 So, I'd like to introduce those new panelists who have just joined us, Kevin Ashly, he's director of Digital Curation Center and he is an agrient capacity and capability among research communities in matter of digital curation. Previously, he was head of the University of London Computer Center and he was also involved in the National Digital Archive on Data Sets. Ruta Petrom-Skyter is a professor of linguistics at the Department of Lithuanian Languages without Ruta's Magnus University and she's also a vice chair of research council of Lithuania. She chairs her committee on social science and humanities. So Ruta will be wearing two hats, a researcher hat and a funder hat. And you already know Lars, so I won't be introducing him. And last but not least is Damien Le Carpentier and he's project director of research infrastructure unit at CSS in Finland and he's also a project manager of EUDOT. And if you don't mind, we'll start with some questions to the panel which I'll ask and then there will be an opportunity for you to ask the questions and we continue with this topic of infrastructure and sustainability. And my first question is perhaps to Ruta and Kevin. So perhaps to Ruta as a researcher, how do you as a researcher benefit from infrastructures which are already openly available and what's your experience with them? So perhaps I know if you could start and please use the mic as I think it should be on. Hello to everybody. Well the question is self obvious. We benefit in a number of ways, first of all, as data users and then as data providers. As data users, we are very happy to have an access to a wealth of publications that would have not been accessible otherwise still having quite limited library funds. As data providers, we could add with our own publications especially from journals that opened up recently and were submitted to DOAJ. And I hear it from my colleague researchers with joyful exclamations that their papers published some time ago have been kind of recycled. They were published elsewhere recently with a very good impact and new possibilities to start a scientific dialogue. And if I could add it from the funder's perspective, well that's also very fascinating because we have a possibility to disseminate the results of nationally funded projects on national open access repository and also internationally. And what I'm extremely keen off is the possibility to get some feedback for the evaluation, for further funding activities, meaning evaluation that based on research community commands for the open access papers. Thanks a lot for that, perhaps you can continue. So my reflections on the benefits are really, I guess really from my perspective as a citizen rather than given that I'm not a funder, I'm not a publisher, I'm not one of the other actors you mentioned in this area. And I'm thinking particularly partly because my organisation care is mainly about the management of research data and the answer to the question is going to really be thinking about open access to data infrastructures. There's no question that I know of one reason I benefit as a citizen is that the money and it's mainly public money that's spent on research is shown to be far more effectively spent if we ensure that the knowledge about even if data itself can't be open, the existence of data, the metadata record about the data can be public knowledge and that's made available and where it's tied to publications about the data. We have lots of evidence that this reduces the incidence of research fraud. That's good for all of us in particular case, it's good if you or I need a medical procedure that depends on the results of research, I'm more comforted if I know it's less likely that there's some fraud behind the treatment I'm receiving. But I also know that research is likely to happen more quickly and that's also a more effective use of public funds if we make data openly available and for individual researchers and for the university that I work in. We know also that our research is more likely to be more highly cited if we can make our data available and if we can link it to the publications. And in some cases also one has to, in the past when I did carry out research, you don't always get a publication out of a research project. Sometimes you collect a lot of data and you don't find what you hope to find and you don't end up with something that's usable for you. But it's quite possible that the data you collected does tell somebody else something interesting that there's a story in the data that you can't find as an individual but somebody else can and if there's infrastructure to make that data available even though you didn't publish anything somebody else might do so. And that's good for one's own reputation and it's good for society as a whole if we can do that. Thanks a lot, Karen. Perhaps you can keep the mic because my next question will also be perhaps to you and to Ruta. So we've talked about those benefits and how does it relate to potential funding sources? Okay, I can speak in particular then about experience in the UK where there's been a lot of debate at the moment about who is going to pay for keeping this data available. Most of the research funders now have requirements that are tied to grant funding that says you need to make the data available for some period of time, a minimum of 10 years. Not just 10 years after the research is done but 10 years after the last time the data was used which of course if it's good data and has much reuse potential that could be many, many more than 10 years on. But from what we can see at the moment it's not really that there's a lack of money to pay for this. The money is there in the system one way enough and also the evidence we have for instance on the value of existing research data centers, things like the social science data archives, the archaeology data service in the UK shows that we get much more value out of those data centers in terms of more efficient research than the money we put into them. So this helps convince funders that this is money well spent although there's always an argument that if you spend 100,000 pounds or euros on keeping a data center going that's money that you could have spent on research but you didn't. And it's important for us to get that evidence that says no actually in the end that does make more money available for research because the research becomes more efficient it only seems to be taking money away from the research budget. But certainly there are arguments also about that it's important for universities to use the money that they get out of the overheads from research grants and although the details of this are different in every country there's always some system like this where there's money that doesn't just pay your direct cost but pays for some elements of keeping the university running. Universities initially see again that this obligation to keep data as an additional cost the more they realize about how it increases the citation rates of their research that makes their own research more effective they begin to see this not as a cost but as an investment and again that they will get the return back from that funding. So it's it was a worry and it was a problem for many people but I think it's rapidly moving away from that. What could I add from a funder's perspective actually we have a very favorable laws for open access for both data collected during the project period and for publications. Well publications are easier to deal with we have where to put them openly they're both on institutional and super institutional level but data is our headache at the moment we are in some kind of a crossroad we have to work out our open access data policy to be conscientious and to stick to our own requirement to have the data open. So we have a few possibilities and we try to follow the first one. Well to include our data and to participate in S3ERIC projects like Clarine European Social Survey it says that well we have a roadmap for that and our research council has a special commission for the expertise of national partners willing to join the infrastructures but it's good for social sciences and humanities but it's subject limited so to say. Not all the research data that were acquired during the project period could be fed into these big infrastructures that also will check the data for quality and provide quality sales. So we have to have something of our own either on institutional level or on national level to deal with the data it's a well kind of an urgent need because we started five six years ago the competitive financing of research that kind of implied this openness of data so now we have the first results the first data still under the embargo period but okay we have to decide and well open air well strategies and seminars they're very helpful for us to decide how to deal with it. On a final note I'd like to say that well okay it's a kind of generational gap with our researchers younger researchers they are willing to search for well existing data and to use it while the older generation they prefer self or home made data and this is a big headache because they are so varied so different and so difficult to make to make into something compatible. Thank you. Moving on to our two other panelists Lars and Demi and the question I'd like you to answer we've heard a lot earlier about stakeholders that we should be engaging with stakeholders so my question is who are those stakeholders we should be engaging with and we I mean those who support open access infrastructure being that publications and data with whom should we collaborate and partner how we should set up governance and coordination system so perhaps if you could elaborate a bit about that. I could give a go on that. There's work going on in knowledge exchange which is a collaboration between national agencies in Finland, Netherlands, Denmark, UK and Germany. There's been produced a number of reports there'll be a meeting in February where potential funders will be invited to discuss how to actually establish a coalition of shared services and funders to find a way forward for establishing this open access infrastructure and of course there's a lot of important services out there that are developing and operating in splendid isolation but we want a more comprehensive open access infrastructure to develop. The big challenge here is that despite that science Europe and even the Global Research Council are talking about promoting open access and so forth it's very difficult for them so far to realize that creating an open access infrastructure requires investments and this is really a big task for us to make the case for that. It's kind of strange because if we look at the current infrastructure for the predominant model for academic publishing the research funders are already funding that. I mean Thomson Reuters, subscription agents, what have you, it's funded indirectly via the research funders so now we should have to find a model where we can create a coalition and define of course which elements in an infrastructure would be necessary. So we have a huge task there but definitely my target here is to bring the research funders to the table and make them realize that this despite their mandates and support for open access repositories and germs this will not in the long run work unless we have investments in the things that would make this work together and be a comprehensive open access infrastructure so it's the research funders that should come to the table and pay the bill. Thank you. I think about research funders I'd like to add researchers to the table as well so I'm speaking here with my UDAT hat on so we're trying to develop a European data infrastructure to take care of research data but research data is not really ready for publication so very various digital objects coming from instruments, sensors, high performance computing machines and so on and in this area there is no real business model, no best practice for sharing data, there is no citation mechanisms, no credit mechanisms to credit you for sharing the data so I think all is to be set up but what is clear when you talk to the researchers and research communities that we are working with is that they want that this data is reusable immediately so then they're not so much interested with archiving this data or preserving this data but they are really interested in having this data immediately available for their research purposes so that means that if you want to have the services sustainable and useful then you have to take into consideration how this data fits in the researchers scientific workflows and this I'm not sure that the funders are the right people to make this recommendation and to provide this knowledge and I guess the research communities themselves and the researchers are probably have a stake here in trying to define and provide the requirements for open access. Very briefly because I think others have said it and certainly I agree with Damien's point that the researchers themselves are key stakeholders here and finding either ways to communicate the benefits to them of using infrastructure or ways to put tools as you say that fit into their workflow rather than require them to change is really important. I think there are a couple of obvious examples from other projects elsewhere that illustrate this so one from a project called Data One in the US some of you may have heard of it's a simple tool called Data Up that's an add-on to Microsoft Excel so a spreadsheet that's very, very commonly used, spreadsheets are really common tool in many areas of research for collecting data. People who work professionally with data tend to look down on spreadsheets you know they're not proper databases they're not proper ways to manage information but the reality is that that's what they use. So the folks in Data One said okay let's not tell scientists don't use spreadsheets let's give them something that makes them produce better data using the spreadsheet so it just encourages them to supply a little bit of extra metadata to supply column headings that make more sense and instantly they don't really have to change what they do or the tools they're using but they're producing better data and more reusable data as a result and more tools like that I think are absolutely key and I chair also I think your reflection about that difference it's not I think always between generations of researchers but that's a common split that we see for some research and another tool there that I think illustrates that is FigShare so that's a simple tool that makes it straightforward to share the information behind figures it's a commercial service although it's free to use if you make your data open and that's got really high take up amongst PhD students and early career researchers it's a free way of making data open I think it's interesting to reflect there that's backed effectively by nature publishing through a company called Digital Science and they must believe that there's money to be made there somehow or other I think it's an open question yet how they're going to make that money but it's a worry I think for us as funders or as universities about how easily perhaps an entire generation of data producers are going to move towards commercial platforms like that and perhaps we're going to lose control of what is really a public good would only be fair if we take a question or two from the audience do we have any questions you would like to ask now I hope you are awake and everything then I'll come back to my list of questions and please think about yours here we're talking about sustainability perhaps once again we can address this issue how can really make those open access infrastructure sustainable right now well also in 10 years time what should be our immediate steps and perhaps some projections for the future and it's been already mentioned here that some kind of collective action is needed because if every individual infrastructure will be trying to sustain itself by itself it probably won't work perhaps if you could elaborate I don't know maybe the status of Lars and Damien and Kevin I don't know if Ulta would like to add something I think it's important that we should bring the major open access service providers to the same table with the potential funders and try to define what are the bits and pieces that are needed for a future open access infrastructure and try to sort out how should this be governed how should this be funded this will of course not happen tomorrow and of course it is quite a challenge to create such an infrastructure by committee so to speak because the existing infrastructure for the traditional publishing has not been created that way that has grown out organically and based on commercial incentives so we have to try to find another way of doing that on the one hand provide this platform for open access basically for free and on the other hand try to sustain it by developing a business model that generates income or revenue in some sense I don't have any specific answers but we have to pursue that road I think there is another issue that could be raised by funders if we include the topic of data and reliable resources of data into the evaluation model of project proposals that could be helpful I believe just because reliable resources they help to generate reliable research that is fundable in general that's part of methodology and well I think it has to be included in the legal pattern in the evaluation pattern we simply have to talk about it as an important issue but what can we call the reliable resources from my own perspective I know that usually the most sustainable open access data resources they are based either on personal initiative they are cheap but people are keen on the idea and they spend their time on it and then these mega huge resources something on the national level well it's not always sustainable we can see many ghosts on internet well something like data from previous projects that are not sustainable anymore and you see and become sad immediately so I think if we talk more about it as some kind of basis for the project to be funded it could help finally but again just to add some reflections and thoughts about infrastructure I think first I'd echo that point made by Ruta that sometimes small scale initiatives are the ones that are going to deliver us the right sort of infrastructure although I think there are certainly cases for the sorts of things you described last in my Blas where collectively we only need one of something and for Europe or for the world and it's important then to get agreement amongst all the stakeholders about how we do that one thing rather than trying to produce many of them there are other things where if we try to begin by building a huge infrastructure we're almost doomed to fail and I'm struck for instance by again thinking about two things what's happened in the UK with infrastructure in universities on which we can build a national layer has mainly been driven by competition and I say this as somebody whose politics are socialist but I recognise that sometimes competition produces the right results Edinburgh does something not because of a national initiative to do something or because it persuades itself because it's frightened that Oxford is going to do it more quickly that's what drove our university to do something and there are so many other cases like that where I mean you can see it but if you have a few leaders and you have good examples for them it encourages others to follow and infrastructure can emerge in that way by lots of small initiatives and the other thing I'm struck by is a parallel European project which some of you may have heard of called SIM for IDM which is trying to look at how to get the right infrastructure for research data management to emerge in Europe and one of its starting principles is that we have cultural and political differences across Europe that mean that it's very unlikely that one solution will work everywhere and that it's important to recognise the different starting positions the different cultural attitudes and the different political situations to have a defined way in which you get interoperable infrastructure where the means to the ends are different in different nation states and that same is true in different research disciplines or in different universities where you need to use different mechanisms to get to the same end Thank you So perhaps we should look for a more dynamic and flexible model where different components mix in the end the infrastructure I was also wondering is there a small trap with this sustainability debate and maybe we're thinking in a too long term and we are trapped by previous models where we wanted to set up something for forever and I'm not sure that it's the right paradigm but the moment with researchers and research that changes constantly and the instruments and things define what we think as a reasonable sustainability perhaps it's not 20 years but perhaps it's 3 years, 5 years and after that we see because after that the people and research practices will have changed so there's no point in investing on a very big static infrastructure for 20 years if after 5 years it becomes obsolete In ministries, governments no one really who can put money on the table for more than 5 years obviously it depends on the countries but it's harder and harder to get any commitment more than 5 years so perhaps we should have a bit more trust into the people who are building the tools, building the infrastructures and have some clear milestones and not looking always after a model that might not be viable after all A question from the audience because we'll be wrapping up our discussion any burning questions I guess we can thank our panelists and we'll be moving on with wrapping up the meeting so Mika will take the floor to wrap up and let's thank our panelists