 Good morning all. Thank you Pedro. It's my pleasure to to present the second session and inviting Jean-Claude Bruggemann that is the head of unit of data open access and foresight at the Director General of RTD and share of the open science task force. Jean-Claude Bruggemann is very well known and we've been working on framing European policies for several years. He was among other things he was very instrumental on the consultation on science 2.0 as it was the name at that time open science in 2014 too. Jean-Claude Bruggemann in the foresight. Thank you. Okay so I'll do that like the pop stars I take the microphone so it's a it's a pleasure. So it's a pleasure to be here I think it's the first presentation I did at a scientific conference outside my home country which is Belgium was in Barcelona but that was in the Stone Age so a long time ago so it's really nice to be and it was at this university but I can't remember which room it was but anyway it's a pleasure to be here so and I should have been to give you a little bit of a hint so I should have been at the RDA funders forum meeting but I told this to be more interesting so I'm sorry for the RDA people to be here because I think this is very important what the topic you are addressing and is and the deliberations that follow and it has already been hinted several times by the previous speakers are crucial for the future of open data in particular and the open science cloud in general it is my personal conviction as a policymaker which I am now and I'm not a lawyer that if we cannot sort out two non-technical issues one is well they have always technical ramifications but in essence they are not technical one is the legal and the other is the economic the business case of repositories and open science cloud if we cannot sort it out in Europe relatively quickly I don't I'm not talking about five or ten years but even sooner we will probably lose it we will lose the opportunity to create a commons for open research data in Europe it's my personal view but I think there are a lot of indicators and we can come back to that in the discussion if you want there are a lot of indicators pointing in in that direction so I was asked to give you a bit of an overview of what we are doing and what we are planning and I think that in a nutshell the issues that were presented by the by the excellent previous speakers and which are on the table now will only will only increase will only become more complex the more we push forward to the policies on open access so an unopened data so it is really only the beginning of where we stand rather than the end so just to give you a little bit of of history and I think it's it's it's quite quite nice to hear that in this area the Commission is not always criticized but always sometimes even thank for what we did and I think together with the community we have done a long we have walked a long way to promoting open access and you see the evolution there and from the next framework program onwards meaning 2020 2020 and onwards from then on it will be simply default across the ball publications data and so so it is it is a it is actually remarkable that that we were able not we Commission but we as community to to to go so fast ahead if you look at it in retrospect so that that is an important perspective to be taken and I think we should congratulate ourselves as European research community on that so where we are now the the the issues for open data are very well embedded in the regular in the regulatory provisions for the framework programs you see them listed there it is it is a legal issue so there is a framework for that that exists and the previous speakers have already alluded to that as well so that's not so important what is important to to to remember what is important to to look at is the experiences we had with the open research data pilot so I'm not I'm not talking here about publications and the open access obligations which is a different discussion we are now focusing on on the open research data part of our policies on open access and in particularly the pilot so as you as you know the pilot was introduced for a limited amount of areas research areas in in the firm in the framework program from the from the start it had the robust and opt out so provisions for robust opt out depending on IPR confidentiality privacy the the points that were raised by the previous speakers but also security issues so that is an important position that has been taken and where we learned a lot and it also explains why even though open data becomes default the opt outs will will remain there because they learned there's some significant lessons for the future and they came across they came they came much into the into the concerns of businesses and legal advocates for not going for a mandatory open access policy across the ball without having exception so that is important we mainly look at and I think it is really important in our discussions on open data that we say what are the data we are talking about the pilot and also the future so the follow-up so the default version of the pilot so if you want is mainly looking at the data underlying publications and so we are not talking about data in abstract to because in a certain sense as all data in a certain sense our personal in a certain sense all data can be used for research whatever you produce can be research you can correlate today anything to anything else so in a certain sense everything that is digital and everything becomes digital is a data source so if we start solving ex ante what is the status of all this potential new data to be developed for the use of research I personally think we will never get there so we try to focus rather limited on the data underlying so open access the data underlying the publications that were made into open access and that makes it a little bit more manageable to come up with solutions because you're talking about a much more restricted corpus of data that is being set in FP7 so the previous program before is on 2020 they were close to 200,000 publications so it is not neglectable and it is not only the data lying to the leading to the publication sorry but also everything that you need to make those data to make sense out of those data so it is quite significant in a sense but anyway so we are focusing on data underlying the publications and a DMP in the pilot and also in the future so data management plan is therefore is therefore obligatory for the projects that do not opt out of the of the of the obligation so that will also remain in the future costs are fully eligible it's an important issue to make the DMPs and whether projects opt out or not does not affect the evaluation that was a very important lesson to be learned because you could also say as a policymaker well if you don't want to opt in we or if you opt out we consider this as a negative trademark as a negative evaluation for your project so that hasn't been done and that came that was very much appreciated because the point was mentioned in particularly by industry so the industrial consortia or the consortia involving industry very much appreciated this possibility so the mantra as said by the first speaker is indeed as open as possible as closed as needed now here here are a bit of the of the reasons why from the projects deciding to opt to take part into the pilot so the so the more or less the rate of opting out was 30 35 percent and from those areas not taking part in the pilot on a voluntary basis 14 percent opted in so you can say that on average the the take up of the pilot was quite positive even more closer so you can say that the the take up of the of the of the pilot was was quite positive all in all so a significant amount staying there 65 percent depending on the areas but anyway that's the average and a relatively interesting amount opting in now if you if you imagine this on a larger scale and you go for the default version from September onwards we have 30 billion of research money remaining in the framework program and if we have the same amount in the next framework program which would be 80 billion it means that we are producing for around 70 billion worth of open data in a certain sense because that will be the totality or the percentage of of the of the of the the framework program that is taking part in the open data so not opting out so that's that's that's a significant volume of of of data and I would say it's a quite positive positive effect it's not 100 percent but nothing is perfect certainly not the world so so I think it is very encouraging what you can see there is also what are the main reasons for opting out so the the colored bars give you the main reasons why from the 35 percent that did not want to stay into the pilot why they opted out and then you see that in the in the main reason for opting out is IPR so this is of course concerns in particular from the from the from the industry to a lesser extent again 20 percent of the 35 percent opting out is related to privacy and is related to all the issues that were so eloquently discussed before that an almost equal percentage is there as a category saying that well our project does not generate data so we don't have to take part in in the data pilot which which by the way my personal view is a very strange way of of thinking science without data is a little bit even even the data is texted text is also data so but anyway the categories foreseen and and and so on so it was listed so you go there for to have an overview of all the different reasons so in a certain sense the the issue of of what has been discussed this morning and what you will discuss the rest of the day it is a relatively limited amount of projects on the total volume of so many billions of research money that decide to opt out for reasons related to personal data that's what you see there so you have to put the things into proportion of a very an overwhelming part of research projects do see the benefit of open data and as we said as open as possible as closed as needed if you give the provisions to opting out it seems to me that's that the industry and and and the privacy advocates and the health people and so on are very relaxed about about the way that we do the policy so I think this is a quite important lesson that we learned from from our pilot now as I said I mean our view is is so we no longer only talking about openness about access to data but also about how make these these data reusable findable accessible and interoperable and the famous fair acronym so the next step in the policy is not not only open but also make it fair and that is of course a very important part of what we want to do in in the future DMPs and in the policies that we will start to implement after the summer when the pilot becomes default and where we will go into the full default in the framework program in the in the in the in the next framework program so fair as you know is much more than openers fair is really openness is only let's say addressing the a of fair the accessibility fair is about findability interoperability reusability because that is what you need all those four together and that made operational is what you need to have reproducible science and that's actually what open science is all about my openness as such is quite meaningful if you can't reuse it so that is that is where what where we are now working on how to translate fair principles into the DMPs into the templates that we will provide for the data management plans in the future and that will as for those who have been working on it or with it with us that will create some some significant change in what is required not in terms of I would say bureaucracy but in terms of granular granularity of what you will provide in your DMPs and it will also require by the way that we train our project officers to to make them able to judge what what will be provided there so the the fair the the the fair data management plans that that will become standard in in the next frame in in the last part of the favorite program and in the next program so the things that you will get there well you see them so that you will provide a template which is a service not an obligation but but that can always change of course but for the moment that's that's that's the policy we try to make the standard DMPs as light and flexible as possible by by by using by guiding the submitter via a set of questions to the kind of answers that that he had that he has to give it should be a DMP per project and not per data set because that is all in that is really I think the most simplified solution possible and I think much more than enough in terms of fair principles and it should be of course the living document it is something that can change over time you can opt in and you can opt out across the the ball of a project so there are examples of projects that's opted out and during the course of the project after two years decided to opt in into the into the data pilot because they thought well what we produce is not sensitive or not IPR related so let's do it the other way around happened as well so that is also something we have to make sure that we can foresee in the future well you get the guidelines there so that's the portal that's not going to this details now we as I said we as a pilot we we also collected some experiences of the DMPs that were provided in of the of the projects part of of the pilot and the key lessons learned you see them there and there are a few good papers on that consultable on on our website so so overall we we we we realize that additional guidance on on data management is needed for all groups of actors in in the research projects that is something to to take very serious I to be honest I thought the level of awareness of data management to be a bit higher three four years ago when we started working on this and I think we we have realized that outside the community of data people and there are 1.7 million researchers in Europe of which a few thousand of data people so there are much more people not working on what we work so there is really a little bit of agnosticism to put it elegantly on on DMPs and data management and what it should be so there is a lot of work to be done there in terms of raising awareness guidance training follow-up and so on the second big lesson is that aspects such as data preservation RPR standards and so on are all too often not very well documented in the in the in the answers of the of the DMPs that we have have received so far so it is a little bit it is a little bit superficial way of of arguing why why for a reason XYZ a certain project cannot comply with with the with the default of with the with the pilot and we we would like to go a little bit more into details to understand first of all better why but secondly also to make at least the the the standard the standard answer well it's it's it's not it's not it's not it is not manageable from an IPR point of view so we don't take part in the pilot to to avoid a little bit that kind of great we answers nevertheless we there was also a significant amount of significant proportion sorry of the DMPs that we could look at that were quite excellent if not if not to say more than excellent and for which we we really can take a lot of best cases as examples for for future projects to be to be submitted so overall the the the experience with the DMPs is let's say relatively positive but it could be better to to answer like the weather and it's always good but could be better but in in in this case I think we have to do much more effort there and we are working really with a lot with our colleagues to to try to to remedy the the lessons that we learned there so fair data if we now look to the future and the fair data issue it will not stop with with the the provisions of open access and and open data in the in the fair in the fair work program if you look at it on the long term but it starts today it is the DNA of the European Open Science Cloud that's that's the acronym at the European Open Science Cloud which is an idea and a project launched by our commissioner 18 months ago which wants to offer a seamless European trusted third party for doing data research across boundaries physical and across disciplines that's what the science cloud wants to offer if you picture it it wants to offer let's say a commons as an environment for research and data for data-driven research in in Europe that means that means that all the discussions and all the issues that were listed before will become articulated to the level X if you try to realize this trusted third party across boundaries and across nations in Europe and that's why we say it is the DNA that if those data in the science cloud are not findable accessible interoperable and reproducible well done you don't have a European science cloud you might have many different clouds and many of them probably private but you will not have a European space for it so that's why it is so important for future discussions so that's our commissioner where he where you have his view so which he made public in the presidency event of in Amsterdam last year that's the the famous white paper where the cloud in the communication sorry where the cloud initiative and our part of science cloud was was announced but I think maybe most of you have are familiar with this document but that politically speaking April 2016 was the official launch of the idea so we are almost one year later and very close to making the cloud a reality by developing several policies which I will not not in detail tell you now but if need be in the discussion so the science cloud as such European open science cloud has three actually four levels of policies to be to be addressed at one is the easiest one so to speak is the technical one the bottom one then the more you go up the more it becomes when the more you go into the upper layers the more you have a mix of technical and non-technical issues to be addressed they are listed there but if you if you translate that to fair you can very easily see that in the block of open data content the fair policies are extremely important so that how to translate that again in in European open science cloud environment is is what is on the agenda this year and next year and in detail of course for the infrastructures to wind up so what is next so well the on open access and open data the decisions have been taken it is mandatory from September onwards so what is next is to translate these these issues and first of all to to fine-tune them the policies and the lessons learned but then translate them into the environment of the open science cloud and that is where you see a few of the initiative listed listed there and if you want I can come back to that in details later on but it you can see from that that it is very high on our political agenda our policy agenda so what is our what is our to wind up now I'm really winding up so a few slides to to conclude so what is our common challenge as research community policymakers scientists data database builders infrastructure people I really think that we must make sure that not only we go for better science but also for more productive more reproducible science that we can in ten years time come up with slides that you can no longer see the difference anymore how many of these kind of initiatives are there it has been proven that open data as an enormous value not only for the quality of science and the speed of science and as a consequence also the speed of innovation but it is also a very economical sensible thing to do so we must really keep that in our mind and try to multiply these arguments we I also think that we have and and that's where I'm so interested to listen to your discussions I think we have to try to come up with definitions of data which are dynamic and which fit open science a static definition of what our data and who owns them is probably but I'm not a lawyer is probably something you can never achieve because the reality is continuously changing what was open science 10 years ago it's not open science 15 years ago and we are increasingly realizing that if you want to reproduce science you not only need to open data you need open methods you need to open algorithms you need you need to actually open software how that will fit into a data approach which is fair quite frankly I don't think that's such an easy thing to answer and I'm really looking forward to to the kind of answers that you that you can help us giving there because I open there is is a community of people that works on this day in day night so to speak at least day in so that that is a very important challenge I would say if to make this dynamic data approach possible this is a joke and we should not give in on quality so that's for the good understander let you read this and then so finally let's let's move fair from a concept to an operational reality for doing data-driven science at the European level to make open science a reality and as I said in the beginning let's do that as soon as possible because I don't have to convince this kind of audience that there are a lot of private players on the market so if you are really concerned about the European Commons then we must make sure that the rules of the European Commons for research data are determined by the community and not by the providers from what kind they are so working on these rules which is the legal the legal work work is is something that is very important as something we are as commissioned very much looking into into your direction to to come up with brilliant ideas thank you for your attention thank you Jean Claude very interesting presentation I think we have time for one question before going to the panel discussion thanks a lot Jean Claude I just can't resist answer I ask you a question so I if you would have been at the funder's forum you would have been discussing this on a global level and you decided to come to the European level and my question to you is that do you see Europe at such a strong position really that we in Europe only can push the fair principles in on a global scale or do they come in you had the links to some global efforts or some things but you were mostly speaking about the European case so how do you see the global view in the fair field I do like me Jagger again so now it's it's I think for the moment we have a strong position and I also think for the moment that we are Europe is being followed by in terms of what we are doing is on the attention span of the policymakers globally I actually as some of you know why we there is also a G7 initiative at the group of seven rich so-called richest people richest persons of countries sorry and they're not always the same so so they're working on this we there will be an in a ministerial conference in I think in October now end of September in Rome and it deals with a global culture of what there are two topics on the agenda on on the ministers for science and technology one is a global culture for research data where a lot of the fair things are are being proposed by us and where most of our counterparts in in the States and so on Japan think exactly the same and the second issue is actually the incentives but that's another issue so I think for the moment we we do have a strong position there and we are still 31 32% of the scientific production in the world so if you move as a block on that position there is a highly likelihood that that others will follow how it is called is it varies across the globe but I think in the science world at least I think that that most of what we are defending in Europe is is is shared policy wise and if you then translate it to the political discussions that's of course not a discussion this being said I think we have to realize that other continents are moving fast also so this is a little bit it always reminds me to to the situation in in in mobile telecommunications or before that but the beginning of the Internet you suddenly have several blocks and at a certain moment one standard becomes operation so we have to watch that moment and make sure it is our moment