 Thank you very much for the introduction and to the organizers for the invitation. This is very kind Allowing me to talk a little bit about our collaboration with Openair and I will show some results of that and specifically talking about How you can track reuse for fp7 funded Publications and how this can maybe also serve Is a starting point for discussion how to collaborate on these things in the future Just a few months ago. I was contacted by the open-air project And we discussed how to identify papers funded by the fp7 framework published in plus Which is for those who don't know plus is a large open access publisher focusing on medicine biology and we specifically wanted to not just identify these publications, but also specifically match papers to Grant numbers and then also and I think that's quite new and hopefully interesting to you also Show the impact of these publications not just list them and of course in case of plus everything is open access But also try to show the impact that these publications have had so far And the first step really was to identify the publications and Who has tried this as a publisher or as from a repository perspective? This is really hard Plus doesn't have a field that authors fill in when they submit a Paper which says please give you give us your grant number and your funding agency So this is all free text the only advantage we have is that because we're open access We have a open search API so you can do a search specifically financial disclosures from all plus articles and we had a very Relaxed search strategy, so we identified about two and a half thousand papers that not only had the words fp7 in there But a lot of combination with you and European committee and so forth and then The the second step was really to match this to To grants and to EU funding and that was really mainly the work by Harry at the University of Athens and without him We wouldn't have found so many publication in the end and this is As of end of July we found 1166 open access papers published by plus funded by one of six nine twenty four fp7 projects and It's of course very important to to have a list of these papers and make this list available in different ways But the next step really and this is what my job at plus is is to really show what impact these papers are having and this is a Composite figure just taking a few of the metrics. We are collecting so for example as of November 8 These 1166 papers have been accessed on the plus website and it's of course not Always you can read these papers about two million times. They have been downloaded close to 500,000 times They have been bookmarked in Mandalay, which is one of one of the more popular reference managers 14,000 times have been mentioned in Facebook have been cited close to 5,000 times and have also been mentioned on Wikipedia and this is just a small list of things we can track and The further analysis of this is of course much more work So I will just show you data of one of the many projects that we found and the engage project Which is your P network for genetic and genomic epidemiology is The the project that had the most papers Published in close. This is 30 publications and here for example, you see just the locations of first and last authors just to get an idea that Engages with a European consortium the project lead is in Helsinki. So happy that there is at least One first author in this set of papers Of course, this is just the plus subset of engage papers not the papers by other publishers In the next slide This is from these 30 papers from the engage project over time The total views on the plus website, which is both html views and PDF downloads And the circle size is the number of citations. So obviously this very much depends on how old the article is so that citations accumulate Start to accumulate after about two years And the colors are the different plus journals. So orange is plus one and green is plus biology and related journals So this is citations, but you can also look at other metrics. So this is mentally a social Bookmarking to reference manager and this looks similar, but you can also see that Of course, you don't have to wait two years and you can also see one paper here in the middle which has a fairly high number of Mandalay bookmarks and that's a methodology paper how to do a micro error experiment showing that Reuse doesn't mean citation, but in some cases it's just something that a lot of people bookmark because they might need this method and Without citing it in the final paper You can then of course go into more detail. So I picked one of the highly cited papers This one for example has been cited more than 100 times from the engage project. You can see the downloads Social network activity Mandalay site you like and other things So for example, this paper has been linked to From Wikipedia. So this is just a small article about a transcription factor But if somebody's interested in this transcription factor and wants to read more He finds further down not on the screen Further reading a link to this paper and because it's open access he can read that And this is just one of many ways how you can demonstrate reuse Just to give you a little overview of what plus is collecting and we have started doing this in 2009 and for for the user statistics that goes back to 2003 when Plus started as a publisher. So this is user statistics both from The plus website and from PubMed Central This is citations from different sources and This is social web activity and of course list this list is This is growing the last edition was Wikipedia two months ago and of course There are big differences in how often these you find things Obviously every article will be downloaded at some point. So the the lower dark blue that's user statistics Citations depend very much on article H. So this is all 60 odd thousand plus articles Published until November 8 and some of them were published just a week before or three months before so If you look at articles that are two years old you 80 or 90 percent of them are cited And it's important. We not not just track numbers, but you we look at actual citations So you can go to the plus website and see who was citing this paper It's of course a service that you find in many other places and for the social media. There's really a big Variations or Mendeley as a bookmarking tool used by scientists is the most highly used Facebook and Twitter have Have increased in use. Obviously the paper was published in 2003. There was no Twitter and I'm not sure where there was Facebook and Comments on the journal website science blog post these things unfortunately not as popular as we Would like and that's true for other publishers as well. So scientists have Difficulties or don't really go to places and comment on papers This is just a preliminary analysis by looking at papers That That are of more general interest so for example a lot of medical topics published by plus and then then also more specific Papers that are really more interested for a small scientific community You can really see different users patterns and you can use this to Demonstrate different kinds of reuse so for example scholars typically site papers. They use Mendeley they go to Pub med center, which is a discipline specific repository. They download a paper Where is the casual user that just wants to learn something about the disease or? something that's just of general interest in science may just go to the website Look at the web page use Facebook or Twitter and with these different Metrics you can demonstrate different kinds of users and It's important That you don't fall in the trap that we have fallen Long time ago and still don't get out that you try to make your job easier by just looking at journals So to say this paper was published in that journal and that's therefore it has this impact factor and therefore it's More important than this other paper And that's true for all the other metrics as well. So this is just showing the number of Scopus citations for all plus one papers published in 2009 and you see that this is a very wide variation and about 10% of them have been Sighted at least 29 times and there are other papers that have not been cited as all So to say this paper was published in a specific journal in 2009 doesn't really tell you anything about the impact You really have to look at the article level Most of these metrics are really found in one place. So whether it's citations Maybe they're collected by different services by that sort of universal Or whether it's a specific service like Wikipedia or Mendeley, but user stats, of course are collected in different places and I have placed a question mark for the institution repositories because for plus papers. We just don't know What is the percentage of People that for example download a PDF of a plus paper. What's the percentage people that go to the plus website that go to? PubMed Central as a Discipline specific repository and I should mention that of course PubMed Central Europe has the same full-text content And how much is really read in institutional repositories? You can see that the plus website is The place where most users go And it would be really interesting to understand Whether the percentage I mean both what's what's the number of views via institutional repositories and also what are the factors? That are relevant. So of course whether papers open access or not whether it's a specific subject area and many other factors But for plus articles we we don't really know It's important that all these metrics are Openly available So this is not just open access articles But the metrics are open data that everybody can use and the tools to use these metrics They're also open. So for example all the visitations we did We used with with a statistical and visualization language called are and the tools of We used are openly available so you can do your own analysis or do analysis of of similar content from other places and This is my last slide to sort of think about How you can use this information to implement as a service and of course Partners in the open-air project have harvested user stats from different repositories have aggregated this information and have really worked out Standards for collecting user stats and how to work together, etc But user stats is really just a small part of what you can Find in terms of reuse about an article and you really want to look at other things from citations to social media and the sort of easiest way to do this is of course to do this for Articles where this information is available. So this includes a number of publishers now on this list is sort of increasing every year But right now this is not really standardized so it's different data formats from every source and you will have many holds in The information so there are many publishers where you don't get this information yet So now the possibility would be to go to one of the emergent service providers Provide this information across Publishers and across disciplines But we don't want to fall in the trap to just build a commercial service model Similar to citation databases. This should be open and freely available So another strategy would be to build a service which is probably more centralized than collecting user stats from many different places and one starting point was be to use the software from plus or impact stories and other service and We both provide our software as open source that other people can use and the last slide is and of course aware that this is the sort of Start of open-air plus and there's more things than journal articles in plus is really a Publisher of journal articles, but of course there's all kinds of content especially data, but also other research outputs like software and Presentation etc. And of course you can also track all these things and People have started to do that using basically the same tools Data Has Of course a bigger challenge in that it's Persistent identifiers for data are sort of emergent and I see young browser there from data site And this example here is from dry it which uses data site dy's, but for many data there. They are just Local identifiers or many different identifiers, so it's more difficult to track reuse of data And of course reuse of data is not as common as we use of publication So for example in this example, you see that the numbers are just much lower So this is just at the beginning, but it's possible to do this and I think It's a chicken and egg situation when you can track reuse and I think it's important to think beyond data citation and also think about Usage and and other reuses that this will of course encourage more researchers to Publish their data. Thank you very much