 When we think about data management, we're also thinking a little bit about why we're talking about it. What is it about the world that has changed that has been that this has now come to the top of the charts as something we needed to do? And I just thought it might be nice just to sort of step into this a bit because it really goes very much towards what it is we're trying to do with people, how we're trying to support them. So one of the things I did when I first started working in this area was really to look at a whole range of materials that were out there and this fourth paradigm was one of the ones that I thought was really interesting that we started off with Archimedes in his bath noticing the water is coming out and that was the beginnings of science and then people started to theorise about science. We developed computational science where people were able to do simulations and so on but we're into this giant area now of data exploration where the data is so enormous that it cannot be processed by a human that computers are needed to do the analysis and to analyse the work and manage the data for people. And in that kind of world we really, really need good management of data for people even to be able to do the job of science. So that to me was a real breakthrough moment as well. I hope I'm on the right slide now trying to get onto this one. Yes, I found this really interesting talk online something you can probably still listen to and this guy Jeffrey Bolton was saying that when he was a young scientist starting out he might do five or six experiments and he'd be able to write that up in a paper but it's got to the point now where the results coming back from scientific experiments are not just six or ten, they're thousands of experiments coming back and it's impossible to shoehorn all of the work that comes out of that research into a published paper. So to some extent the published paper which is of course the Holy Grail for scientists and researchers is however really just now more of an advertisement and the science is actually underneath and if people don't have access to the data they cannot understand the science. So just as we've had a great push in the last few years about open access around complications I think we're going to see a greater push and it's starting to get underway now for open data because if people don't have access to the data they're not going to be able to see the research, understand the research, use the research and we've seen the growth of data only journals. These are the game changes as far as I see and I think this really impacts on our role in trying to support people with research data management or in e-research. Simply that we've got these enormous projects now where we've got huge amounts of data coming in, we're answering bigger questions because we've got the ability to do that. There's a huge number of data sets out there already and it's easy to share those with people because we've got tools to enable us to collaborate. We've got new tools so we can visualise data, we can actually mine data, we can get into that 5 million book corpus that Google Books has produced and we can find out about the use of words and so on. So there's these new tools out there. We've seen the growth of crowdsourcing where people like getting involved in projects like the Galaxy Zoo, identifying the kinds of galaxies that are out there. We've seen crowdsourcing to help us correct the Australian newspapers in digital form. So there's all of these new things coming on board that people are interested in and finding out about and that really impact what we do. The expectation that data will be open and also the code of conduct. And obviously all of us need to understand the code of conduct because it's very much about data has to be managed, it's the responsibility of the researcher to manage data, but it's also the responsibility of the researcher to get that data out there to disseminate the stuff that they found out in their research. But to do that it's very important that they have the skills to do that. It's one thing to beat people over the head with a stick and say, you must do this, you must do that. But equally it has to be easy for them to comply. So the government put quite a lot of work into research skills for an innovative future to try to decode and define the kind of skills people are going to need if they're going to be able to do this work. Three of the things that I found really helpful in trying to help me understand how the world has changed in particular disciplines were these three reports coming out of the research information network in the UK and while they were looking at case studies of researchers in those, in those, in the UK and not in Australia, it's very useful if you're trying to get a handle on how things have changed here just to read through those case studies. And I think a lot of what is relevant for the UK is also relevant to us. In the humanities for example there's very little take up of tools that might be out there to do digital humanities. It's not that people aren't interested but they're rather frightened that if they do start going down this road that there won't be anyone to help them, that they won't have anyone to ask. And so even though they'd like to get into this area there's a bit of a gap there. In the sciences it's a bit of a different story. They're probably not as frightened of tools but they are unwilling to take them up if they are not going to be supported. Certainly there's a whole range of new roles coming in the life sciences. These roles like statisticians, people who can model data, who can curate data, bioinformatics and so on. And they're very interested in people supporting them with that work as long as you're close to where they are. So they don't want the librarian sitting in the library doing this. They'd actually like that person physically based in their team. In the physical sciences a bit of different story where we expect these people to be very good with tools. But apparently there's a very poor understanding of how to search for information in this area. So that's a place where there could be a role for us. And they certainly need help with metadata. They're creating a lot of data sets and they need people who might be able to help them work with that. Another thing that was really helpful to me in understanding the changing world was this big exhibition that came from the British Library. They've taken it down now. It's no longer available. But it's available on the Internet Archive if you can find it through there. And you can watch some of these interviews with people from different disciplines. But one of the things that they talked about a lot that came up as a thread all the time was they saw that the whole business of supporting e-research really was about teamwork. It was about multi-disciplinary teams that you might have a software engineer who can design something. But you need someone who can build a database. You might need an artist to help you visualize how you're going to present the data or to build a kind of virtual reality world if that's the way you're going to go. Or you might need someone to crunch numbers. But there's a whole range of skills that are needed. And one of the things I thought about that is we don't have to do everything. But there certainly are roles that we could fulfill. And we definitely have a place. This is my sort of academic in the pinball slide where I'm trying to explain to people, yes, it can be not as easy for an academic or researcher to find the help that they need. And certainly when I started out in trying to build support services in data management, I felt that there were quite a number of people in this position where they've come on board, they want to build a database, or they're trying to share the data they've got, or they want to do a longitudinal study, but they need help with designing the way they get data or about surveys and so on. And where would that person go for help? And obviously people turn to their colleagues or they might turn to local IT. But is that necessarily the best place to go for advice? Maybe those people aren't up to date. If they go to ITS, they're a little bit focused on kind of workstation support or keeping the network going. They're not necessarily focused on research computing, which is what that person really needs. We do have a new research unit here, the Research Computing Centre, but at the moment they're a little bit more focused on high performance computing and those very high volume users, not so much on the long tail, or the people are a little bit timid about what they want to do, not quite sure, haven't really got their act together. QCIP, obviously, whom I work for, can help, but have to be invited to come on board. And basically I just felt that people were kind of in this pinball machine where they'd get sent from pillar to post, but didn't really get any kind of useful joined up advice, and it meant that sometimes they were a little bit frustrated. I came on board with the library, possibly about two years ago now, to try to build a service to support data management. The idea was to, first of all, that I came, and it was just me, there were no other people, but the idea was to work out what the library wanted to do. So very much about scoping a service, working out if there was a role for our UQE space repository to play in data management or not, but generally to kind of have a bit of a look around and figure out what was needed and what we might be able to do. A couple of the ways that I started to sort of dig for information was we first of all started to do a survey with RHD students, they were coming along to a lot of our events, and it was quick to get them to just run through a little survey about how they were managing the data they had. It enabled us to then start having that conversation with people about, well, you know, you are supposed to manage data and there is the code and have you heard about the code and so on. There wasn't anything we could use to say that you must do this, we were just trying to raise awareness. And we did run another survey, but I might talk about that a bit later on. We had no real support materials for data management. We had nothing really on our website to say that we were interested in this work. So part of what I started to do first and foremost was to build that material. We had a new unit in the library called Scholarly Publishing and Digitalization Service and we had, under that umbrella, was Data Management and also Scholarly Publishing and also the Bibliometrics, being able to help people with metrics for grants and so on. So we decided our research support pages should encompass all of those three things and one of the things I built onto the page was a template for a data management plan and a checklist for a plan as well, so that they could work through a checklist before they started filling in the plan template. The other tool I developed was a data assets inventory, I think I copied this from ones I found online, but so that people in schools who were starting to worry about the data they had, wanting to get sort of a picture of the data they had, could have a tool to help them record that. And the main way we started to disseminate information was through fact sheets. So we developed 20 for data management. I think there's nine for metrics now and 15 on Scholarly Publishing. This is what the website looks like, the data management part of it anyway. It tells them about the code and links through to there and it just gives them the downloads. If they want the checklist, they want the plan template, they want the data assets inventory. We also have the data management questionnaire that we ran online because we're always happy to get more feedback and usually that was a way of us finding people who said yes I would like more information, I'd like some help. Then we've got their details captured and then we would follow up with those people. So we've left that online and people do fill that out and lead us to give us information that way or ask the library. The other thing I did was set up an eSchoolership blog which was really to discuss all of those issues as well. When I was training staff within the library we had a whole lot of what needs to be included in the plan and one of the things when I was training staff to do the job of going out and talking to people about data management was that they needed to be able to talk through all of these particular issues. But in this single one of those issues was covered in a fact sheet so that if they needed to talk to people about data formats or version control or how to restore data after you backed it up then there was a fact sheet to do with that. The way that I trained people was trying to be in a very non-threatening way. I felt that it was very important for people to move along a particular continuum so that they didn't feel like they had to be an expert within five minutes that some people are much more ready to take on work than others and I felt it was important that people move along at their particular pace. So I developed this thing I call grasp which was about that it is a continuum and people will go at a different pace but within a year or so we'd like to have moved from the gauge where we're sort of scoping what is needed to a position of being able to advise and support people and hopefully get to the gold standard of what I really felt we were aiming for as a partner. So all of those kind of roles that came out of understanding what it means to support a research we would be partnering in that. So at the early stages at the referral page stage it would be more a raising awareness, running surveys maybe putting conducting data interviews and getting stuff on to research data Australia and just starting to develop those basic materials to enable us to offer checklists and templates for everyone. At the advise and support area trying to help people with the grant about scoping their data maybe helping them write a plan if they're going for a funding application they need to have a plan help them with that maybe give a bit of advice about data that might be in an obsolete format or maybe help with specific tools and services as well. At the gold standard partner point I would really hope that we might be helping with methodologies for maybe a crowdsourced project or digitizing material for people maybe a non-line transcription project of non-digital material partnering on bids and so on and maybe embedding stuff because I felt that there was a role for that as well. Probably where we're at now is we have these specialists at the top of the tree people who are the team leaders in those three areas the subject experts who people who come on to comment and then basic most librarians just have a working knowledge but not a great knowledge. Just in terms of logistics of how all of this rolled out we offered part-time to librarians who were interested in the area of data management and building this service they came on internal succumbents and then they helped me develop the work that I was doing and help in training staff to do the outreach. We also succumbed one staff member to CUSIF as an research analyst believing that that person can bring those skills back to the library. I had a role on the UQ's research data management policy group that group met and we did quite a lot of work in trying to nut out what the policy would be. We contributed to ceding the commons and we also in the course of doing that data management policy we surveyed 200 plus staff about their data management habits and we were able to present some hard evidence that we really had a bit of an issue which was really useful. We developed training that we offered to the grad school during research week and things like that but we also got training into the UQ staff development program because we thought that would kind of legitimize what we were doing so we developed a data management and a data sharing. I developed some branch classes with generic content that people could run so that a class on collaboration tools, a class on what is data management just so people could start to roll that material out and get it more widely known and we also got the research office to agree that grant recipients would be told that they should have a data management plan. I thought that was quite important but the other thing really much is what I call building the web and really this is my last slide which is that if you're going to stop people from feeling like the pinball you've got to understand all of the people who touch research and might be involved some peripherally or crucially in data management and make an alliance of all those people and get all those people to see that we all need to work together to tell a single story to the researchers who are trying to do this work so the library with an established service culture was seen as a place where people can come in the front door ask a question. We can help with data management plans, we can help with data descriptions and metadata but on the whole we're probably going to be referring people onto these other units for things like ethical advice or storage or help with patents and so on but if we didn't build that network then we would really be leaving people to the mercy of the pinball machine and I thought that that was a bad way to go. Okay that's all I really got to say as I say here I'm happy to chat if I've got any material that you might like to repurpose I'm happy to do that and if you're in Queensland I'm happy to help you get started and that's the end of me.