that can be shared only under NDAs and strict rules we can have a very big ecosystem of anonymized data where they can be shared freely, this data might be reduced accuracy because of the anonymization but they will make up for it by who will make up for it for research by having large quantities and the ability to do results, build models on let's say millions of data records instead of few hundreds of thousands, this is a very novel tool there are no established commercial solutions, there are a few I think academic tools that do anonymization of course Amnesia has unique features, it is unique in the approach of some things so it is one of the very few solutions in this area and mostly it is to be used by the data curator who wants to share their data, I think I told you a bit of how Amnesia is made, you can find it in amnesia.openware.tu, this is the online version, we do have limitations there due to limited resources, anonymization needs a bit like data mining is an expensive procedure you can use the standalone version through the graphical interface, you can incorporate it to information systems just using the backend tension through a REST API or you can even use it through our command line interface. Just a few screenshots to get an idea, the first thing you in Amnesia is you upload a data set, there is a wizard to pass this data and make Amnesia understand what the data is, it's a bit similar to the import wizard of Excel for text files so people using this kind of tools will be familiar, so the steps is that we put the data, we have it here, we create a generalization hierarchy, this means the rules by which that Amnesia will use to abstract the data to remove any kind of identifying combinations of data attributes and finally, depending on the algorithms, Amnesia will provide a whole lattice of different solutions where the user can choose different ways to anonymize the data where different trade-offs between data accuracy and privacy will be given and basically different areas of the data will be preserved. For example, depending on the research, you might be more interested on the area where a person lives or their age and you can decide which one of the two you will preserve in a better way. So the positioning of Amnesia in the open air ecosystem is, I think it's a bit clear, it's next to the data source. Basically, the data owner or curator anonymizes the data before making them fair before offering them to the community. The idea is that the data owner has personal data that cannot be shared, so it anonymizes the data with Amnesia and gets statistical data with the anonymized data and then they can be inserted in the air ecosystem. So we expect that this happens, for example, before uploading them to Zenodo. Amnesia actually has in-head support for Zenodo, so you can anonymize the data on your premises and then directly upload them through Amnesia to Zenodo. So take away, Amnesia does data anonymization, which is a different thing that should do anonymization. The resulting anonymous data from Amnesia are no longer personal data, they're not restricted by the GDPR. Amnesia, as it is now, it is a free tool, you can download it and use it on your premises and it is one of the very few solutions that do data anonymization. There are several solutions in commercial tools that help obscuring some properties of the data, doing masking, doing hashing of some field, but none of them does data anonymization in the sense that GDPR defines it. So I guess that is all, thank you very much. I see the chat flushing, so I understand there will be questions, but I think Androniki will organize them. Yes, Manol, most of the questions are being answered, so just have a look and pay attention to the chat, because if there's any question, you can answer directly. Most of the speakers have already answered, so it's quite active. Okay, thank you. I'll stop sharing. Yes, and now we move on to the Q&A, as I already said, most of the questions are being already answered, so we're going to record the chat, so in case you have not seen all the questions, because we are running a little bit late, but now I need you to go to Menti, we have some questions for you, but in case you have some questions to raise to one of the speakers in the reactions button, you can raise your hand or open your mic, so feel free. Yeah, please, let's see what you answer also in Menti, because it will be really nice for us to understand what are your interests, what is that you like, what is something that which service you would prefer to use, and to understand more about you, your profile, you know, preferences, and to see that you really did understand what we described here with the four services. I see moving, oh, Zanotto is a winner, I see here. Yes, I've already put the code for the ones who haven't seen them, go to Menti.com and then use the code, go to the chat. There's a question for Manolis, I don't know if you want to answer directly or open the mic, Manolis. Hi, I will open the mic, yes, so the question, it's not just for quantitative data, it is also for qualitative data. For qualitative data, the difficulty is that someone must usually create the generalization rules, these rules for abstracting the data by hand or in the case where, like the example I presented where I see the code, which is a predefined ontology for medical diagnosis, they can be used, they can be taken from some online source, so this is a more difficult challenge for qualitative data, but Amnesia does both, and we do have documentation online and we're adding new material all the time about how to use the Amnesia tool. There are a few webinars that are also online and we plan to add some more videos with tutorials. So as we see Zanotto is the most known service here. You know, it's going to be one of our goals maybe. We can repeat these questions at the end, close in the middle of the project also in another event, and see how many they will use also the rest of the services. So can we move to the, yes, yes please. So I'll do globally access the value or the interests of these services. Let's see out the reactions. I've already retype again the access of Formanti and the code, but you have here on the top of the slide also. Again, it will be nice to have more than 30 answers to have a good view. Yo, you're 21, let's see. It's like the racing of the services. Zanotto is also the most, let's say, general purpose services among the four, right? Yes. It's usually by all researchers in terms of functionalities while the other three are kind of scoping down to specific categories of researchers or officers or so that's more reasonable. Yes, and Argos and Amnesia are more recent, yes. Well, it's not just the the the tiny, because these are functionalities that are useful and known, but if you take 10 researchers, how many of them need to anonymize their data? Maybe half to one, right? Not one. And it's the same for Argos. How many need to manage the data management plan? Well, they all need, but typically is one that does it out of a research group, right? And it's the same thing for the sciences. Managing a journal is not an easy thing and requires experience, skills and time, right? So again, these are key services for those who need them. But if you ask them to approve, this is the right percentage, I think. I'm actually surprised that the whole demo is so high. We can move to the last one. So we have one more question for you. In which way these services are useful in your daily activities? Your daily activity, so something that makes your life easier. So what you do, your work, saves time, effort, money, resources, very important. Just try to think like you have one of these services and what will be really nice for you? Find the interesting publications, okay. Free and professional, great tools I can rely on. Save, produce data and services for Zanotto and Amnesia. And don't end up in jail with Amnesia, because if you use other people's data, you have a problem. Serving project outputs all in one place for Zanotto communities. Supporting compliance with European Commission requirements. There's another repository for the results. The point of users science should be free, sorry, free. Yeah, open free, open access with the proper licenses. Can you scroll down? Yes. Zanotto with community. I see many people say about Zanotto community functionalities. So this is very important. This is a good feedback for Zanotto also. Yes. I think it's a key functionality and today is not reaching its full potential, it will be, because I know that Zanotto is working on an evolution of the concept of community. So Alex, if you want to say more. Yes, exactly. So currently we've come communities act a bit more as collections, let's say, so they're not very, it's sort of a very collaborative, let's say, feature. But we're going in this direction where, for example, there's the different roles in communities, there's many members, they can curate things, they can, there can be a discussion basically happening on the metadata and what kind of records so it's also a big part of having quality metadata is to have curators that know how to describe the objects very well. So this is kind of the direction we're going now. Towards the end of this year or the beginning of next year we will start seeing some of these new features. Thank you. I also see about Amnesia. Yes, Manolia is another two more questions for Amnesia. But I don't know if you want him to answer it publicly or just in the shot because we are running a little bit late. So we're moving on. Another important aspect. Two minutes. I just want to mention one thing, another important aspect of Zanotto, of which many of you may not be aware of, is that it offers the position facilities, as well as search facilities, also via APIs. So you can integrate your services to Zanotto. So if you have a thematic service that produces a dataset, you may easily extend your service to publish these datasets in Zanotto via APIs. And this is what many scientific services are starting to do out there. They publish their software, their software releases, they publish the results of an experiment. They publish their research objects as a whole within Zanotto to get a GUI. And this is done transparently from the user point of view. So we're hoping to discuss this further if you need it.