 Okay, um, I'm just about people as well. Um, people doing infrastructure. Basically, we had done an architecture project, or part of the university had, and sitting in the Commons. So we had a little more preparation. What we did realise is that really critical to this project working was that it needed to be embedded in institutional infrastructure. And we had already tried to go and talk nicely to the research office, and yes, that was pretty good. So we decided that up front we really had to establish some project governance, which gave ownership of the project to the university, not to the library. And so the project actually lives with the chair of the research coordination group and his, um, faculty, faculty of engineering. And people on the governance board, you can see on the list there, from the analytic centre, where scientists put data into instruments, which it fits down into some of our biggest data effects projects. And then grant management, research policy, research strategy. More. More to interrupt for a moment. Um, is there a microphone you can get closer to? Okay, we'll just put it a little bit higher. Is that better? That's a bit better. Grants management, office research, several areas of IT. Um, IT, I guess, I'm still going to do that, but IT is constantly going through change. So we needed to be insured, to ensure that what we were doing actually was very alive with their shifts as well. And a member of the group also is the university librarian and the director of the library. So that's at the side of this, people. The people actually needed from the top down, as well as the bottom up to see in a moment, needed to actually govern the project and drive the infrastructure. So this is a simplified version of the infrastructure. Um, you can see the elements in it, the litters, that are highlighted in pink, green and grey. These exist or will exist at the end of June, when our project is completed. Most of them are in testing at the moment. So I'll go through them quickly, but what I'm going to do is demonstrate them through research use cases, rather than through the infrastructure. Coming back to people again, it's the researchers, and this is the bottom up approach to it. It's actually the researchers who are using the infrastructure. So we visit for them not to make something pretty and clever. I think it's quite pretty personally, but hey. Specifically based data systems on the bottom left, these were from the data capture projects. And further projects have been developed as well in other areas of faculties at the university. These can go directly into Fedora Mint, which is our repository. Two along at the bottom is the data metadata deposit. It's simply the web-based deposit form to put RISDS records in. We graph metadata feeds and the management layer from the UNSW Data Warehouse. So grants and people information comes once a day for people. We actually get the same feed as the research publication system gets, and grants and information once a week. As you can see on the right, there's lots of different ways of discovering this metadata. It's not just going to RDA, and that's why it's really important that it's actually designed to serve lots of purposes for researchers and indeed for the university administration. So the parts that are not highlighted are either not part of our project or not completed or both. That's future plans. So at the top, the storage layer IT are working on this hopefully as we speak. And along the bottom research data management plan deposit, which you'll see in my research use cases, the data deposit is part of the storage component and a reporting service, and I'm not going to explain that as we go along as well. So getting back to people, and I promise this is the last of these ones, this is who is responsible for the various areas within it. So while the UTAIS presentation was talking a lot about external partnerships and working with other areas, we really have focused on embedding ourselves as much as possible within UNSW systems and infrastructure. So as you can see, we've got the IT area, there's going to be financial operations, there's lots of the warehouse, lots of faculties, Division of Research at the bottom right, there's a research profile, there's a research gateway. The next two should be part of the library, but there's sort of a blue line that didn't go away. The institutional repository, which is UNSW, we link to the publication records. And the data discovery is going to be a user interface based on Primo and Kalibera software, which will make the Fedora repository searchable and external responsibility there as well. Okay, so while we're on people, I'm not going to show you pictures of anybody, but this is the team who has done a great job of it all, so I can take very little front of front evidence. Okay, so now, oops, sorry about that, I'm going to go to some research use cases. Okay, so I'm just going to walk through these. I mentioned before that this has been built for the people. It's been built for the people in the administration of areas who basically have provided infrastructure that sits around it for the researchers at the bottom end of it and for the team that I just displayed who actually created it. So I'll just focus through the system by showing you how to use. This is a very common scenario. You're probably already by now. The researcher wanting to do something with their data, so help. There's a couple of ways to get the data into Reddata. Mededata Deposit is the web-based form. It goes through the Deposit Service API. Or alternatively, it can come straight from other systems, so the analytics centre has had a project, ACData, which was a data capture project. That metadata comes straight through to the Deposit Service API and is then integrated in Reddata with several records. Okay, a record is created in Reddata. We're assigned as the system identifier. At this stage, it's a handle. It's linked to associated parking and activity details from the UNISW data warehouse on the left side. The record is able to be extended or edited by the researcher. So the data that's mededata deposit is also an editing form. So people can go back once they see the data and send it. Yeah, it's not on display. But, yeah, they can add to the record or change things. This is the dedication. Okay, so the record can be displayed through the UNISW Discovery Interface. This is the preview when I mentioned. And research started in Australia. And links to researchers' research gateway profiles. So that would be just a link, not an actual display of the record. It's kind of one of the things that underpins the design. Well, we've got things to be displayed in several cases. We want to minimise the amount of data which we certainly budgeted locations and efforts overall. Okay, so for the research can be uploaded and attached to a mededata record at the same research data, sorry, as the best data record was created or at a later time. Additional data can be added and dashed out of place. So from the left is, we'll go directly through to, or should not directly, sorry, mediated by red data, the mededata, to research responsibility of IT. So they'll basically take within the mededata record of the mededata at the data storage service point where it will then communicate with the storage. So information in the mededata record will assist the data storage service in determining where the data can be located, either locally or externally. They're actually building a local data storage, external bulletproof things like RDSI. So there may be something in the mededata that says, okay, this looks like an important data set. Have you thought about RDSI? Do you have a candidate for that? Or any other number of data storage facilities recorded and captured to inform the data storage system? Sort of straightforward, not that straightforward, but that's a fairly common scenario about a depth mededata record that goes to research data Australia, but also data storage, all through existing enterprise systems data like Julia, the data warehouse, and also putting a mededata once, which can then be used and reused. We're going to look at an extended mededata system, whether we can actually send people off with the mededata to external storage as well. So I won't go into that in detail with that, certainly on our work plan. So this doesn't exist yet. This is what we're going to do with it next. I'm a researcher preparing to submit a grant proposal. I need to develop and submit a data management plan. How do I do that? And part two, at the end of the project, I cite my data. So we put together really two use cases of one in the interest of past money. So created a data management plan through the web interface, stepping from the left there. Now we plan to create a number of templates based on particular needs. For example, probably the first thing we'll do here, based on discussions with the graduate research school, is create some sort of a plan for new PhD students. So new post-grad isn't necessarily a requirement, but it will be a possible way for them to actually think about and create a data management plan. So the research will be prompted for information which aligns this requirement of particular grant and project types as well. So if they happen to be going for an NSF grant, then we may in time have a template which actually aligns with that as well. So okay, this is what these guys want. So enter this metadata. When submitted, the data management plan will be managed in real data, and will be used as a basis for subsequent data management activities for that project. Okay, so it's the same pop, same fedora as the metadata went in for the data set records. So the RISC-S, so obviously we'll map various metadata schemes within it. If you already have a grant, this information will be available in a feed from Julia, the data warehouse, to ReSata, and you'll be able to link it to Julia's system, by the way, not a person. You'll be able to link it to the data management plan. If the grant is subsequently awarded, you can link the grant details to the data management plan later. So again, we're using existing metadata. Yes, obviously if the grant doesn't exist, they can create a record anyway. And if the grant doesn't exist in the data warehouse, though it's a small faculty grant, the record may not exist in there. They can make a record up, or create a new record. Lots of the data management plan can be exported through the reporting service. Second from the right on the bottom there is the reporting service. Now this is a multi-faceted feed. We would like and time for it to enable people to actually construct a report based on any combination of metadata within ReSata. And this could include aggregate reports for heads of school, or dean, if the dean of science wants information about all the assets from his faculty for the last five years, then there will be a report that can generate that. There will be a report if the graduate research school wants a particular report to be submitted for their PhD students, then that can be generated through the reporting service. So again, looking back, thinking back to the first slide, the membership of the board, we're taking our leads from the requirements that are giving us. Okay, so once the researcher has collected data and is ready to publish, publish research output publications here, a DOI for a data set can be requested as well. They could request one earlier, but this is the point at which we'll start to think about it. The DOI has created through and site my data service at data sites, and allows the data set to be cited in the publication. One of the features we've included with that is after they request the DOI, we're giving them from our system a 14-day cooling-off period, if you like, because once someone gets a DOI, it's supposed to be permanent. It's a fairly serious business, and we don't want someone's supervisor coming along asking what they thought they were doing when, in fact, they've already got the DOI. So those who are containing DOI certations can be entered into UNS Works, with the institutional repository, with their publications. The publication's records will be linked to the data set that go in the web data. That was a bit quick, wasn't it? So next, basically, what we're doing, I think, importantly, for us, is that while the funding certainly came from our hands, and thank you very much, but the purpose of the project really is to establish a structure within the university that actually will be used because of the system of existing and future systems. Yeah, and to actually look at how it extends beyond storing that data and trying to explain this in the search data of Australia, which is a very integral part of the system. So that's it. Thank you very much, Morg. Again, I wonder if there are any questions. Please use the chat box for that. But to start with, I'll ask you a question. When you were talking about this at the Sydney Roundtable, you mentioned that discipline-specific metadata was one of your key challenges. Can you talk a little bit about that? That's on the future planning area as well. What we want to do is map to metadata that has been used, for example. We're working quite closely with a company about social sciences, and I've actually been working over the years with people of Australia and what was social science data archive from now, Australian Data Archives about the possibility of getting metadata or data records from one of our research centres to Australia Data Archives. So we've been looking at BDI, which is Data Documentation Initiative, and mapping that to RISC-S, what else is in there, Morg, so that if people are using some sort of metadata given for their data, that can be mapped to what's in the metadata records. And I think I mentioned at the beginning the possibility of taking data, the metadata with them. So, for example, one thing I have been talking to a few people about is whether we can actually, we can generate a report from RISC-S, which generates XML or CSV file in CDI for records which in fact have CDI mapped to what's in the data, and that that could then be packaged and taken along to Australian Data Archives, for instance, with their data sets. Rather than again capitalising on the metadata of that creation already. Does that answer that? Yeah, that's great. And I've got one more for you before Lewis has a question for John. And that is that when, again, I'm aware that at the University of New South Wales, you've got a team of data librarians. I think you were saying there were ten of them at the moment. Can you talk a little about that sort of backup infrastructure that you're building? Outreach librarians and data librarians. Sorry, what did you do? Outreach librarians. Okay. Very much your concept of mediated deposit and what their role is in the mediation. Yeah, I'm not sure if the term mediated deposit came from and it wasn't mine, but I'll convert with it. What they're doing is they're talking, they're developing a package of resources for the outreach librarians, which they will then take out to the faculties. We're not doing that through ourselves. And so I guess by mediated deposit that will involve the faculty librarian that's been working with people within faculties to put metadata out for them. This can be designed for data entry from the faculty. The library is having a minimal role in doing that. Initially, of course, once we're rolling it out from the early doctors who get a little more hand-holding than others will get, we will do a lot of work with them. The plan is that the outreach librarians will be responsible for liaising with the faculty to get the metadata out. I've been in the Graduate Research School. We've been talking with them about contributing to their resources for the new graduate students. That's great. Thanks, Moor. Now, Lewis has a question for John. Yes. John and Lynn, I'm wondering whether you could talk a little more about two elements that I think are key from your project. One of them being that you've successfully managed the relationships with a research department in Utah which is already doing the research data management properly, and they have systems in place where they're recording metadata. The system, I think, is geonetwork and how you are working with them to get their metadata records into your institutional metadata store. The other aspect that I think is also interesting to the rest is that you have a vision on how you see the metadata stored activity developing in the future after the project. I was wondering if you could just describe these two aspects in a bit more detail. I'll head over to Lynn for the first one about the IMS and the geonetworks, and then I'll cover up on the second point first. So when we started the project, IMS had done an earlier and project at their scale. So they're a centre at UTAS, a discrete centre for marine and Antarctic studies, and they were quite mature and in how they've gone with publishing metadata records. So when we came onto the project, they already had 241 records published in RDA. So initially, I met with Peter on a few occasions, the data manager from IMS, to just get a feel of what they would like to do. And so he decided that he would really like to move his records when we have our established data repository through the institutional centralized repository store at UTAS. So I guess the technical part Sue can probably explain, because they were pushing out to two data centres, the AODM, the Australian Oceanographic Data Network, in a different schema, and also pushing out to RDA in RibCS with Sue, our technical officer on the project, has been working quite closely with IMS to have set up an automatic harvest from the... And import, yes. Yeah, import into our... Yeah, so at the moment, we're sort of halfway through that, and so Sue can probably comment on the more technical aspects, but what we hope to do is set up an automatic weekly harvest from the marine and Antarctic centre view, and it will pre-populate probably about 80% of the metadata elements straight into Redbox, and then data librarians will come in and just look at the bibliographic accuracy and do all the party curations, activity curations, and the actual publishing to RDA. So we're really quite excited about that, and so is IMS. So, Sue can probably comment a bit of where we are at in terms of ensuring we get the right RibCS and the work that you've done on that. Yeah, because IMS is doing RibCS a very long-time goal, and the version is just 1.1, and at the moment, they are working on to upgrade the RibCS file forward to 1.4, which is suitable for our Redbox. And in that opportunity, I will try the post-load into our Redbox and make sure everything is mapped correctly into our Redbox data hub. So we've done some tests. Yeah, the test was successfully, but just 80%, as Lin said, 80% data mapped over, and we try to improve the data quality and maybe 100% we could. So, is that answer the question, Luis, or not really? Yes, yes, yeah. Okay, so for point two, which is the forward-looking stuff, I guess this comes about in trying to pin down the business owner for the Redbox solution as it was, and I started thinking about, well, there are other things happening. When I went to the Redbox community day and a half, I called in on the Vivo Research and Profiles conference on the way through, which was happening in Melbourne when I was traveling, so that was pretty convenient. And I saw the Vivo product there, and I'm thinking, well, UTAS does have a portal, and I'll use that word, guardably for its researchers, when you see what Griffith are doing in University of Melbourne are doing in terms of the support of their researchers in making their research or information discoverable and accessible, and their outputs discoverable and accessible, and then I'm looking what's happening in the UTAS space and thinking our systems are very stale. Is it possible in terms of when I'm searching for or trying to pin down the governance model and the business owner within UTAS for the continuing operation of the Redbox system that we should be looking more strategically and looking at some opportunities there where we can point in place something that will replace our existing stale systems and come up with a business structure which will enable that to happen. So I started talking with the librarian here. Fortunately, there's a new person come on board within the research office who's taking a fresh start there and I'm making suggestions to both of them and I did a demonstration of the Griffith Researcher portal for them, trying to wet their appetise and you can see people start sitting up very straight when they see what's happening in other areas and realising the UTAS is perhaps falling behind. Certainly the head oceanographer at IMS he was very impressed with what's happening in that space and again wants to use the governance model that I set in place for the ongoing management of Redbox to also start to look at other systems as well. So yeah, I can't really say much more than that that I've wetted people's appetites. They've seen what else is happening in other areas and they're looking to use this project as a springboard to make other things happen as well. Does that answer your question, Luis? Yes. Thanks, John. There's a question from Sharon. I'm not sure if Sharon has a microphone. Before you do, the question in case somebody is not aware of what the question is. A shout from Sharon at UTS is that she's interested in that more is taking the approach of mapping to DDI and pushing to ADA. Yes, ADA. Since DDI is so much more granular than RISC-S, are you anticipating a gap between metadata entered in RISData and what ADA requires? Lord? Absolutely. There's a huge gap. But as I said in the presentation, what we want to do is ensure that what researchers have done with RISData can be reused. So we're simply saying that, okay, if you've gone through the trouble of putting this much in, someone which is collected from the grant report itself through the data warehouse, I will have done what they've put in, some of it maybe from the grant application and may actually be from the project level. Sorry, the more granular level. So to do this, most of what we'll do with DDI is the study level variables. The aim is simply that they take with them what's been done already rather than creating another rainbow that looks the same, but who knows, may or may not be. Does that answer your question, George? Yep, that's great.