 So you should be right to go Rebecca. Okay, thanks Jerry. Excellent, we can hear you. Thank you. Thanks very much Hans-Yoget. It's kind of hard to follow your presentation about such very clear practical training in data management with what I'm about to talk about. We are at a point at Swinburne and we were when we did this particular session where we had a research repository and the ability to describe and promote data sets and send them to Research Data Australia, but we didn't have data storage and we only had a draft data management policy. It's actually very difficult to enforce a data management policy without data storage and it's hard to encourage good data management practices without a policy. So part of the purpose of this session was to discuss with researchers what they would like the policy to look like, what would be practical for them. So the three presenters were me. I'm the research services librarian at Swinburne so I look after all of the services that we provide to researchers from the research repository for publications through your data, copyright advice, all those kinds of things. We also had Terence Bennett who is technically employed by the College of New Jersey in the US but he was on succumbent to us as a research data librarian. And we also had Sonia Petal, a research fellow from the Swinburne Faculty of ICT and the Dean was present as well. The first thing I'd like to point out is that the reason I've been asked to speak to you today is the difference in our presentation that we had a researcher talking with us as well. But Sonia is a researcher with a research interest in data management and policy modelling. So I wouldn't say that the kind of presentation that we did would work with every researcher or any researcher. Sonia's research is also really multidisciplinary so she was able to give examples right through from anthropology to computer science to quite hardcore data and that meant that everyone there was covered no matter which discipline they came from. The reason I want to talk about this style of presentation is that I think researchers opened up far more because we had a researcher stand up and honestly say this is really hard to do and far from perfect. From experience when I've encouraged researchers to openly evaluate library research services like the repository they've been really reticent to criticise anything in front of me in case they hurt my feelings and they probably would if they said anything really mean. But having Sonia facilitating the discussion definitely made a difference to getting people to open up and making it more interactive. This is just a quick discussion about what an agent-based modeler does just in case not everyone knows. What they like to do is they like to model systems and how people interact with them and how policies grow around systems and how these three things interact. The significance of the way that we ran our training workshop slash sessions slash free for all at Swinburne is that it was a workshop not training as we know it. We probably asked more questions than we answered because we didn't have the storage and the policy and everything to point to and say this is how you've got to go about it. We just wanted people to start thinking about how they would look after their research data. We were enabling discussion but probably not really dominating it. We also went with the approach of facilitating rather than leading. So this was a user-centred discussion about benefits. The purpose was, as I mentioned before, to receive input about how we might design a data policy that suited researchers. I think it's unusual perhaps for researchers to contribute to how we think about data policy. I think a lot of that sort of thing is worked out by support areas like the library and IT and research and maybe even the intellectual property office. But it was a great opportunity to say if we designed a data management policy that you needed to comply to, how would you like it to look? So we asked researchers to think a lot about the big questions. We're still right back at that point and to tell us what was getting in their way. It was a two-hour workshop with cupcakes, which was probably the most popular part. I've included a few slides in this presentation that came from our training session so you can see the kinds of things that we discussed with researchers. So we discussed the value of data management in terms of sharing. So sharing with yourself in the future as Anne likes to put it or sharing with another. We also talked about how we can promote research through the repository because that was something we were actively, practically able to do at the time. We asked this question of researchers and a lot of them looked a little bit guilty. When did you last see a computer with a floppy disk drive? Well, I haven't got one and I was willing to stand up and say, and I've got stuff that's sitting on a floppy disk drive that I'll probably never be able to retrieve. And I think that relaxed people a bit. The value of having a researcher in the room was that we could actually give some real-world concrete examples of what data looks like and how we might need to manage it and what the differences between kinds of data are for access and ownership and even the ability to share. So this was one of Sonya's slides about the kinds of data she had collected through one of her projects. It was everything from quite rudimentary things like the BARC reports that have to go in with the project and keeping an eye on the budget and that sort of thing, right down to analysed data, photographs which could never be shared because they were children, server log files, which nobody feels particularly sad about if they are public, but just the breadth of the kinds of different data really opened people's eyes a little bit to what it was that we were thinking about when we talked about research data. We didn't specify who the audience should be. We talked when we promoted the event about getting people along who were interested in data management policy and how we might go about developing one. We did get lots of HDR students, lots of PhD candidates and lots of early career researchers. The project officer in the research office who at the time was drafting the data management policy also came along which was really good because she could be there and hear in person what people's pain points were. But because of the audience that we attracted we decided to choose to focus on promoting data as early in your career we went with the idea that all promotion is good promotion and the ability to promote your research through sharing your research data has to be beneficial and that seemed to work quite well with people. We asked them if they were sharing their data and how they were sharing their data and what we could do to make it easier for them to share their data and we did find out that there are quite a lot of wounds quite close to the surface which was a little bit of a shock for us. We knew it wasn't easy but we didn't know quite as much detail as we found out from this workshop. We didn't for example know about some of the competing interests at work so we realised that there's a really vast literature on data sharing and the impediments to that particularly in certain disciplines. But we hadn't thought so much about what the relationship between sharing data and accessing data is. So when we sat down and said why would you share your data, how would you share your data some of the questions we got were I can't access the data I want, why would I share my data and they gave quite concrete examples of data sets that were held at other universities and actively not shared with them either by email or through a repository. And that was a discussion that was quite hard to move past. They weren't making a distinction between accessing data including paid data and sharing their own it was all part of the continuum to them which was interesting because it's not really the way that our experience with making research data sets available through ANS had told us. Data was also seen as a commodity by many especially the engineering researchers who were present so they were interested in commercial data that they are couldn't get access to as mentioned before or was worth tens of thousands of dollars and they wanted to know why they would share their data freely if it was clearly of some monetary value to them. We were also perhaps not quite as aware as we are now of the issue of commercial and confidence data. So we realised because I come from a publication's repository background I've got that perspective already but we realised that one of the reasons that we sometimes have to embargo these is that there's confidential material in there that can't be shared because it could identify someone. But it never occurred to us that there's research that some of our academics are working on right now with industry that they can't even let their competitors know they're doing let alone sharing the data that's an outcome of that and that was an eye-opening experience for us that idea that they don't even want someone to know that it's happening. We made a point of emphasising the benefits of sharing data either with oneself in the future or with others and the idea that that ties into how you promote the research that's coming out of your work. Some of the PhD candidates were interested in the fact that we might want their data along with their theses. But we also found it helped to be realistic about the obstacles. There's no point pretending that it's easy to share data especially in a situation that I've described before where we had no policy or storage to support our discussion. But also that researchers really appreciated as demonstrating that we understood where they were coming from and that their concerns about sharing their data are legitimate. And we were able to change some perspectives just because we were prepared to talk about the negatives. So it could in the future be a great outcome for us as data trainers or people encouraging people to share data if we are able to address these issues in a public forum and ameliorate them. But we can only do it if we actually do say, well, you know, there are some reasons why you might want to share your data or you might want to restrict your data or those kinds of things. So we put up this slide in the presentation so that we could show that we were aware of some of the problems that, you know, there were time restrictions and the terrible issue of missing incentives as far as data. So it's very clear return on publications. You get points, but nobody rewards you for sharing your data. So we put this slide up merely to show that we understood. We found that tools were really helpful. It actually, in our situation where we don't have a lot of practical tools to help researchers, that having even one or two was really useful. So we have a very active supercomputing function here at Swinvern, which is available to the astrophysicists who run it, but also everyone else at the university. And that's quite long-standing and that's a really useful service. But there's not a lot of thinking at the university other than what's happening in the library about how to manage the data. Our data management checklist, we went through very briefly in the session and talked about how if you complete this checklist then you should be able to have the rudimentary elements required for a data management plan. Without a policy it's very difficult for us to enforce people making those plans, but it does at least allow them to think about the things that they need to think about if they want to look after their data in the future. And our checklist is influenced by Monash and QUT and available from our website. We found that a good example is gold. And especially a local example if possible. And luckily we had one and it just kind of came up in conversation, something that someone said made me think about it. We have this software evolution data set which is prepared by one of our researchers and he makes it available through a website for the project with clear details on how to cite the work and also under a Creative Commons license. So the reuse qualifications are very clear. We've cut a lot of this data set and it's now available through Research Data Australia. But it was nice to be able to show that someone at Swinburne had been able to do this and it wasn't so very hard. And that there was also something to be gained from it. The dean of ICT who was present at the time talked about how in particular in the computer science sphere you can tie the open data movement into other open movements like music and computer science where it was shown over many years that exposure was sometimes of more value than royalties. And that was a particularly good concluding point for those who'd been so concerned about the monetary value of their data. We might have been there to talk about the value of good data management and planning but we still had to be willing with the kind of session that we'd set up to change the tack of the discussion if it was going in a particular direction. So we had to adapt our presentation on the fly to take account of the audience being mostly early career researchers. So they weren't awarded research grants with the ARC and therefore it was difficult for us to say that they were as subject to the responsibility of good data management as say their professors. We would have had to think a bit differently if the senior professors had been present. So the draft data policy was prepared incorporating feedback from this session. It hasn't actually been ratified yet but we made a start. Terence has since returned to the US and Sonya's been teaching but she's returning to research next year. Swinburne has a metadata stores project at the moment and once we have working software installed it would be good to use this as a catalyst for having a follow up session with some more practical examples more like ANU is doing. At the moment the library research data page is the only one in the university that provides advice on where to go for data analysis support, software training in SPSS and R and other statistical software packages and also a list of data repositories in particular subject disciplines that might be useful and the checklist is available there as well to assist with planning. And my advice to you if you're starting out in data management training is talk to someone else who's tried it before you. They're my contact details. Are there any questions?